Economics

Please skim and understand the concept and main idea of the following reading, no need to stress on small details and do the 2 page essay in the following format. attached as Breifing note.  

BIS Papers
No 67

Fiscal policy, public debt and
monetary policy in emerging
market economies

Monetary and Economic Department
October 2012

JEL classification: E52, E62, H63

Papers in this volume were prepared for a meeting of senior officials from central banks held
at the Bank for International Settlements on 16–17 February 2012. The views expressed are
those of the authors and do not necessarily reflect the views of the BIS or the central banks
represented at the meeting. Individual papers (or excerpts thereof) may be reproduced or
translated with the authorisation of the authors concerned.

This publication is available on the BIS website (www.bis.org).

© Bank for International Settlements 2012. All rights reserved. Brief excerpts may be
reproduced or translated provided the source is stated.

ISSN 1609-0381 (print)
ISBN 92-9131-152-9 (print)
ISSN 1682 7651 (online)
ISBN 92-9197-152-9 (online)

BIS Papers No 67 iii

Contents
BIS background papers
Fiscal policy, public debt and monetary policy in EMEs: an overview
M S Mohanty …………………………………………………………………………………………………………. 1
Is monetary policy constrained by fiscal policy?
Carlos Montoro, (OĘG�7DNiWV�DQG�James Yetman ………………………………………………………. 11
Developments of domestic government bond markets in EMEs and their implications
Aaron Mehrotra, Ken Miyajima and Agustín Villar ………………………………………………………. 31
Central bank and government debt management: issues for monetary policy
Andrew Filardo, M S Mohanty and Ramon Moreno …………………………………………………….. 51
Contributed papers
Fiscal policy, public debt management and government bond markets:
issues for central banks
Miguel Ángel Pesce ………………………………………………………………………………………………. 73
Fiscal consolidation and macroeconomic challenges in Brazil
Carlos Hamilton Araújo, Cyntia Azevedo and Sílvio Costa …………………………………………… 91
Macro policies and public debt in Chile
Sebastián Claro and Claudio Soto …………………………………………………………………………. 103
Monetary policy, fiscal policy and public debt management
People’s Bank of China ………………………………………………………………………………………… 113
Macroeconomic effects of structural fiscal policy changes in Colombia
Hernando Vargas, Andrés González and Ignacio Lozano ………………………………………….. 119
Some insights into monetary and fiscal policy interactions in the Czech Republic
Vladimír Tomšík ………………………………………………………………………………………………….. 161
The importance of fiscal prudence under the Linked Exchange Rate System
in Hong Kong SAR
Hong Kong Monetary Authority ……………………………………………………………………………… 173
The impact of public debt on foreign exchange reserves and central bank profitability:
the case of Hungary
Gergely Baksay, Ferenc Karvalits and Zsolt Kuti ……………………………………………………… 179
Sovereign debt management in India: interaction with monetary policy
R Gandhi …………………………………………………………………………………………………………… 193
Fiscal policy, public debt management and government bond markets in Indonesia
Mr Hendar ………………………………………………………………………………………………………….. 199
The interaction between monetary and fiscal policy: insights from two
business cycles in Israel
Kobi Braude and Karnit Flug …………………………………………………………………………………. 205
Public debt and monetary policy in Korea
Geum Wha Oh ……………………………………………………………………………………………………. 217
Banco de México and recent developments in domestic public debt markets
José Sidaoui, Julio Santaella and Javier Pérez ………………………………………………………… 233

iv BIS Papers No 67

Fiscal policy considerations in the design of monetary policy in Peru
Renzo Rossini, Zenón Quispe and Jorge Loyola ……………………………………………………… 253
Fiscal policy, public debt management and government bond markets:
the case for the Philippines
Diwa C Guinigundo ……………………………………………………………………………………………… 269
A framework for fiscal vulnerability assessment and its application to Poland
7RPDV]�-ĊGU]HMRZLF]�DQG :LWROG�.R]LĔVNL ………………………………………………………………. 285
Fiscal policy, public debt management and government bond markets:
issues for central banks
Elizaveta Danilova ………………………………………………………………………………………………. 295
Aspects of fiscal/debt management and monetary policy interaction:
the recent experience of Saudi Arabia
Abdulrahman Al-Hamidy ………………………………………………………………………………………. 301
Development of the government bond market and public debt management
in Singapore
Monetary Authority of Singapore ……………………………………………………………………………. 309
Fiscal policy, public debt management and government bond markets:
issues for central banks
Lesetja Kganyago ……………………………………………………………………………………………….. 315
Fiscal policy and its implication for central banks
Suchada Kirakul ………………………………………………………………………………………………….. 325
Globalisation of the interaction between fiscal and monetary policy
0HKPHW�<|U�NR÷OX�DQG�0XVWDID�.ÕOÕQo ............................................................................... 335 List of participants .............................................................................................................. 351 BIS Papers No 67 1 Fiscal policy, public debt and monetary policy in EMEs: an overview M S Mohanty1 1. Introduction During the 1980s and 1990s, the vulnerability of EMEs to shocks was often exacerbated by high fiscal deficits, underdeveloped domestic bond markets, and large currency and maturity mismatches. In many cases fiscal and monetary responses were procyclical. Debt management policy played very little part in either the choice of an optimal debt maturity or in stabilising the economy. Since the beginning of 2000s, however, the role of fiscal and monetary policy has started to become more active. Fiscal deficits and public debt levels in EMEs as a whole have declined substantially. Domestic financing has increased, and the share of foreign currency debt has fallen dramatically. And the average public debt maturity has lengthened significantly. What do these developments mean for monetary policy, particularly in the context of the recent global financial crisis? Has the threat of fiscal dominance in EMEs lessened, just when it has grown in the advanced economies (BIS (2012))? Have fiscal and monetary policies in EMEs become more countercyclical than in the past? Has the development of domestic bond markets helped? What role have central banks played in debt management and what are the implications for monetary policy? These questions were the focus of discussion at the 17th Annual Meeting of Deputy Governors from major EMEs held at the BIS in Basel on 16–17 February 2012. The meeting addressed three issues: (i) the fiscal constraints on monetary policy; (ii) the impact of local currency bond markets on central bank policies; and (iii) the role of central banks in public debt management. The current volume brings together the papers prepared by the BIS staff for the meeting as well as the contributions of central banks.2 One major finding emerging from the meeting was that improved fiscal positions helped many EMEs to rely on countercyclical fiscal and monetary policies to stabilise their economies during the recent global financial crisis. Anchoring medium-term fiscal expectations was crucial, but it was not by itself sufficient to insulate the economy from the shock. Greater access to domestic financing and the consequent reduction of currency mismatches, enabled by the domestic currency bond market, played an important role. Yet these conclusions came with a number of caveats. Although fiscal dominance has fallen in many EMEs, contingent liabilities and the costs of ageing populations pose serious medium- to long-term fiscal risks to many EMEs. In addition, although government debt levels have moderated, the volume of securities issued by central banks has expanded substantially, largely reflecting interventions in the foreign exchange market. Not only is the combined gross debt of the official sector (the government and the central bank) now large in many countries, but a considerable part of this debt consists of short-term securities, which are not characteristically very different from monetary financing. The implications of these balance sheet developments for price and financial stability require careful monitoring. 1 ,�DP�WKDQNIXO�WR�.HQ�0L\DMLPD�DQG�(OĘG�7DNiWV�IRU�WKHLU�FRQWULEXWLRQV�WR�WKLV�RYHUYLHZ��DQG�WR�3KLOLS�7XUQHU�IRU� comments. 2 The last time the Deputy Governors discussed fiscal/monetary interaction was in 2002 (see BIS (2003)). 2 BIS Papers No 67 The rest of this overview summarises the key points from the discussion and the background papers along the three organising themes of the meeting. 2. Fiscal constraints on monetary policy For much of the past three decades, fiscal policy remained a major concern for monetary policy in EMEs. Unsustainable fiscal deficits and public debt levels created the spectre of fiscal dominance in many countries, leading to high and volatile inflation and elevated risk premia on government debt. An unfavourable exchange rate dynamic – linked to weak fiscal and monetary policy credibility – exposed EMEs to destabilising capital outflows. As VXPPDUL]HG�E\�<|U�NR÷OX�DQG�Kilinç in their paper, such a fiscal setting was associated with low levels of financial development, a high degree of dollarisation, and high exchange rate pass-through. The consequence was that both fiscal and monetary policies tended to be procyclical in many countries, accentuating rather than damping economic volatility. Shift to countercyclical fiscal and monetary policy However, as argued in this volume by Montoro, Takáts and Yetman in their BIS background paper on “Is monetary policy constrained by fiscal policy?”, many EMEs have grown out of this procyclical policy bias over the past decade. A significant decline in fiscal deficits and public debt has reduced the problem of fiscal dominance, and made countercyclical policies more feasible. While EMEs’ average fiscal deficits as a percentage of GDP fell through the 1990s and the 2000s, reaching 1.8% during 2000–07, the period before the recent financial crisis, the reduction in gross public debt as a share of GDP was even more impressive. By measuring the degree of policy cyclicality from two separate fiscal and monetary policy reaction functions (from a Taylor rule), the authors show that in a majority of EMEs both fiscal and monetary policies were used to smooth output volatility during 2000–11. The scale of monetary and fiscal easing implemented by several EMEs in the worst phase of the recent global financial crisis was simply unthinkable during the 1980s and 1990s. Several country papers in this volume discuss the factors heralding this change. In most cases, measures to strengthen medium-term fiscal sustainability and monetary policy credibility played a decisive role. Brazil provides an interesting example of a dramatic turnaround in an economy that was once considered to be very vulnerable to crisis and procyclical policies. As noted in the paper prepared by Araújo, Azevedo and Costa, Brazil’s policy flexibility was enhanced by a number of critical policy reforms in the 1990s and 2000s, including the switch to an inflation targeting regime; concerted actions by the central bank and the Treasury to reduce the magnitude of short-term and various types of index-linked debt in the economy; and the introduction of the 1999 Fiscal Responsibility Law to strengthen financial institutions and transparency as well as to reinforce the goal of maintaining consistent primary surpluses. The paper by Braude and Flug demonstrates the marked difference in Israel’s responses to the 2001–03 and 2008–09 global shocks, which were dictated largely by the initial fiscal conditions facing the country. In the earlier period, high public debt and weak fiscal credibility meant that any increase in the fiscal deficit quickly translated into higher government bond yields. Even a modest reduction in the policy rate was considered by investors as unsustainable, causing sharp currency depreciations and subsequent monetary tightening. By contrast, during the 2008–09 global recession, the government allowed its fiscal deficit to rise and the central bank cut policy rates sharply. Improved fiscal and monetary credibility ensured that financial markets had little doubt about the sustainability of countercyclical policies. The discussion and country papers also confirmed that many commodity-exporting countries have been able to reduce their vulnerability to the potential volatility associated with BIS Papers No 67 3 commodity price cycles. In Chile’s case, as discussed in the paper by Claro and Soto, the introduction of the Fiscal Responsibility Act in 2006 proved to be a major turning point for the economy. It mandated the government to adopt a structural budget balance target, ie a fiscal balance corrected for fluctuations in revenue and expenditure due to business cycles. Similarly in Peru, as discussed by the paper by Rossini, Quispe and Loyola, fiscal rules providing for a nominal deficit target and a maximum limit for growth in non-financial public sector expenditure were critical in reducing the net debt of the public sector (public sector liabilities minus public sector assets). Together with accumulated surpluses in a separate fiscal stabilisation fund, the new fiscal framework has strengthened the role of monetary policy. The meeting also focused on the challenges facing economies with fixed exchange rate regimes. As is well known, when the exchange rate is fixed, fiscal policy is often the sole macroeconomic instrument that the authorities can use to address output volatility. But there was a view that, to mitigate risks to the fixed exchange rate regime, countercyclical fiscal policy should be used sparingly and only under exceptional circumstances. And, such stimulus must not compromise the medium-term sustainability of fiscal policy. The paper from the Hong Kong Monetary Authority discusses the central role of fiscal reserves in Hong Kong’s currency board arrangement. Historically, the government has followed a very prudent fiscal policy with a view to accumulating substantial fiscal reserves. An essential purpose of such reserves has been to underpin investors’ confidence in the fixed exchange rate, but they have also helped to cushion the economy against adverse shocks. Saudi Arabia has followed a somewhat different strategy. As noted by Al-Hamidy in his paper, the government has pursued an active fiscal stabilisation strategy by paying off debt when oil prices are high and spending more when they are weak. Notwithstanding the recent positive role of fiscal policy, there was a broad agreement that, beyond allowing the automatic stabilisers to work, the use of countercyclical fiscal policy should be limited. Some participants argued that crisis times are very different from normal cyclical downturns, when monetary policy is expected to do much of the output smoothing. To the extent that extraordinary monetary easing in advanced economies has helped many EMEs to pursue an aggressive stabilisation policy, it is unlikely that they would be able to repeat the recent experience in other times. In addition, authorities should try to avoid the unintended consequences of fiscal policy on the economy, which could arise from difficulties in measuring the cyclical stance in real time, uncertainty about fiscal multipliers and lags in ILVFDO�SROLF\��7KH�SDSHU�SUHSDUHG�E\�7RPdžLN�SURYLGHV�VHYHUDO�PHDVXUHV�RI�F\FOLFDOO\�DGMXVWHG� budget deficits for the Czech Republic, highlighting some of these issues. Nevertheless, there was a view that countercyclical fiscal policy could be used selectively to reduce some of the monetary policy challenges stemming from capital flows. For instance, fiscal tightening could be substituted for monetary tightening to address inflation pressures when capital inflows are attracted by large interest rate differentials. As Araújo, Azevedo and Costa show in the case of Brazil, a contractionary fiscal policy brought about by spending cuts couOG�KDYH�VLJQLILFDQW��SHUVLVWHQW�HIIHFWV�RQ�LQIODWLRQ��<|U�NR÷OX�DQG�Kilinç make similar arguments for using countercyclical fiscal tightening in Turkey. Fiscal policy and interest rates Another aspect of fiscal and monetary policy interaction explored at the meeting was the impact of fiscal policy on interest rates. In theory, the impact depends on whether the private sector is Ricardian or non-Ricardian. In a Ricardian world, fiscal deficits and debt have no consequences for interest rates, as the private sector saves the full extent of discounted tax liability implied by a rise in the fiscal deficit. In a non-Ricardian world, however, changes in fiscal deficits can lead to changes in interest rates. The classical mechanism is the “crowding out” hypothesis, where higher fiscal deficits, with an unchanged money supply, lead to higher interest rates. In economies with fiscal 4 BIS Papers No 67 dominance and a reliance on foreign credit, the mechanism that prevails is the default risk premium on government debt. For instance, in 7XUNH\��DV�QRWHG�E\�<|U�NR÷OX�DQG� Kilinç, external bond spreads had risen above 10 percentage points during the 2001 Turkish fiscal crisis. Several Latin American economies saw similar bond spreads during the 1990s and 2000s. Several country papers and Deputy Governors found that stronger fiscal balances and lower debt levels were followed by lower interest rates in EMEs. Indeed, one of the findings of Montoro, Takáts and Yetman is that the estimated equilibrium interest rates for EMEs (represented by the constant term of the Taylor rule) have been negatively correlated with the budget balances as a percentage of GDP. Although the link is weak, their results are consistent with a permanent reduction in interest rates in EMEs. Vargas, González and Lozano reach similar conclusions for Colombia. They note that not only have the country’s sovereign spreads fallen sharply following the recent fiscal consolidation, but they have also become less sensitive to global risk aversion. According to their estimates, about 60% of the decline in Colombia’s EMBI spread between 2002 and 2011 (excluding 2008 and 2009) could be attributed to local factors, particularly reductions in government currency mismatches and the government debt-to-GDP ratio. As noted by the authors, a permanent reduction in the long-term interest rate would have important implications for monetary policy not only by driving down the natural interest rate (the rate that would prevail with zero inflation and output gaps) but also by leading to changes in the equilibrium real exchange rate. Nevertheless, there was also a view that the recent developments in long-term interest rates should be interpreted with caution. Real long-term interest rates have fallen across the world, and disentangling global and local factors is difficult. A prolonged period of very easy monetary policy in industrial countries, the strong demand of EME central banks for highly rated bonds, and global risk aversion have driven real long-term rates to zero, or even negative. These conditions will not last forever. Future fiscal risks Worries about the medium-term sustainability of fiscal policy in EMEs surfaced prominently in the discussions. First, fiscal deficits and public debt levels are still high in a number of EMEs (for instance, in Hungary and India). Second, questions remain about the measurement of fiscal balances and public debt in several countries. The paper prepared by the People’s Bank of China points to a number of issues regarding the coverage of the fiscal balance. In China, although the government budget covers central and local finances, not all items of local government revenue and expenditure are included; in addition, the reported budget balance excludes the profits and losses of state-owned enterprises. Third, although explicit government liabilities have moderated in many EMEs, contingent liabilities remain high. Future liabilities related to implicit government guarantees to the financial system are difficult to assess accurately in many countries As pointed out in the paper prepared by Kirakul, in Thailand growing state-sponsored programmes have led to a sharp rise in implicit liabilities in recent years. In China, the People’s Bank of China notes that some of the local government liabilities, which are not covered by government debt statistics, require careful monitoring. Finally, many EMEs are ageing fast, and a large part of population, currently outside any social security systems, has to be ultimately covered by a formal pension system. This will put considerable pressure on the fiscal system in future. As Montoro, Takáts and Yetman summarise in the annex to their paper, the old-age dependency ratio in EMEs is expected to rise from an average of 11% in 2011 to 27% in 2040. While the impact of this rise will vary across regions and countries, depending on pension systems, going by the experience of industrial countries, the share of health and pension expenditure in GDP is expected to rise steadily in EMEs in the next decade. However, public policy reform can greatly reduce the BIS Papers No 67 5 fiscal burden of ageing populations. As the paper by -ĊGU]HMRZLF] and .R]LĔVNL shows, ageing-related expenditures are projected to decline in Poland thanks to pension fund reforms, while in all other EU member countries they are set to increase. 3. Local currency bond markets and central bank policies Following a series of financial crises in previous decades, many EMEs started to develop local currency bond markets in the beginning of the 2000s. Local currency bond markets help achieve several objectives: completing markets; reducing currency and maturity mismatches; diversifying financing sources; and strengthening the monetary transmission mechanisms. In many EMEs, central banks have often played a critical role in nurturing these markets. One important issue is how far these markets have developed in the past decade and what difference they have made to central bank policies, particularly in the conduct of monetary and financial stability policies. The meeting provided an opportunity to study these issues. The BIS background paper on “Developments of domestic government bond markets in EMEs and their implications” by Mehrotra, Miyajima and Villar provides a brief review of developments in this market. As reliance on foreign debt declined, the total stock of domestic debt securities issued by emerging market governments increased from about $1 trillion in 2000 to $4.4 trillion by 2010. The average remaining maturity of government local currency debt has roughly doubled over this period, from 3.5 years to seven years, with the longest debt maturity issued by EMEs being 28 years in 2010 compared to 14 years in 2000. The authors note that the expansion of domestic currency bond markets has been led by many factors including better domestic policies, lower inflation, reduced external financing needs and higher domestic saving in EMEs. Implications for the conduct of monetary policy One reason why local bond markets matter for monetary policy is that they increase the scope for long-term domestic currency financing, thus reducing currency and maturity mismatches. With borrowers’ and lenders’ financial health becoming less sensitive to changes in the exchange rate and interest rate, monetary policy can squarely focus on stabilising output and inflation. In the past, to prevent widespread bankruptcy among firms, many EMEs with large foreign currency debt were forced to raise interest rates during a downturn. Several country papers and the discussion at the meeting confirmed that the development of domestic bond market has led to a reduction in currency mismatches in many EMEs. Mehrotra, Miyajima and Villar present a number of indicators for currency mismatches in EMEs (Table 3 in their paper). Their finding is that since 2000 currency mismatches have fallen sharply, particularly in Asia and Latin America where most countries now enjoy net foreign currency asset positions. The paper from Peru (Rossini, Quispe and Loyola) argues that the government’s switch from external to domestic financing prompted the de- dollarisation of the banking system, shifting the focus of monetary policy away from the exchange rate. The papers from Israel and Colombia discuss similar evidence for the impact of recent reduction of currency mismatches on monetary authorities’ response to adverse shocks. The discussion also pointed to a number of challenges facing EMEs in monitoring currency mismatches and reducing the risk of future build-ups of foreign currency debt. One source of concern was that demand for foreign currency loans could increase on expectations that interest rates in emerging economies would remain above those in advanced economies causing EM currencies to appreciate. A second source of concern was that speculation about future exchange rates could prompt firms to shift currency mismatches to imperfectly monitored and regulated derivative markets. One view was that commitment to a floating 6 BIS Papers No 67 exchange rate was essential to prevent excessive currency speculation. Another was that central bank intervention in the foreign exchange markets should be made more predictable, so that markets have less scope for speculating on the exchange rate. Implications for the monetary transmission mechanisms A well developed sovereign yield curve is important for pricing risker assets and strengthening the interest rate and wealth channels of monetary policy. It also increases the role of the expectations channel of monetary policy as anticipation of central bank actions gets priced into forward curves, with implications for the borrowing and lending rates in the economy. The discussion was generally supportive of the view that the recent initiatives to deepen bond markets have strengthened the transmission channels of monetary policy. In most EMEs, governments have made efforts to reduce reliance on indexed-debt and floating-rate debt and increase financing through fixed rate debt, leading to the development of a domestic yield curve. Based on econometric work, the paper from Colombia argues that lower government currency mismatches and a deeper fixed rate domestic public bond market seem to have strengthened the response of market interest rates to monetary policy shocks. However, as noted in the paper by Guinigundo from the Philippines, possible interest rate repression stemming from reduced issuance of government securities could lead to distortions of the yield curve. In the Philippines, yields on Treasury bills, which are often used as a reference rate for pricing other loans, have fallen sharply because the government has rejected bids in auctions. This had led to confusing signals about monetary policy. This problem is likely to be even more severe in countries with persistent fiscal surpluses. Mainly to develop a domestic yield curve, some fiscal surplus governments have opted to issue bonds by overfunding their budgets. As noted by the paper from the Monetary Authority of Singapore (MAS), the government of Singapore is an interesting example of this trend in that its debt issuance is wholly unconnected to its fiscal requirements. While the MAS issues government bonds regularly to develop the yield curve, supplemented recently by its own bonds, it retains the proceeds from the sale of securities in a special government account to meet interest payments and repayments. Most Deputy Governors felt that diversification of the investor base is critical in boosting liquidity and reducing bond market volatility. As noted in the paper by Sidaoui, Santaella and Pérez, a more diversified investor base in Mexico has reduced the impact of idiosyncratic shocks on bond prices. The authors attribute this development to the growth of domestic institutional investors (particularly pension and mutual funds), leading to reduced concentration of bond holdings in the hands of any one investor category. A diversified investor base has contributed to a more stable pattern of investment by institutional investors. Several participants argued that greater foreign participation in domestic bond markets can on balance have positive financial stability implications, particularly in the long run. However, other participants highlighted that foreign inflows could numb monetary policy transmission and increase financial stability risks. In many EMEs, strong foreign capital inflows have compressed domestic yields, reducing the impact of monetary policy tightening. In addition, foreign inflows could be driven by carry trade incentives and are inherently volatile as highlighted in several country papers (eg Indonesia, South Africa and Thailand). Therefore Thailand has introduced a withholding tax on non-resident investors, while Indonesia has adopted measures to manage capital inflows and resultant excess liquidity. In this respect, the paper from Chile suggested that short-term financial volatility from foreign ownership may be mitigated by allowing domestic funds to invest abroad. As home bias increases during times of stress, domestic pension funds in Chile can absorb the foreign selling of domestic bonds. BIS Papers No 67 7 4. Central banks and public debt management The meeting’s final session focused on central banks’ involvement in debt management and its macroeconomic and monetary policy implications. Presently, the issue is being debated in advanced and the emerging market economies, as central banks have expanded their balance sheets sharply. Views differ widely about the role central banks should take in debt management (see BIS (2012)). In the advanced and emerging market economies alike, governments – or central banks on their behalf – manage public debt with several objectives in mind: eg to keep interest costs and refinancing risks to a minimum, ensure an adequate supply of risk-free assets in the economy and maintain a stock of short-term securities so that banks can adequately manage their liquidity risks. Yet microeconomic objectives are not the sole purpose behind an active debt management policy. In recent years, central bank interventions in debt markets have been motivated by macroeconomic considerations too; that is, to gain more control over the long-term interest rate or the exchange rate. The working assumption behind this motivation is that different assets held by private agents are imperfect substitutes for each other. Consequently, the central bank affects their relative prices (ie the asset returns) by changing the quantity of their supply. In the case of the yield curve, the central bank can alter the relative supply of short- and long-term bonds to manage the term structure. In emerging markets, as noted by Filardo, Mohanty and Moreno in their BIS background paper on “Central bank and government debt management: issues for monetary policy”, central banks have become a major issuer of domestic debt securities in the past decade. The authors highlight three salient trends about the size, issuance and maturity of the outstanding stock of debt securities. First, official debt securities issued by EMEs (government plus central banks) as a whole have increased from 19% of GDP at the end of 2000 to 29% of GDP at the end of 2010; the debt securities issued by central banks constitute 10–40% of GDP in several countries. Second, most debt securities issued by the central banks are short-term, with an average maturity of less than one year. Finally, partly reflecting central bank issuance, the share of outstanding short-term debt securities in total official debt securities remains high in EMEs, at about 37% of GDP at the end of 2010. Several perspectives on these developments were discussed at the meeting. One concerned the primary motivations behind central bank debt management and whether such motivations systematically differed from those of the government. There was a consensus that in many economies, central bank debt issuance has been driven by exchange rate and monetary policy considerations. When central banks intervene in the foreign exchange market to resist appreciation pressures on the exchange rate, they issue their own debt securities to banks to ensure that short-term interest rates do not fall below their policy rate target. At an operational level, the participants emphasised the need for central banks to have sufficient financial resources to absorb potential financial losses when altering the size and composition of the debt. On the one hand, debt issuance by central banks exposes them to interest rate risks, potential losses stemming from the positive interest rate carry (issuing high-yielding domestic bonds to finance low-yielding foreign assets) and the costs of rolling over debt at inopportune times. On the other hand, the increased scarcity of government securities implies that central banks have to rely increasingly on their own securities for sterilised intervention. Despite the potential benefits of actively managing the debt structure, some central banks have recently reduced or suspended issuance of their own securities. For instance, as noted in the paper by Sidaoui, Santaella, and Pérez, the Bank of Mexico has stopped issuing its own securities in order to allow for more domestic issuance by the government. This is expected to enhance market liquidity for government debt and lead to further reductions in currency and maturity mismatches. Similar efforts have been taken by the Bank Indonesia to reduce issuance of its own securities (see the paper by Hendar). 8 BIS Papers No 67 A second perspective that arose in the discussion was the role of short-term debt securities in influencing the effectiveness of monetary policy. There is a long-standing view that issuance of short-term government securities is akin to monetary deficit financing. Banks can, in particular, easily sell or leverage up on short-term securities and then expand credit to the private sector (Tobin (1963)). Converting short-term debt into long-term bonds (“funding”) reduces the liquidity of these assets because bonds cannot be liquidated without a capital loss. This is why selling long-term government bonds is regarded as non-monetary financing. For instance, the authorities in many advanced economies have adopted limits on short-term debt issuance by the government.3 In order to assess the magnitude of this potential source of liquidity, Filardo, Mohanty and Moreno update Tobin’s analysis for EMEs. They find that the share of highly liquid liabilities (the sum of monetary base and short-term debt of maturity of less than one year, adjusted for mandatory reserve requirement on banks) in official sectors’ combined liabilities ranged between 15% and 90% in EMEs at the end of 2010. They also find a positive correlation between this liquidity measure and the expansion of bank credit to the private sector. The paper by Gandhi notes that, in the Indian context, large government borrowing requirements have created significant challenges for the Reserve Bank of India in coordinating debt management and monetary policy. The RBI typically has to cope with persistent structural liquidity in the banking system, which needs to be managed carefully to ensure smooth transmission of monetary policy actions. Central banks that issue debt can use it to help reduce their reliance on other, more distortionary monetary policy tools. While the issuance of short-term debt may be seen as inflationary in some situations, it also reduces the need to resort to non-market policy instruments such as reserve requirements that impede the development of financial markets over time. In addition, the development of a market in short-term local currency debt can have positive effects on that of the inter-bank market for collateralised lending. A final perspective discussed was the role of active debt management in influencing the yield curve, particularly in volatile financial market conditions. The relevant channel is the term premium, which can change depending on the relative demand and supply for various securities. In emerging markets, as noted by Filardo, Mohanty and Moreno, given their relatively underdeveloped bond markets and a shortage of EME high-quality assets, the term premium is likely to be more sensitive to changes in demand for various debt maturities. A rising share of foreign investors in EME local currency bond markets has added further complexities. There is significant evidence that the spread between the short- and long-term yields has fallen sharply in many EMEs in recent years, although it is not immediately clear whether such a trend reflects an anticipation of future monetary easing or a reduction in the term premium associated with strong demand for long-term government paper. The discussion at the meeting also highlighted the impact of strong capital flows on domestic capital markets and raised the possibility that active debt management could be used as a policy tool to enhance financial stability. As noted by Oh in his paper, the recent drop in term spreads in Korea because of increased demand by foreign investors created significant challenges for the monetary authority as it was raising policy rates in response to upward inflation pressures. It remains an open question whether more active debt management could achieve a better balance of supply and demand for various debt maturities. 3 Provided, of course, that the authorities do not support the bond market. Patel’s classic paper is lucid on this point (reprinted in Khatkhate and Reddy (2012)). Until the late 1970s, many central banks used a liquid asset ratio to control bank credit. The policy was based on the view that long-dated bonds absorb liquidity from the banking system, thereby acting to tighten monetary policy (see Allen (2012) for an application to the UK). %,6�3DSHUV�1R���� �� � � References $OOHQ��:���������³*RYHUQPHQW�GHEW�PDQDJHPHQW�DQG�PRQHWDU\�SROLF\�LQ�%ULWDLQ�VLQFH�����´�� LQ�³7KUHDW�RI�ILVFDO�GRPLQDQFH´��BIS Papers, QR�����0D\��� %DQN� IRU� ,QWHUQDWLRQDO� 6HWWOHPHQWV� �������� ³)LVFDO� LVVXHV� DQG� FHQWUDO� EDQNV� LQ� HPHUJLQJ� HFRQRPLHV´��BIS Paper��QR������ BBBBBBBB��������³7KUHDW�RI�ILVFDO�GRPLQDQFH"´��%,6�3DSHUV��QR�����0D\�� 3DWHO��,���������³2I�HFRQRPLFV��SROLF\�DQG�GHYHORSPHQW��DQ�LQWHOOHFWXDO�MRXUQH\�E\�,�*��3DWHO´�� '�.KDWNKDWH�DQG�<�9�5HGG\��HGV���2[IRUG�8QLYHUVLW\�3UHVV�� 7RELQ��-���������³$Q�HVVD\�RQ�SULQFLSOHV�RI�GHEW�PDQDJHPHQW´��LQ�³(VVD\V�LQ�(FRQRPLFV�±� 9ROXPH����0DFURHFRQRPLFV´��1RUWK�+ROODQG�� BIS Papers No 67 11 Is monetary policy constrained by fiscal policy? Carlos Montoro, (OĘG�7DNiWV�DQG�James Yetman1 Abstract In this paper we analyse how fiscal policy has affected monetary policy in the emerging market economies (EMEs). We find that most EMEs have pursued countercyclical fiscal and monetary policy over the past decade, with little evidence of fiscal dominance, in contrast to earlier periods. Our results also suggest that stronger fiscal positions are weakly associated with lower equilibrium real interest rates, and smaller deficits with lower inflation. Overall, improvements in fiscal policy in EMEs appear to have increased the effectiveness of monetary policy. Keywords: Fiscal policy, monetary policy, Taylor rule JEL classification: E63, H63 1 The authors thank Tracy Chan, Emese Kuruc, Lillie Lam and Alan Villegas Sanchez for providing research assistance. 12 BIS Papers No 67 1. Introduction Fiscal policy and public debt matter for monetary policy. Not only can they influence interest rates and the level of aggregate demand, but they may also affect monetary authorities’ ability to control inflation. During the 1980s and 1990s, public debt levels in many emerging market economies (EMEs) remained high, constraining monetary policy. However, over the past decade fiscal positions in EMEs have generally improved. Public debt levels have fallen or moderated and governments in several economies have accumulated large holdings of financial assets. Many economies have adopted formal fiscal rules and most have abolished direct central bank financing of deficits, reducing the threat of fiscal dominance. Notwithstanding their strong medium-term growth prospects, EMEs’ fiscal positions are still exposed to financial and external demand shocks. In addition, many EMEs are likely to face significant fiscal pressures from ageing populations over the long term. Furthermore, contingent liabilities from government-owned corporations and the financial sector require careful monitoring. What are the implications of fiscal developments for monetary policy? In this note, we discuss three key aspects of this question. First, have EMEs left behind the era of fiscal dominance? Do they consistently pursue countercyclical monetary and fiscal policies? We argue that the ability of EME policymakers to conduct countercyclical economic policies represents a major advance, and one that contributes to global economic stability. But of course, countercyclical monetary and fiscal policies are not sufficient by themselves for good macroeconomic outcomes – in fact, many advanced economies facing economic crises today do so in spite of their countercyclical policies in the past. Second, is the long-run real interest rate related to fiscal deficits and the level of government debt? If so, further improvements in fiscal sustainability measures might lower real interest rates. Conversely, poor fiscal performance may have negative implications for long-run growth. Further, a fiscal deterioration could raise the spectre of a return to fiscal dominance and so complicate central banks’ efforts to control inflation. And third, what is the relationship between inflation and the government deficit? Are fiscal policies an important determinant of monetary stability? The rest of the note is organised as follows. In Section 2 we discuss factors influencing the relationship between fiscal and monetary policy. In Section 3 we present some preliminary empirical evidence on the three questions set out above. Using estimated Taylor rules, we show that both monetary and fiscal policy were generally countercyclical in EMEs over the past decade. Furthermore, equilibrium real interest rates are generally lower when fiscal deficits or government debts are lower. Finally, lower fiscal deficits are also associated with lower inflation in EMEs. The final section concludes. 2. Factors influencing the relationship between fiscal and monetary policy Countercyclicality of fiscal policy Some components of the budget balance vary with the business cycle, independently of policy decisions. Such automatic stabilisers include many types of tax revenue and social transfers. The structural, or cyclically adjusted, fiscal deficit is a measure of the hypothetical fiscal stance if output were to equal potential. Table 1 shows general government fiscal deficits and cyclically adjusted deficits as a percentage of GDP in EMEs. For 2011, by the latter measure, the fiscal stance in EMEs appears to be more expansionary than suggested by fiscal deficits, with some exceptions such as China and the Czech Republic. Also, headline fiscal surpluses invert to deficits in BIS Papers No 67 13 Chile, Hungary and Hong Kong SAR in 2011 after controlling for the effects of the business cycle. Table 1 General government fiscal and cyclically adjusted deficit1 Fiscal deficit Cyclically adjusted deficit 1990–992, 3 2000– 072, 4 2008 2009 2010 2011 2008 2009 2010 2011 Emerging Asia5 0.2 1.0 0.0 2.7 1.2 1.2 0.2 2.2 1.4 1.8 China 2.3 1.8 0.4 3.1 2.3 1.2 0.0 2.4 1.5 0.0 Hong Kong SAR -2.0 0.0 -0.1 -1.6 -4.5 -3.7 -0.2 2.2 1.4 2.2 India 7.7 8.0 7.2 9.8 9.2 8.7 9.3 10.8 9.7 9.1 Indonesia 1.0 0.0 1.8 1.2 1.6 0.2 1.7 1.2 1.6 Korea -2.0 -2.1 -1.6 0.0 -1.7 -2.3 -1.8 -0.7 -1.7 -2.4 Malaysia -0.1 3.7 3.2 5.3 3.7 5.1 4.9 5.8 6.1 5.4 Philippines 0.5 2.4 0.0 2.7 2.2 0.8 1.7 3.5 3.5 2.1 Singapore -21.0 -10.1 -5.6 0.5 -5.1 -7.3 -5.3 0.1 -4.8 -7.1 Thailand 2.2 0.4 -0.1 3.2 0.8 1.9 0.8 2.1 0.4 1.8 Latin America5 2.2 1.8 0.8 3.6 2.8 2.6 1.2 2.1 2.1 2.9 Argentina 2.7 4.6 0.8 3.6 1.6 3.3 1.1 1.7 0.7 3.3 Brazil 5.9 3.5 1.4 3.1 2.8 2.6 2.2 2.6 3.8 3.2 Chile -0.6 -2.4 -4.1 4.1 0.3 -1.2 1.1 4.1 2.0 1.2 Colombia 1.6 1.8 0.0 2.5 3.1 2.1 2.1 0.7 2.2 2.5 Mexico 3.1 2.1 1.1 4.7 4.3 3.4 1.3 3.8 3.8 3.2 Peru 0.4 -2.2 2.1 0.3 -1.9 -0.8 0.9 1.1 -0.9 Venezuela 1.8 -0.1 2.6 8.1 5.9 5.3 CEE5 4.8 4.3 2.4 5.8 4.3 0.3 3.9 4.0 3.9 3.1 Czech Republic 5.6 4.0 2.2 5.8 4.8 3.8 3.2 4.5 3.9 3.1 Hungary 3.3 6.6 3.7 4.5 4.3 -4.0 5.2 2.7 4.8 4.6 Poland 4.1 4.3 3.7 7.3 7.8 5.2 4.7 6.9 7.9 5.5 Russia 5.9 -4.6 -4.9 6.3 3.5 -1.6 -3.9 3.4 2.2 -1.6 Turkey 5.0 2.4 5.6 2.7 0.3 3.9 4.0 3.2 1.8 Other EMEs5 2.5 0.6 0.5 5.3 4.6 4.0 3.2 5.2 4.4 4.2 Israel 5.0 3.4 6.0 4.6 4.0 4.0 5.3 4.3 4.2 Saudi Arabia 2.5 -10.8 -34.4 4.6 -6.6 -15.2 South Africa 0.6 0.5 5.3 4.9 4.6 2.3 5.1 4.5 4.2 EMEs5 2.3 1.8 0.4 4.3 2.8 1.7 1.5 3.1 2.7 2.4 1 Overall fiscal deficit as a percentage of GDP and overall cyclically adjusted deficit as a percentage of potential GDP, respectively. 2 Mean 3 For Hong Kong SAR, 1991–99; for the Philippines, 1994–99; for Korea, Thailand, the Czech Republic, Hungary and Poland, 1995–99; for Brazil and Chile, 1996–99; for Argentina, 1997–99; for Russia, 1998–99; for Saudi Arabia, 1999. 4 For Turkey, 2002–07. 5 Simple median of the economies shown. Sources: IMF, World Economic Outlook and Fiscal Monitor Databases, April 2012. Even so, there are issues with the accuracy of cyclically adjusted balance measures in EMEs. As discussed in the background paper from the Czech Republic, they can be very sensitive to underlying assumptions about the level of potential output. A second problem relates to the adjustment of budget balances for commodity price changes. To be 14 BIS Papers No 67 meaningful, the structural budget balance must correct for exceptional movements in the terms of trade. This factor is particularly important in economies with a large share of production related to commodities such as mining, energy (including oil) and agricultural products. The methodology for adjusting for commodity prices parallels that used to construct a cyclically adjusted deficit, and amounts to adjusting tax revenues to those that would be received were commodity prices at equilibrium levels. Some economies already use an estimate of equilibrium commodity prices to estimate structural budget balances. Since 2002, Chile has used a rule-based fiscal policy whereby the structural budget balance is adjusted for cyclical movements in the prices of copper and molybdenum. According to the background paper by the Central Bank of Chile, an escape clause on the fiscal rule was put in place in 2009 to allow more scope for countercyclical fiscal policy during the recent global financial crisis. In 2010, Colombia introduced a targeting rule on the structural primary balance adjusted for the effects of cyclical oil prices. And Peru uses the structural budget balance adjusted for the cyclical effects of mining and energy prices as a guideline for multi-annual macroeconomic planning. According to the IMF (2009b), variation in commodity prices from equilibrium levels reduced the fiscal deficit by 0.7 percentage points of GDP in 2008 and raised it by 1.8 percentage points in 2009 across EMEs. Fiscal sustainability Fiscal sustainability is often defined in terms of measures of gross or net debt, as well as the change in debt given by the current and the expected future primary balance. Data for gross debt are more readily available than for net debt, and represent the total stock of outstanding government debt. Net debt is the difference between gross debt and financial assets owned by the government, although precise definitions vary by economy.2 Gross debt influences interest rates because it represents the total stock of debt that governments need to roll over. However, investors’ perceptions could also depend on net debt, especially in economies where the government holds a large stock of financial assets. In general, central banks regard net debt as the more appropriate measure of underlying government indebtedness since the financial holdings of the government can be liquidated to offset a portion of the gross debt. The difference between gross and net debt widened in many developed economies in the wake of the international financial crisis as a result of government purchases of financial assets, a process that is likely to reverse in the coming years as governments reduce their holdings of such assets. However, there are limitations to net debt as a measure of fiscal sustainability. In some cases, a portion of the government’s financial assets represents the government’s future obligations – government debt held by pension funds for government employees, for example. While these holdings may clearly be used to offset debt issued by the government, the future pension obligations that they are intended to finance would then need to be funded from some other source. Also, gross debt may be an important indicator of short-term fiscal vulnerability if there are limits to markets’ ability to absorb the sale of financial assets held by the government, especially during times of financial stress. As the government needs to refinance its gross (rather than net) debt as it matures, its ability to refinance its existing debt stock depends not only on the total level of debt but also on its maturity structure. As Graph 1 shows, gross debt in major EMEs varies widely. The graph also shows that, while net debt is a little lower than gross debt for most economies, in some cases the 2 For economy-level data, please refer to Tables A3-A4 in the Appendix. BIS Papers No 67 15 difference between the two is very large. Poland’s net debt is less than half of its gross debt, and in Saudi Arabia gross debt of 10% compares with net debt of –50%. Graph 1 General government debt 2010 As a percentage of GDP AR = Argentina; B = Brazil; CL = Chile; CN = China; CO = Colombia; CZ = Czech Republic; DE = Germany; FR = France; GB = United Kingdom; HK = Hong Kong SAR; HU = Hungary; IL = Israel; ID = Indonesia; IN = India; IT = Italy; JP = Japan; KR = Korea; MX = Mexico; MY = Malaysia; PE = Peru; PH = Philippines; PL = Poland; RU = Russia; SA = Saudi Arabia; SG = Singapore; TH = Thailand; TR = Turkey; US = United States; VE = Venezuela; ZA = South Africa. 1 As of 2009 for Russia; net debt data of Argentina, China, Czech Republic, Hong Kong SAR, India, Indonesia, Malaysia, Peru, Philippines, Singapore and Venezuela are not available. Sources: IMF, Government Finance Statistics; IMF, World Economic Outlook; CEIC; national data. Comparable data for the six largest advanced countries (by GDP) is reported in the right- hand panel of the graph. All the advanced economies are more indebted than all but two of the EMEs in net terms. This suggests that the EMEs are currently in much better shape than the major advanced economies in terms of debt sustainability. As fiscal sustainability is primarily about the expected future path of public debt, it is natural to consider public debt projections as a measure of fiscal sustainability. Current debt levels provide the starting point for such projections. These are then combined with assumptions about the future. The key variables are the economy’s expected growth rate, government spending levels and interest rates. Given the inherent uncertainties regarding these variables, any debt projections should be interpreted with caution, and the major underlying assumptions critically examined. Table A2 in the Appendix contains past and projected levels of public gross debt published by the IMF for EMEs. Debt levels increased in many economies between 2006 and 2010 as a result of the international financial crisis. However, in almost all EMEs gross debt is projected to be lower as a percentage of GDP in 2016 than in 2010. In terms of levels, the projections in the tables suggest some vulnerability. Gross debt is projected to remain close to 60% of GDP in India, Brazil and Israel beyond 2016, and above 70% in Hungary. This leaves fiscal sustainability in these economies somewhat vulnerable to a spike in interest rates, for example. In Singapore, high gross debt is less of a concern due to the large offsetting asset positions held by the government. Overall, the projections suggest that debt remains sustainable in most EMEs, at least for the next five years. However, ongoing population ageing that is projected to accelerate beyond then (see the discussion in the Annex and Graph A1) may pose a challenge further down the road. 16 BIS Papers No 67 Contingent liabilities of the government One key factor that all the previous analysis ignores is “invisible” public debt that may not be captured in standard debt statistics and may be very difficult to forecast. This latent debt reflects obligations to public corporations as well as explicit or implicit government guarantees. These contingent liabilities may also reduce balance sheet transparency and increase the risk of negative debt surprises, as the note from Thailand argues. Large state-owned corporations are a major source of invisible debt. These corporations play an important role in many EMEs, especially in sectors considered to be natural monopolies. For example, Indonesia’s state-owned Pertamina is the world’s largest exporter of liquefied natural gas, while Indian Railways is the country’s largest employer. State-owned entities benefit from the expectation of backing from the fiscal authority, resulting in lower financing costs. PetroChina, which is 87% state-owned, pays a spread of 160 basis points over Chinese sovereigns; by comparison, the private sector ExxonMobil pays 265 basis points over US sovereigns. Lenders have come to expect the government to prevent failures of state-owned firms, implying a potential liability for the fiscal authority. However, the debt of state-owned corporations does not generally appear in government debt statistics. Banking is another source of invisible public debt. State-owned banks account for a large share of many EME banking systems. China’s largest banks are majority-owned by the government. In India, state-owned banks hold over 75% of all deposits, a market share that has been growing since the beginning of the international financial crisis. While the debts of these institutions are not counted as part of public debt, the fiscal authority is likely to bail them out if necessary. As the background note for the case of Hungary shows, foreign currency-denominated private debt can also create challenges. Even private sector banks may benefit from implicit government guarantees. In India, private sector banks are largely free from the fear of failure as the government guarantees to take over banks’ uncovered liabilities if necessary. In late 2008, many governments in advanced economies resorted to significant bailouts of private sector banks, substantially swelling public sector debt. In earlier crises, Turkey’s public debt-to-GDP ratio rose from around 30% in 1999 to nearly 70% in 2001, and that of Thailand increased by two thirds as a result of the Asian financial crisis. While it is impossible to predict the potential cost of implicit guarantees to the financial sector in future, clearly a well regulated and well capitalised banking system plays an important part in ensuring fiscal sustainability. More generally, maintaining a precautionary debt buffer below the limit of what is sustainable is prudent in the light of implicit liabilities. 3. Consequences for monetary policy In this section, we formally analyse the three questions asked at the outset. First, we examine the cyclical properties of fiscal and monetary policies. Second, we examine how fiscal deficits and outstanding debt stocks might affect the real interest rate. Finally, we take a look at how fiscal deficits might affect inflation. Monetary and fiscal stabilisation In the past, EMEs often found it difficult to implement countercyclical policies. This was particularly the case for central banks. Monetary policy was frequently subordinated to the requirements of an expansionary fiscal policy, a condition described by Sargent and Wallace (1981) as fiscal dominance. And fiscal expansion during economic upturns left little scope for countercyclical policies during downturns. However, the era of fiscal dominance appears to have ended in most EMEs; monetary and fiscal policies appear to be countercyclical. We now examine this question further with statistical analysis. BIS Papers No 67 17 One way to measure how far monetary policy is countercyclical is to estimate the correlation between the business cycle and the real policy interest rate, controlling for other relevant factors. The Taylor (1993) rule offers a straightforward way to do so. The policy rate is modelled as responding to several variables: ( *) ( *) *i y y rp a b p p= + - + - + (1) where i is the nominal policy interest rate, p is the rate of inflation, p* is the (explicit or implicit) inflation target, y-y* is the output gap, r* is the “equilibrium” real interest rate; a and b are parameters that represent the degree to which a central bank responds to output and inflation developments, respectively. The intuition behind the Taylor rule is straightforward: a monetary authority should adjust the policy rate one-for-one for changes in inflation (p) and should respond positively to business cycle fluctuations (y-y*) and the deviation of inflation from the inflation target (p-p*). In particular, a larger a captures a more countercyclical monetary policy, while a negative value would imply a procyclical monetary policy.3 For fiscal policy, Taylor (2000) provides an analogous approach. The fiscal balance, measured as a percentage of GDP, is split into structural and cyclical factors: * ( *)b b y yg= - - (2) where b denotes the general government budget balance as a percentage of GDP, b* the cyclically adjusted deficit, y-y* the output gap and g the degree of sensitivity of budget balance to the output gap. The coefficient g can be used to measure for the degree of countercyclicality; the larger g becomes, the more countercyclical is fiscal policy. Similarly, as in the case of monetary policy, a negative g would imply procyclical fiscal policies. The degree to which monetary and fiscal policies are countercyclical is estimated over the 2000–11 period for a subset of EMEs that have adopted inflation targeting. To better match the data in the EMEs under investigation, equation 1 is extended to include an exchange rate term to reflect EME concerns about exchange rates in monetary policy-setting. In addition, an autoregressive term is added representing the preference of policymakers for smoothing interest rates. The two modifications yield the following empirical specification: [ ]1 1(1 ) * ( *) ( *) ( ) *i i y y e e rf f p a b p p d e- -= + - + - + - + - + + (3) where, in addition to the variables defined in equation 1, the subscript (–1) denotes one- quarter lagged variables, f is an autoregressive parameter reflecting the preference of a monetary authority to smooth policy rate adjustments over time, e is the bilateral nominal exchange rate vis-à-vis the US dollar, d is the parameter reflecting the monetary policy response to exchange rate movements, and e is the error term. The time and country subscripts are omitted for ease of representation.4 Notice that a remains the parameter of interest, because it captures the long-run countercyclicality of monetary policy. 3 Furthermore, a larger b might also signal that monetary policy is more countercyclical in responding to output deviations to the extent that these output deviations also appear in the inflation rate (via, for instance, the relationships captured in the Phillips curve). 4 Potential output (y*) is estimated on quarterly output data (y) between 1999 Q1 and IMF projections up to Q4 2013 using the Hodrick-Prescott filter. 18 BIS Papers No 67 In an analogous way, equation 2 is also modified to incorporate policy preferences for smoothing: 1* ( *) (1 ) ( *)b b b b y yy y g x-- = - - - - + (4) where, in addition to the variables defined in equation 2, y represents the policy-smoothing preference for fiscal policy and x is the error term. The time and country subscripts are again omitted for ease of representation.5 As in equation 3, g remains the parameter of interest because it captures the long-run countercyclicality of fiscal policy. For each inflation targeting EME, equations 3 and 4 are estimated jointly using the method of seemingly unrelated regression for the 2000–11 period. In order to provide some context, similar estimates – without the exchange rate term in equation 3 – are also obtained for advanced economies.6 Table A5 in the Appendix shows the estimation details. Graph 1 presents the point estimates of a and g and offers a cross-country perspective on the countercyclical characteristics of monetary and fiscal policies during the 2000–11 period. The vertical axis measures a , the degree of countercyclicality for monetary policy, while the horizontal axis measures g, the degree of countercyclicality for fiscal policy. Consequently, policies which fall into the first quadrant ( 0a > , 0g > ) are countercyclical and policies which
fall into the third quadrant ( 0a < , 0g < ) are procyclical. Policies in the second ( 0a < , 0g > )
and fourth ( 0a > , 0g < ) quadrant are ambiguous and their cyclicality depends on the relative strength of monetary and fiscal policies. The results show that most EMEs were able to pursue countercyclical policies during the decade as the dots representing individual economies are either in the first quadrant or near its border. This impression is confirmed by a more formal statistical analysis. The last column on Table A5 in the Appendix shows the probability that both monetary and fiscal policies were countercyclical (ie 0a > and 0g > ). The probabilities are close to unity for around half
of the EMEs in the sample, and are below one half in only two cases. The evidence suggests
that EMEs as a group were able to pursue countercyclical monetary and fiscal policies.
Naturally, the policy mix varies considerably. While most EMEs used both monetary and
fiscal policy to lean against the business cycle, some relied more heavily on one policy than
the other. For example, Thailand and Turkey relied heavily on fiscal policy while the Czech
Republic and Indonesia looked more to monetary policy. The degree of countercyclicality
also varied markedly from country to country. For instance, Chile pursued the most
countercyclical fiscal policy among EMEs. This may reflect policy preferences for output
stabilisation (as laid down by Chile’s fiscal responsibility law) and also the need to stabilise
output in the face of volatile copper prices. Yet, fiscal policy is not necessarily dictated by
commodity prices: Russia pursued a less countercyclical fiscal policy despite its exposure to
oil prices. It seems that policy preferences as well as economic and institutional frameworks
have all shaped the policy mix applied by EMEs over the past decade.

5 Quarterly budget balances are seasonally adjusted and, where not available, are extrapolated from yearly
figures. The structural budget balance (b*) is estimated on quarterly budget balance data between Q1 1999
and IMF projections up to Q4 2013 using the Hodrick-Prescott filter on quarterly budget balances (b). This
estimate of b* is used because it is available for all countries, allowing a consistent methodology. This choice
does not seem to affect the results: using OECD estimates where available instead does not materially affect
the estimates of g .
6 The exchange rate term is not used for advanced economies, because exchange rate concerns appear to be
less relevant for policymakers there. Importantly, this estimation choice does not materially affect the
estimates of a and thus our conclusions.

BIS Papers No 67 19

Graph 2
Countercyclical monetary and fiscal policies1
2000-20112
Emerging market economies Euro area Other advanced economies

AT = Austria; AU = Australia; BE = Belgium; BR = Brazil; CA = Canada; CH = Switzerland; CL = Chile;
CN = China; CO = Colombia; CZ = Czech Republic; DE = Germany; FI = Finland; FR = France; GB = United
Kingdom; GR = Greece; HU = Hungary; ID = Indonesia; IE = Ireland; IT = Italy; JP = Japan; KR = Korea;
LU = Luxembourg; MX = Mexico; NL = Netherlands; NO = Norway; NZ = New Zealand; PE = Peru;
PH = Philippines; RU = Russia; SE = Sweden; TH = Thailand; TR = Turkey; US = United States.
1 Seemingly unrelated regression estimation of equations (3) and (4). For details, see Appendix Table A5.
2 Years without an (implicit) inflation target were excluded. 3 The horizontal axis shows how countercyclical
fiscal policy is in output stabilisation ( g of equation (4)). 4 The vertical axis shows how countercyclical
monetary policy is in output stabilisation ( a of equation (3)).
Sources: IMF, World Economic Outlook; OECD, Economic Outlook; Bloomberg; Datastream; JPMorgan Chase;
national data; BIS calculations.
To put the EME results into perspective, the centre and the right-hand panels show the
results for the advanced economies. The centre panel confirms that policies were also
countercyclical in the euro area. Not only did the common monetary policy turn out to be
countercyclical in all countries for which estimates were possible, but fiscal policy was also
countercyclical in all countries except Greece. Interestingly, the estimates show that, on
average, countercyclicality in the euro area was similar to that of the EMEs, although slightly
stronger. Unfortunately, further interpretation of the euro area results is not straightforward,
as euro area countries do not have monetary policy independence.
Policies among other advanced economies were so much more countercyclical that the
scales needed to be recalibrated on the right-hand panel. In particular, Japan and some
English-speaking economies (Australia, Canada, the United Kingdom and the United States)
stand out for their markedly countercyclical fiscal policies. For most of these countries, the
phenomenon seems to be explained by the huge scale of the fiscal packages enacted after
the Lehman failure. In any case, policy, especially fiscal policy, seems to be substantially
more countercyclical in most of these economies than in EMEs.
In sum, both monetary and fiscal policy was countercyclical in most EMEs over the past
decade. Although the estimates vary from country to country, the degree of countercyclicality
compares with that in many advanced economies.
Fiscal deficits and government debt: effects on interest rates
Fiscal policy might have substantial effects on monetary conditions, and thus on monetary
policy, beyond its direct countercyclical effects. In particular, sustainability concerns due to
large deficits or high debt levels might put upward pressure on long-term interest rates.

20 BIS Papers No 67

The left-hand panel of Graph 3 shows that budget balances display a weak, inverse
relationship to estimated equilibrium real interest rates. The horizontal axis shows the
estimated structural general government balances as a percentage of GDP (b* from
equation 2) while the vertical axis displays the estimated equilibrium real interest rate (r*
from equation 3). The negative trendline implies that larger surpluses (or smaller deficits) are
associated with lower real interest rates, as the crowding out hypothesis would suggest,
although the relationship is weak.
Graph 3
Budget balance, government debt and equilibrium real interest rates
2000–111
Budget balance – equilibrium real interest rates2 Government debt – equilibrium real interest rates
3

BR = Brazil; CL = Chile; CN = China; CO = Colombia; CZ = Czech Republic; HU = Hungary; ID = Indonesia;
KR = Korea; MX = Mexico; PE = Peru; PH = Philippines; RU = Russia; TH = Thailand; TR = Turkey.
1 Years without an (implicit) inflation target were excluded. 2 The horizontal axis shows b* from equation (2), ie
the average general government net lending as a percentage of GDP, averages based on annual data. The
vertical axis shows equilibrium real interest rates, ie r* from equation (3), averages based on quarterly
data. 3 The horizontal axis shows the average general government debt as a percentage of GDP, averages
based on annual data. The vertical axis shows equilibrium real interest rates, ie r* from equation (3), averages
based on quarterly data.
Sources: IMF, World Economic Outlook; OECD, Economic Outlook; Bloomberg; Datastream; JPMorgan Chase;
national data; BIS calculations.
Furthermore, the right-hand panel of Graph 3 shows that equilibrium real interest rates are
positively associated with government debt. The horizontal axis displays general government
debt as a percentage of GDP while the vertical axis shows the estimated equilibrium real
interest rate (r* from equation 3). Higher government debt is associated with higher real
interest rates and vice versa as the crowding-out hypothesis would predict. In a similar vein,
the background paper from Colombia finds that lower structural deficits lead to lower risk
premia. However, the relationship is weak – and the underlying theory ambiguous. While
government debt can crowd out private investment, strong private balance sheets might also
enable the government to maintain large debt levels with low interest rates. High UK
government debt throughout the 19th century could be one example of this. Again, in spite of
some general trends, EMEs display large heterogeneity as both panels of Graph 3 confirm.
The inflation effects of fiscal deficit
Fiscal policy choices may affect the ability of monetary policy to achieve inflation stability.
The well known extreme case is fiscal dominance, when fiscal policies force the central bank
to abandon its price stability goal. Under a fiscally dominant regime, as defined in Sargent
and Wallace (1981), central banks may not be able to counter inflationary pressures
effectively. For this reason, Blanchard (2005) argues that inflation targeting would not have

BIS Papers No 67 21

been appropriate in Brazil in the early 2000s. In contrast, in a monetarily dominant regime,
fiscal policy accommodates monetary policy, rather than being subsumed by it. The
background note from Singapore outlines a special case of this. There, sound fiscal policy
allows the central bank to manage the exchange rate, which is its primary monetary policy
instrument.
Many EMEs have taken steps to reduce the threat of fiscal dominance, especially in the last
10 years. However, even in the absence of direct monetisation, fiscal policy might still affect
inflation. Excessive fiscal deficits can contribute to economic overheating and higher inflation.
For instance, spending may be systematically higher in election years, as Drazen (2004)
documents. Furthermore, inflation expectations might increase when the medium-term path
of public debt is perceived to be unsustainable.
More conservative fiscal policies are indeed weakly associated with lower inflation. Graph 4
shows average fiscal deficits (on the vertical axis) and average inflation (on the horizontal
axis) during the 1990s (left-hand panel) and the 2000s (right-hand panel). The positively
sloped trend (blue line) shows that a higher fiscal deficit is associated with higher inflation.
Interestingly, the relationship is more positive when high-inflation economies such as
Venezuela and Russia are excluded from the sample (lower two panels), although there is
substantial variation across EMEs.
4. Conclusions
Returning to our three questions: first, can EMEs consistently pursue countercyclical
monetary and fiscal policies? Our analysis suggests that, indeed, most EMEs have been
able to pursue countercyclical policies over the past decade. Furthermore, EMEs which
leaned against the business cycle generally relied on both monetary and fiscal policy to do
so. In fact, the degree of countercyclicality is only slightly below that seen in most euro area
countries, suggesting that EME policy frameworks have matured substantially – although it
must be noted that EMEs vary considerably in their policy preferences, economic structures
and institutional frameworks.
Second, is the long-run real interest rate related to fiscal deficits or the level of government
debt? Our results suggest that stronger fiscal positions (lower deficits and lower debt levels)
are weakly associated with lower equilibrium real interest rates. This implies that further
improvements in fiscal sustainability could also yield lower interest rates. Conversely,
deteriorating fiscal outcomes would be likely to have negative implications for long-run
growth, as higher interest rates crowd out domestic investment, complicating the stabilisation
role of central banks.
And third, is steady-state inflation related to the government deficit? Empirical evidence
suggests that conservative fiscal policies are weakly associated with lower inflation,
especially once high-inflation outliers are excluded from the sample. This suggests a
cautionary interpretation to recent evidence of declining fiscal dominance in EMEs: the
apparent decline may simply reflect a run of good fiscal outcomes. The corollary is that
deterioration in fiscal performance may see a return to fiscal dominance.

22 BIS Papers No 67

Graph 4
Fiscal and monetary policy interaction
In per cent
1990s1, 2 2000s
1, 3

1990s excluding Russia and Venezuela 1, 2 2000s excluding Russia and Venezuela1, 3

AR = Argentina; BR = Brazil; CL = Chile; CN = China; CO = Colombia; CZ = Czech Republic; HK = Hong
Kong SAR; HU = Hungary; ID = Indonesia; IL = Israel; IN = India; KR = Korea; MX = Mexico; MY = Malaysia;
PE = Peru; PH = Philippines; PL = Poland; RU = Russia; SA = Saudi Arabia; SG = Singapore; TH = Thailand;
TR = Turkey; TW = Chinese Taipei; VE = Venezuela; ZA = South Africa.
1 Simple average. 2 For Hong Kong SAR, 1991–99; for Korea, Thailand, the Czech Republic, Hungary and
Poland, 1995–99; for Brazil and Chile, 1996–99; for Argentina, 1997–99; for Russia, 1998–99; for Saudi Arabia,
1999; 1990–99 otherwise. 3 For Turkey, 2002–10; 2000–10 otherwise. 4 Corresponding to general
government; as a percentage of GDP. 5 Annual changes in CPI.
Sources: IMF, World Economic Outlook Database, September 2011; national data.

BIS Papers No 67 23

Annex:
Additional fiscal sustainability issues in EMEs
Pension liabilities and demographics
Additional caveats to debt as a measure of fiscal sustainability are pension obligations and
changing demographics. In some economies, pension plans operate on a “pay-as-you-go”
basis, with contributions used to fund immediate obligations. When underlying demographics
were favourable, due to high birth rates or immigration, these appeared to be self-funded for
many years. But ageing populations make this pension model unviable, as has been well
documented for advanced economies in Cecchetti (2011), for example.
Although many EMEs currently enjoy a relatively favourable demographic situation,
populations there are also expected to age rapidly in the coming years (Graph A1). Old-age
dependency ratios are expected to increase from an average of 11% in 2011 to 27% in 2040
in the listed economies, and to more than treble in China and Korea.
Graph A1
Old-age dependency ratio1
In per cent
Emerging Asia Latin America Other EMEs

1 Ratio of the population aged 65 years or over to the population aged 15–64.
Sources: US Census Bureau; World Bank.
The effect of ageing populations on debt sustainability will vary widely. In Latin America, the
rate of ageing is expected to be relatively low and pensions are generally well funded. The
background note from Poland provides another positive example: even though Poland is
expected to experience one of the fastest ageing processes in the European Union, age-
related expenditure is expected to fall over the next 50 years due to pension reforms enacted
in 1999 that provide for a switch from defined benefit to partly defined contribution plans. In
contrast, ageing will occur rapidly in emerging Asia over the next two decades and current
pension plans are generally too small to provide a secure, sustainable and adequate
retirement income for current workers. In addition, underlying demographic developments will
translate into increased fiscal demands for health care funding to meet the needs of growing
numbers of retired workers, as the note from Hong Kong SAR outlines.
Demographic developments are also likely to put strains on fiscal sustainability due to slower
future growth. As the note from China argues, ageing populations will result in lower
economic growth rates and therefore a diminished future tax base. Persistent fiscal deficits

24 BIS Papers No 67

that appeared to be sustainable because debt-to-GDP ratios were stable may become
unsustainable.
The case of Japan may be instructive. While productivity growth in Japan has matched or
exceeded that of many other advanced economies in recent years, GDP growth has appeared
anaemic due to low capital accumulation and a shrinking labour force (Graph A2, left-hand
panel). Thus the rapid increase in debt-to-GDP ratios (right-hand panel) reflects not just
significant deficits driving up the numerator, but slowing growth in the denominator as well.
From the late 1980s, when Japan was growing at around 5% and net debt was a mere 13% of
GDP, it took only 20 years to deteriorate to the point where net debt stood at 117% in 2010.
Japan may serve as a cautionary tale as to how quickly debt sustainability can erode away
when population growth stalls and demographic trends start to work against economic growth.
Graph A2
The case of Japan
In per cent
Real sector1 Government debt
2

1 Year-on-year changes. 2 As a percentage of GDP.
Sources: IMF World Economic Outlook; OECD; CEIC; national data.
Sovereign wealth funds
One important variable that is typically excluded from debt sustainability calculations is the
sovereign wealth fund (SWF). SWFs are government investment vehicles that are typically
funded by foreign exchange assets but managed separately from the official foreign
exchange reserves of the monetary authorities. The investment horizon of SWFs is longer
than that of the official reserves, mainly because the primary goals of the latter are liquidity
and security rather than long-run investment returns.
Graph A3 displays the size of total SWFs, by economy. SWFs play a particular role in fiscal
sustainability for non-renewable resource exporters. Any economy where concurrent
government spending is supported with such revenues faces sustainability issues as
available resources are extracted. Well designed SWFs can provide the mechanism to
transform resources into sustainable and stable future income. The background paper from
Saudi Arabia outlines how such assets are part of an intergenerational swap, transforming
natural resource revenues into monetary reserves for the benefit of future generations.
Following this model, Algeria, Russia and Venezuela also have stabilisation funds funded
with revenues from oil, while Chile has a stabilisation fund and a pension reserve fund
funded with earnings from copper. As noted by the background paper for Chile, the
administration of these two funds was delegated to the Central Bank of Chile.

BIS Papers No 67 25

Graph A3
Main sovereign wealth funds (SWFs) by economy and size
December 2011, in billions of US dollars

AE = United Arab Emirates; AU = Australia; AZ = Azerbaijan; BH = Bahrain; BN = Brunei Darussalam;
BR = Brazil; BW = Botswana; CA = Canada; CL = Chile; CN = China; DZ = Algeria; FR = France; HK = Hong
Kong SAR; IE = Ireland; IR = Iran; KR = Korea; KW = Kuwait; KZ = Kazakhstan; LY = Libya; MY = Malaysia;
NO = Norway; NZ = New Zealand; OM = Oman; OT = other economies, including among others (by SWF size)
Mexico, Italy, Venezuela and Indonesia; QA = Qatar; RU = Russia; SA = Saudi Arabia; SG = Singapore;
US = United States.
Source: Sovereign Wealth Funds Institute.
Fiscal rules
Historically, EMEs have faced debt sustainability issues due to aggressive fiscal policies.
However, over the past decade a number of economies have implemented fiscal rules to
improve fiscal behaviour by increasing accountability, transparency and the quality of fiscal
policies, as Fatás (2005) argues. Table A1 shows that 13 out of the 24 listed EMEs have
some type of fiscal rule, and 10 of them have a numerical target for one or more fiscal
variables.

Table A1
Fiscal rules
Feature Number of economies Economies
Fiscal rules 13 AR, BR, CL, CO, CZ, HK, HU, ID, IL, IN, MX, PE, PL
Numerical target 10 AR, BR, CO, CZ,1 HU, ID, IL, IN, MX,2 PE
Escape clause 7 AR, BR, CZ,3 IL,4 IN, PE, PL5
Sanctions 5 AR, BR, CO, PE, PL6
Monetary financing prohibited
by law 4 BR, CL, PE,7 PL
AR = Argentina; BR = Brazil; CL = Chile; CO = Colombia; CZ = Czech Republic; HK = Hong Kong SAR;
HU = Hungary; ID = Indonesia; IL = Israel; IN = India; MX = Mexico; PE = Peru; PL = Poland.
1 Expenditure limits are inserted in a medium-term expenditure framework. 2 Balanced budget on a cash basis
3 The government may change the medium-term expenditure framework only in defined cases. 4 The Defici
Reduction Law excludes public investment or other priority items from ceiling. 5 Rules exclude public investmen
or other priority items from ceiling at sub-national levels. 6 The Public Finance Act includes triggers for corrective
actions when the debt ratio reaches thresholds of 50%, 55% and 60% of GDP. 7 Prohibited from granting credi
to the government, except for the purchase, in the secondary market, of securities issued by the Public Treasury
these securities cannot exceed 5% of the money base.
Sources: Canales-Kriljenko et al (2010); IMF (2009a); BIS (2009).

26 BIS Papers No 67

However, the empirical evidence for the effectiveness of fiscal rules in enhancing discipline is
inconclusive. On the one hand, some studies suggest that fiscal rules have been an
important ingredient in successful fiscal consolidation: economies with fiscal rules have
managed to reduce their public debt levels more significantly than others (IMF (2009a)). On
the other hand, the recent experience of the euro area demonstrates the potential limitations
of fiscal rules.

BIS Papers No 67 27

Appendix:
graphs and tables
Table A2
General government gross debt
As a percentage of GDP
2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016
Asia1 46.8 45.0 45.2 48.5 48.6 46.9 46.1 44.8 43.8 42.9 41.9
China 16.2 19.6 17.0 17.7 33.5 25.8 22.0 19.4 17.1 14.8 12.6
Hong Kong SAR 33.0 32.8 30.6 33.2 34.6 33.9 33.2 30.4 29.7 29.0 28.4
India 78.5 75.4 74.7 75.0 69.4 68.1 67.6 66.8 66.2 65.8 65.3
Indonesia 39.0 35.1 33.2 28.6 27.4 25.0 23.2 21.1 19.2 17.6 16.4
Korea 31.1 30.7 30.1 33.8 33.4 34.1 32.9 30.8 28.7 26.7 24.9
Malaysia 43.2 42.7 42.8 55.4 52.9 52.6 53.1 54.0 54.8 55.6 56.3
Philippines 51.6 44.6 44.2 44.3 42.2 40.5 40.1 38.7 37.2 35.8 34.4
Singapore 86.4 85.8 96.9 103.3 101.2 100.8 98.0 95.7 92.6 90.2 87.6
Thailand 42.0 38.3 37.3 45.2 42.6 41.7 44.4 46.3 49.0 50.3 51.2
Latin America1 42.7 39.5 37.7 40.0 37.7 36.7 35.7 35.0 34.2 33.5 32.9
Argentina 76.5 67.1 58.5 58.7 49.1 44.2 43.3 41.9 41.6 40.1 39.4
Brazil 66.7 65.2 63.5 66.9 65.2 66.2 65.1 63.1 61.5 59.9 57.7
Chile 5.0 3.9 4.9 5.8 8.6 9.9 10.1 9.8 8.7 7.6 7.1
Colombia 36.8 32.7 30.8 35.9 36.1 34.7 32.3 32.3 31.4 31.6 31.7
Mexico 38.4 37.8 43.1 44.6 42.9 43.8 42.9 42.9 43.0 43.1 43.1
Peru 33.1 30.4 25.2 28.4 24.6 21.6 20.7 19.8 19.2 18.7 18.3
Central and
eastern Europe1 37.7 37.1 39.1 44.0 46.4 46.7 46.1 46.1 46.1 45.7 45.4
Czech Republic 28.3 28.0 28.7 34.3 37.6 41.5 43.9 45.4 46.2 46.6 46.9
Hungary 65.9 67.0 72.9 79.7 81.3 80.4 76.3 76.0 75.4 74.3 73.1
Poland 47.7 45.0 47.1 50.9 54.9 55.4 55.7 55.2 53.9 52.2 50.5
Russia 9.0 8.5 7.9 11.0 11.7 9.6 8.4 7.9 9.0 9.7 11.3
Other emerging
markets1 47.8 41.2 39.4 43.2 40.9 40.0 39.0 38.3 37.6 36.6 35.4
Israel 84.7 78.1 77.0 79.4 76.1 74.3 74.0 72.6 70.8 69.1 67.4
Saudi Arabia 27.3 18.5 13.2 15.9 9.9 7.5 5.9 5.2 4.6 3.9 3.4
South Africa 32.6 28.3 27.4 31.5 35.3 38.8 40.0 40.8 41.5 40.7 38.8
Turkey 46.5 39.9 40.0 46.1 42.2 39.4 36.0 34.6 33.5 32.8 32.1
1 Simple average of the economies shown.
Source: IMF, Fiscal Monitor, April 2012.

28 BIS Papers No 67

Table A3
Gross and net general government debt1
As a percentage of GDP
2000 20052 2010
Gross Net Gross Net Gross Net
China 18.0 17.0
Hong Kong SAR 1.4 –0.1
Korea 18.3 28.6 33.4
Philippines 60.5 68.5 52.4
Singapore 26.6 37.4 43.6
Thailand 14.5 26.1 29.7
Argentina 52.5 88.3 52.5
Brazil 56.4 46.1 54.7 40.7
Chile 13.6 3.2 7.3 –0.1 9.2 –7.5
Colombia 38.8 36.5 44.2 39.1 40.2 35.7
Mexico 21.5 21.8 29.9
Peru 45.5 46.0 37.7 30.1 23.5 11.6
Czech Republic 17.8 –33.9 28.4 –16.6 37.6 –4.6
Hungary 56.1 50.8 61.7 57.5 81.3 73.5
Israel 84.3 71.6 93.7 83.8 76.1 68.2
Saudi Arabia 87.2 38.9 9.9
South Africa 43.4 42.6 34.7 30.2 35.1 29.6
Turkey 51.1 42.9
1 For China, Philippines, Saudi Arabia, Singapore, Thailand, Turkey and South Africa central government debt.
2 For Brazil, 2006.
Source: Results taken from central bank questionnaire, complemented where necessary with information from
IMF, World Economic Outlook.

Table A4
Gross and net general government interest payments1
As a percentage of GDP
2000 2005 2010
Gross Net2 Gross Net3 Gross Net
Hong Kong SAR 0.0 –0.1
Korea 1.2 0.30 1.2 –0.8
Philippines 3.9 5.3 3.3
Thailand 1.1 1.2 1.1
Argentina 3.7 2.3 1.8
Brazil 6.5 5.9 5.7
Chile 1.2 1.1 0.8 0.6 0.5 0.2
Colombia 3.5 3.1 2.7
Peru 2.4 1.9 1.1
Czech Republic 0.8 0.2 1.1 0.7 1.4 1.2
Hungary 6.1 5.3 4.2 3.6 4.4 3.8
Israel 5.3 4.5 4.9 4.2 3.4 3.1
South Africa 5.2 3.5 2.6
Turkey 7.0 5.7 4.4 4.0
1 For Philippines, Thailand, Turkey and South Africa central government interest payments. 2 For Chile,
2001. 3 For Brazil, 2006.
Source: Results taken from central bank questionnaire.

BIS Papers No 67 29

Table A5
Countercyclical policy parameter estimates
2000–11
Emerging
economies
a g
standard error
( a)
standard error
(g)
covariance
( a,g)
probability
(g>0, a >0)
Brazil 1.69 0.63 0.96 0.19 –0.01 0.96
Chile 0.57 1.11 0.20 0.20 0.01 1.00
Colombia 1.52 0.54 0.46 0.16 0.02 1.00
Mexico 1.70 0.31 1.00 0.04 0.00 0.95
Peru 0.64 0.20 0.40 0.32 0.03 0.70
Indonesia 1.31 –0.06 1.69 0.43 0.12 0.37
Korea 1.43 0.97 0.36 0.30 0.00 1.00
Philippines 1.25 0.72 1.37 0.43 0.13 0.79
Thailand –0.06 0.49 0.12 0.31 0.00 0.28
Czech Republic 1.72 0.05 1.15 0.34 0.06 0.53
Hungary 1.04 0.27 1.21 0.77 0.07 0.52
Turkey 0.21 0.25 0.68 0.18 –0.01 0.57
China 0.38 0.43 0.11 0.22 0.00 0.97
Russia 0.44 0.52 0.28 0.29 0.01 0.91
Advanced
economies
a g
standard error
( a)
standard error
(g)
covariance
( a,g)
probability
(g>0, a >0)
Australia 1.22 5.29 0.24 1.44 0.12 1.00
Canada 1.30 4.06 0.35 0.54 0.05 1.00
United Kingdom 1.24 4.09 0.21 0.74 0.04 1.00
Norway 4.06 1.85 3.03 0.59 0.19 0.91
New Zealand 2.75 0.98 0.68 0.44 0.07 0.99
Sweden 1.34 0.56 0.52 0.16 0.00 1.00
Austria 1.19 0.77 0.25 0.19 0.01 1.00
Belgium 2.00 0.85 0.32 0.22 0.01 1.00
Germany 1.20 0.66 0.33 0.15 0.01 1.00
Finland 0.82 0.64 0.20 0.06 0.01 1.00
France 1.82 0.95 0.36 0.11 0.01 1.00
Greece 0.51 –0.28 0.37 0.33 0.03 0.18
Ireland 0.26 0.90 0.07 0.84 –0.01 0.86
Italy 1.41 0.57 0.38 0.10 0.01 1.00
Luxembourg 0.65 0.74 0.17 0.19 0.01 1.00
Netherlands 1.43 0.65 0.81 0.27 0.05 0.95
Switzerland 0.82 0.51 0.15 0.06 0.00 1.00
Japan 0.13 5.07 0.05 0.87 0.00 1.00
United States 1.46 4.75 0.50 0.50 0.07 1.00
Note: Seemingly unrelated regression estimation of equation 3 and 4 (without exchange rate adjustment for
advanced economies). Estimates excluded where the null hypothesis that 1f < or 1y < could not be rejected. Years without (implicit) inflation target were excluded; for China, CPI inflation target set by the Central Economic Working Conference; for euro area countries, euro area inflation target; for the United States, 2%. Probability is calculated assuming normality of distribution. Sources: IMF, World Economic Outlook; OECD, Economic Outlook; Bloomberg; Datastream; JPMorgan Chase; national data; BIS calculations. 30 BIS Papers No 67 References Auerbach, A (2011): “Long-term fiscal sustainability in major economies”, BIS Working Papers, no 361, November. BIS (2009): “Issues in the governance of central banks”, a report from the Central Bank Governance Group, May. Blanchard, O (2005): “Fiscal dominance and inflation targeting: lessons from Brazil”, in F Giavazzi, I Goldfajn and S Herrera (eds), Inflation targeting, debt, and the Brazilian experience, 1999 to 2003, MIT Press, pp 49–84. Canales-Kriljenko, J et al (2010): “Weathering the global storm: The benefits of monetary policy reform in the LA5 countries”, IMF Working Paper, 10/292, December. Cecchetti, S (2011) “Fiscal policy and its implications for monetary and financial stability”, June, http://www.bis.org/events/conf110623/cecchetti . Drazen, A (2004): “Fiscal rules from a political economy perspective”, in G Kopits (ed), Rules-based fiscal policy in emerging markets: background, analysis and prospects, Palgrave Macmillan. Fatás, A (2005): “Is there a case for sophisticated balanced-budget rules?”, OECD Economics Department, Working Papers, no 466. IMF (2007): Global Financial Stability Report, October, http://www.imf.org/External/Pubs/FT/ GFSR/2007/02/index.htm. ——— (2009a): “Fiscal rules: anchoring expectations for sustainable public finances”, December, http://www.imf.org/external/np/pp/eng/2009/121609 . ——— (2009b): “The state of public finances: outlook and medium-term policies after the 2008 crisis”, March, http://www.imf.org/external/np/pp/eng/2009/030609 . Mehrotra, A and J Sánchez-Fung (2009): “Assessing McCallum and Taylor rules in a cross- section of emerging market economies”, BOFIT Discussion Papers, no 23, December. Sargent, T and N Wallace (1981): “Some unpleasant monetarist arithmetic”, Federal Reserve Bank of Minneapolis Quarterly Review, vol 5, no 3, pp 1–17. Taylor, J (1993): “Discretion versus policy rules in practice”, Carnegie-Rochester Conference Series on Public Policy, 39, pp 195–214. ——— (1995): “Monetary policy implications of greater fiscal discipline,” in Budget deficits and debt: issues and options, Federal Reserve Bank of Kansas City, 1995, pp 151–70. ——— (2000): “Reassessing discretionary fiscal policy”, Journal of Economic Perspectives, vol 14, no 3, pp 21–36. Zoli, E (2005): “How does fiscal policy affect monetary policy in emerging market countries?”, BIS Working Papers, no 174, April. BIS Papers No 67 31 Developments of domestic government bond markets in EMEs and their implications Aaron Mehrotra, Ken Miyajima and Agustín Villar1 Abstract During the past decade, domestic government bond markets in EMEs have developed further. Market depth has increased, maturities have lengthened and the investor base has generally broadened, although the degree of progress has varied across countries and several deficiencies remain. The expansion can be attributed to improvements in domestic policy management and a reduction in external financing needs. The commensurate reduction in currency mismatches has increased the scope for countercyclical monetary policy. Financial stability has broadly benefited from this development, but the volatility stemming from derivatives markets and greater foreign holding of domestic currency debt present some additional risks. Keywords: Financial markets and the macroeconomy, international lending and debt problems, financial aspects of economic integration JEL classification: E44, F34, F36 1 The authors thank Philip Turner and Madhusudan Mohanty for comments and Tracy Chan, Emese Kuruc, Marjorie Santos and Agne Subelyte for valuable assistance. 32 BIS Papers No 67 1. Introduction Since this topic was first discussed at the Deputy Governors’ Meeting a decade ago, domestic government bond markets in EMEs have developed further.2 Market depth has increased, maturities have lengthened and the investor base has generally broadened, although the degree of progress has varied across countries and several deficiencies remain. In this paper we address three interrelated questions. First, what are the factors promoting these developments and how much further is it possible for these markets to evolve? Second, what are the implications for monetary policy? In particular, has the development of domestic government bond markets reduced the potential for currency mismatches and enhanced monetary policy effectiveness? In the 1990s, for example, when foreign currency debt remained high, many EM central banks had to raise interest rates in the midst of a recession. The experience during the 2008–09 global recession appears to be quite different. Third, how have the domestic bond markets influenced financial stability? In principle, a better developed yield curve for domestic government bonds and improved market infrastructure should help the corporate sector issue more bonds in domestic markets and diversify funding risks, increasing the economy’s resilience to external shocks. In practice, however, the record has not been straightforward, although substantial progress can be seen in some countries. Also, recent bouts of market volatility have highlighted potential new risks as the investor base for domestic bonds becomes more global. These risks need to be managed. The rest of the paper is organised as follows. Section 2 documents the development of domestic government bond markets (excluding central bank issuance) in EMEs. Section 3 discusses potential reasons for these developments. Section 4 addresses the implications for monetary policy. Finally, Section 5 discusses the impact on financial stability. 2. How far have domestic government bond markets developed in EMEs? Over the past decade, domestic government bond markets have expanded in EMEs. To gauge how far these markets have developed and deepened, we focus on the following aspects: (i) size; (ii) composition in terms of maturities, type of instrument, and investor base diversity; and (iii) market liquidity. (i) Size As Graph 1 shows, the stock of domestic bonds as a percentage of GDP has increased in all regions between 2000 and 2010. And this has happened as overall government debt levels have declined or stabilised, suggesting a shift away from foreign currency debt. Domestic currency bonds have grown by about 10 percentage points of GDP to some 30% of GDP in Asia and central and eastern Europe (CEE), and by a few percentage points to some 15% of GDP in Latin America (Graph 1, right-hand panel). In absolute terms, the dollar value of these debts in EMEs as a whole has quadrupled from $1 trillion in 2000 to more than $4 trillion in 2010. Table A1 provides country details, along with evidence that the amount of tradable bonds is, in many cases, typically smaller than the total amount of outstanding domestic debt. 2 See BIS Papers, no 11, 2002. BIS Papers No 67 33 Graph 1 Government debt securities outstanding in EMEs1 As a percentage of nominal GDP Total government debt2 Domestic debt 3 1 Simple averages across the countries listed. 2 General government gross debt. 3 Domestic debt securities issued by government; central bank issues are excluded. 4 China, Hong Kong SAR, India, Indonesia, Korea, Malaysia, the Philippines, Singapore and Thailand. 5 Argentina, Brazil, Chile, Colombia, Mexico, Peru and Venezuela. 6 Czech Republic, Hungary, Poland, Russia and Turkey. Sources: IMF, World Economic Outlook; BIS securities statistics. (ii) Composition Maturity A second dimension of market development is the maturity. The remaining maturity of general government local currency debt in EMEs as a whole has increased on average by more than three years, from 3.5 years in 2000 to seven years in 2010. This was partly due to the extension of the longest maturity, which has increased by 14 years to 28 years on average, led by a 30-year extension in Latin America (to 40 years). Table A2 shows country details. This has happened at a time when average maturities of government debt have declined in advanced economies. That said, the maturities of domestic currency government bonds are considerably shorter than those of foreign currency bonds in many EMEs, as highlighted by central bank papers from Colombia, Mexico and Turkey. Type A third dimension is a move away from types of debt that would increase interest rate and currency risks. Fixed rate issues have become increasingly important, representing around 90% or more of total in Asia in 2010 (Table 1). Despite steady increases, the share of fixed rate issues remains at around 40% of total in Latin America. Inflation-indexed bonds remain important in some Latin American economies. The Central Bank of Chile’s contribution mentions that a long history of inflation explains the prevalence of indexation in the sovereign’s long-term issuance. Inflation-indexed bonds also constitute a large share of total domestic debt in Israel. In contrast, the share of floating rate bonds issued by governments (which exposes them to interest rate risks) has declined over the past decade. However, floating rate government bonds remain relatively common in Latin America, particularly in Brazil and Mexico (30% and 29% of total, respectively). More importantly, the share of domestic foreign currency bonds has fallen notably and is no longer significant in Russia and Brazil, which relied on them heavily in the 1990s. However, in some Latin American economies, part of domestic debt 34 BIS Papers No 67 remains either denominated in or linked to foreign currencies (eg Argentina, Peru and Venezuela). Tables A3 and A4 report similar data for a larger number of issuers. Table 1 Instrument and maturity structure of general government debt outstanding in 2010 (in local currency)1 Overall Long-term (more than one year) Short- term debt3 Floating rate Fixed rate Inflation- indexed2 Foreign currency Floating rate Fixed rate Inflation- indexed2 Foreign currency Asia4 7 87 1 5 6 88 0 5 18 China 1 99 0 0 0 100 0 0 16 Indonesia 18 57 4 21 19 58 0 23 9 Korea 0 97 1 2 0 97 1 2 0 Singapore 0 100 0 0 0 100 0 0 52 Thailand 15 84 0 2 13 86 0 2 14 Latin America4 14 39 19 28 15 34 21 30 17 Argentina 10 8 24 57 11 0 27 61 16 Brazil 30 36 26 8 32 30 28 9 20 Colombia 0 65 21 13 0 64 22 15 13 Mexico 29 51 13 6 31 43 17 9 26 Peru 0 34 11 55 0 31 12 57 9 CEE4 7 74 0 19 9 69 0 23 17 Czech Republic 7 74 0 19 9 69 0 23 17 Other EMEs4 14 54 25 7 12 57 25 8 18 Israel 7 29 50 14 7 29 50 15 9 Saudi Arabia 21 79 0 0 16 84 0 0 26 Total of above4 11 63 12 15 11 61 12 17 17 1 As a percentage share; the total of the shares may fall below or exceed 100% due to rounding. For Indonesia, zero-coupon bonds. 3 As a percentage share of the overall debt outstanding. 4 Simple averages across listed countries. Source: Central bank responses to the BIS questionnaire. Investor base A fourth and welcome development is that the investor base for domestic government bond markets in EMEs is now more diversified than it was five to 10 years ago. As Graph 2 shows, the share of pension funds and insurers has risen as the gradual development of funded pension systems has helped stimulate demand for domestic bonds. The share of pension funds in 2010 exceeded one fifth of total in Chile, Colombia, the Czech Republic, Korea and Mexico, although even at that level it remains lower than in industrial countries. And, the share of foreign holdings has generally increased, including in Indonesia, Korea and Mexico. Foreign holdings remain generally high in CEE (eg the Czech Republic and Hungary), reflecting the region’s increasing financial integration with the rest of Europe. Despite these developments, banks are reported as the remaining key holders of domestic government bonds in EM Asia. In China, banks are almost the sole investor in government bonds and this has not changed over the past five years. The share of bank holdings exceeds 60% in Singapore. Elsewhere, the reported share of bank holdings is typically smaller. However, in some cases, the data may require careful interpretation. For instance, in BIS Papers No 67 35 Colombia and Mexico, part of “other residents” may represent brokers, which are typically owned by banks. Such shifts in investor composition should change the mix of maturities in demand, which could in turn affect the maturity structure of government debt, the shape of the yield curve, and interest rate level.3 Graph 2 Investor base for general government debt1 In percent 1 The shares are based on the data reported in local currencies. Source: Central bank responses to the BIS questionnaire. (iii) Market liquidity A final dimension is liquidity. Trading of domestic government bonds in EMEs has become more active. According to the Emerging Markets Trade Association, in the past decade, trading of these securities surged fivefold in EMEs, reaching some $4.7 trillion in 2010. As a result, trading volume of domestic debt securities as a share of total debt securities doubled during the same period, reaching 70% of total debt traded. However, liquidity remains an issue for many EME bond markets. Graph 3, which summarises data provided by central banks to the meeting, indicates that bid-ask spreads are often in single-digit basis points, especially in Asia and Latin America, suggesting that the costs of executing trades are not elevated. However, even for the most liquid maturities, turnover remains low relative to the average amount outstanding in many EMEs. The turnover ratios, computed as the ratio of the amount traded to the amount outstanding, are relatively high in Hong Kong SAR (29), Mexico (20) and South Africa (15). Market liquidity data from the recent Bond Market Liquidity Survey for Asian economies, and JP Morgan’s Local Markets Guide for a larger number of EMEs, broadly confirm the observations, despite differing in several details. 3 The issue is explored in the background paper “Central bank government bond markets: issues for monetary policy and coordination”. 36 BIS Papers No 67 Graph 3 Indicators of liquidity in government bond markets in 2010 Bid-ask spread (basis points)1,2 Turnover amount as a multiple of amount outstanding1 1 For the most liquid issue. See Table A5 for the underlying data. 2 Bid-ask spreads are expressed in basis points. For Japan, the United Kingdom and United States, average bid-ask spreads of generic 10-year government bonds in 2010. Sources: Bloomberg; central bank responses to the BIS questionnaire. 3. What factors have contributed to bond market development? The expansion of domestic government bond markets over the past 10 years can be attributed to improvements in domestic policy management and a reduction in external financing needs. Questions, nevertheless, arise about the sustainability of debt levels and whether EMEs have grown out of “original sin”. Many central banks have been able to keep inflation at low levels. As a result, nominal interest rates have fallen and become more stable. Over the last decade, yields on domestic government bonds have declined by some 4 percentage points to 6.5%, and their volatility has declined by two thirds (Table 2). In Brazil, domestic government bond yields came down from 26.1% to 11.8%, and, more impressively, their volatility fell from 11.2% to 0.9% during the same period. In Turkey, yields fell from 23.8% to 8.7% and their volatility from 7.7% to 1.1% during the same period. Low and stable inflation has helped to reduce the need for foreign currency borrowing. In the past, investors often preferred foreign over local currency debt to hedge themselves against inflation risks, as they feared that governments would generate surprise inflation to reduce the value of debt. BIS Papers No 67 37 Public debt sustainability has improved considerably owing to sounder fiscal policy, increasing the attractiveness of domestic currency bonds (see the background paper “Is monetary policy constrained by fiscal policy?”). Several EMEs have also taken advantage of these favourable developments to bring onto the government balance sheet some (non- marketable) debt that was previously concealed in various ways. In addition, vulnerabilities associated with foreign currency funding may have prompted several governments to consciously switch to domestic funding. Several EMEs have sought to avoid the consequences of sudden interruptions in capital flows as experienced during the 1990s, which led to major macroeconomic adjustments and episodes of financial crisis. Table 2 Domestic government bond yields1 Average Standard deviation 20022 2005 2010 20022 2005 2010 Asia3 6.5 6.0 5.3 1.0 0.8 0.6 China 4.7 3.5 3.3 1.9 0.7 0.4 India 7.4 7.1 7.9 0.9 0.5 0.4 Indonesia 12.2 12.1 8.9 1.2 1.9 1.3 Korea 6.2 4.6 4.4 0.9 0.9 0.8 Malaysia 4.0 4.1 3.7 0.5 0.3 0.2 Thailand 4.3 4.8 3.4 0.5 0.7 0.6 Latin America3 16.4 12.1 8.6 4.9 1.6 0.8 Brazil 26.1 17.5 11.8 11.2 2.0 0.9 Colombia 13.5 9.7 7.2 1.7 1.8 0.8 Mexico 9.6 9.1 6.9 1.7 1.0 0.7 Central Europe3 6.8 5.1 5.3 1.0 1.8 0.9 Czech Republic 4.6 3.3 3.5 0.7 3.3 0.5 Hungary 8.0 6.7 7.0 1.3 1.1 1.6 Poland 7.8 5.2 5.3 0.9 0.9 0.6 Other EMEs3 17.7 11.7 8.5 4.8 1.9 1.0 South Africa 11.5 7.8 8.3 1.9 1.0 0.9 Turkey 23.8 15.6 8.7 7.7 2.8 1.1 Total of above3 10.3 7.9 6.5 2.4 1.4 0.8 1 GBI EM Broad Diverse, or GBI all maturities. Based on daily data. Standard deviation of daily percentage point change, annualised. 2 2003 for Indonesia and Colombia, 2004 for China and Turkey. 3 Simple averages across listed countries. Source: Datastream. Increased domestic saving has boosted the pool of resources for investment in domestic capital markets and reduced the need for external borrowing.4 IMF data suggest that gross national savings as a percentage of GDP increased by 9 percentage points in EMEs as a whole (to 34%) during 2000–11. The increase was most pronounced in Asia where the average saving rate reached 46% in 2011. 4 Central banks have accumulated foreign exchange reserves, part of which has been financed by issuing their own securities. Table A6 shows that, in a number of countries, central bank debt securities now account for large shares of GDP. 38 BIS Papers No 67 The growth of government bond markets raises two interrelated issues: how far domestic bond markets can or should expand? Have EMEs grown out of so-called “original sin”? As for the first question, Reinhart, Rogoff and Savastano (2003) argue that many EMEs experience extreme duress with overall debt levels that may be considered low by the standard of advanced economies. In their view, EMEs face “debt intolerance”, and can accumulate only a relatively small amount of debt, be it external or domestic. Countries can improve their creditworthiness, but the process is typically arduous and slow. In countries suffering debt intolerance, the threshold for domestic government debt would be low, and any attempt to breach it would expose the economy to considerable risks. In addition, domestic government borrowing could crowd out private sector borrowing. As a result, governments would end up borrowing in domestic bond markets, forcing the private sector to access external markets. Evidence over the past decade has not been quite consistent with this prediction. Many EMEs now have public debt ratios above the 40% mark that was once considered unsustainable. This is because these EMEs have improved the health of their banking system, strengthened their fiscal positions, and accumulated large foreign currency reserves, which have improved their sovereign credit ratings. Turning to the second question, the proponents of original sin held the view that EMEs cannot borrow abroad in their own currencies (Eichengreen, Hausmann and Panizza (2005)). However, the increased take-up by global investors of domestic government bonds appears to have made this proposition less relevant today. For most EMEs, the share is in the range of 10–30% of total government debt, which remains low relative to the 50–70% range for major industrial countries.5,6 Foreign participation in domestic bond markets could accelerate in future as more EMEs have been included in a benchmark local currency government bond index for international investors, and as global investors reassess credit risk in favour of EMEs more generally.7,8 4. Implications for the conduct of monetary policy In the past, heavy burdens of foreign currency debt have limited the use of countercyclical monetary policy. As currency depreciation increased the liabilities of residents with large amounts of foreign currency debt, monetary policy had to focus on propping up the exchange rate rather than stabilising the economy. This was done by raising the policy rate, often very sharply. Matters were often made worse by debt with short maturities or floating rates. The growth in domestic government bond markets and the changes in their composition have contributed to a reduction in currency mismatches within the broader economies of many, if not all, countries. Table 3 reports three sets of indicators to help assess currency mismatches for 18 selected EMEs: 5 The shares are larger for some EMEs, as they are based on the amount of bonds included in the benchmark index or tradable debt, which is smaller than the total domestic government debt outstanding. 6 See “Global fixed income strategy”, 11 January 2012, JP Morgan Securities. 7 The Czech Republic, Hungary, Israel, Korea, Mexico, Poland and South Africa have been included in the JP Morgan GBI Broad Index. 8 However, the tendency of the price volatility on EM local debt to surge during times of stress could discourage foreign participation. This is partly because such characteristics reduce so-called collateral capacity – the scope for the underlying securities to be pledged as collateral for financing. Turner (2012) reports that the Sharpe ratios of EM government bonds have been higher than those of developed economies in relatively calm periods (such as 2002–06), but that they tend to fall sharply during periods of global financial stress. BIS Papers No 67 39 • The first four columns show the share of foreign currency debt in total outstanding debt. The ratio fell for 13 EMEs during the last decade, reflecting the increased importance of local currency government debt. • As the extent of risk stemming from foreign currency debt depends in part on the country’s net foreign currency liability position (that is, foreign currency liabilities minus foreign currency assets), the middle four columns show this measure as a share of exports. A country with a significant net positive position suffers a balance sheet loss when its currency depreciates. During the last decade, 15 EMEs either reduced net foreign currency liabilities or turned to holders of net foreign assets. • Finally, the net international investment position (NIIP) as a share of GDP, reported in the last column, represents a measure of an economy’s balance sheet. The sign is reversed such that, consistent with the first two indicators, positive values signify net liabilities. In 2011, only six EMEs had either net assets or small net liabilities. The rather noticeable discrepancy with the first two measures may stem partly from the fact that, in NIIP, the share of net non-financial and/or non-debt positions could be relatively large. Table 3 Measuring currency mismatches1 Foreign currency share of total debt outstanding2 Net foreign currency liabilities as a percentage of exports2, 3 Net IIP as a percentage of GDP3 2000 2005 2010 2011 2000 2005 2010 2011 20114 Asia China 4.2 2.7 2.5 2.9 –81.1 –110.7 –196.0 –189.5 –27.2 India 6.3 6.6 8.9 10.2 –46.5 –80.7 –57.1 –43.3 13.4 Indonesia 23.9 16.8 17.2 18.0 9.7 –11.2 –30.1 –20.9 40.8 Korea 9.7 7.8 9.8 10.1 –19.5 –31.3 –9.7 –7.5 8.7 Malaysia 15.2 16.2 9.0 9.5 –11.8 –15.5 –25.9 –29.8 –1.8 Philippines 34.3 34.9 26.9 25.2 36.6 30.1 –30.8 –46.4 5.3 Thailand 15.7 10.4 6.4 5.9 –5.7 –36.9 –66.9 –61.7 11.4 Latin America Argentina 46.5 34.4 27.2 25.2 275.3 80.2 –8.1 3.9 –12.5 Brazil 18.1 11.0 7.1 8.3 159.3 49.8 –33.7 –27.8 29.5 Chile 22.1 20.1 20.9 21.5 10.5 14.2 15.2 8.3 4.4 Colombia 29.2 19.6 14.5 15.4 29.9 –7.5 –18.9 –7.6 23.9 Mexico 26.5 17.0 17.0 18.8 4.9 –0.2 –3.4 28.5 Peru 34.1 34.9 35.3 35.7 –28.1 –33.1 –45.5 –35.9 20.0 Central Europe Hungary 33.5 30.5 34.3 34.9 43.5 38.0 46.2 30.9 85.5 Poland 15.2 20.0 20.5 21.0 –41.8 –1.0 32.0 24.6 55.1 Other EMEs Russia 45.3 41.3 21.4 18.9 18.9 –32.1 –90.1 –76.6 –1.1 Turkey 30.9 19.7 17.5 20.2 72.2 45.2 54.0 51.9 41.9 South Africa 10.2 6.9 7.2 8.8 16.1 –33.9 –34.5 –32.4 19.4 1 Using estimates of the currency of denomination of aggregate debt liabilities and assets (domestic as well as foreign). The net international investment position includes non-debt variables as well. Data in other columns represent mainly bonds, deposits and bank loans, but may include non-debt variables are well. 2 For detailed methodology, see Controlling currency mismatches in emerging markets, Goldstein and Turner (2004). 3 A negative sign indicates that international assets exceed liabilities. 4 For Argentina, Indonesia, Malaysia, the Philippines, Russia, South Africa and Thailand, 2010; for Chile, China, India and Peru, latest available quarter. Sources: IMF; national data; BIS. 40 BIS Papers No 67 In sum, economy-level currency mismatches have broadly declined in most Asian, Latin American and other EMEs. This contrasts with central Europe (Hungary and Poland), where currency mismatches appear to have increased. Nevertheless, the actual degree of such mismatches could depend on how far these balance sheet exposures are hedged in derivatives markets. Notwithstanding the reduction in currency mismatches, the choice of funding in local or foreign currency depends on several factors.9 The desirable level of foreign currency borrowing should be assessed against the country’s foreign currency revenues and assets (Goldstein and Turner (2004)). Also, the relative costs of borrowing in different currencies matter. In addition, the issuance of long-dated local currency bonds could be very costly if investors charged higher interest rates to compensate for inflation, currency depreciation and default risks as well as broader macroeconomic volatility. Graph 4 Currency mismatches and short-term domestic interest rates1 Brazil Korea Sources: Datastream and BIS estimates. 1 Policy and money market interest rates are in percent, shown on the left-hand scale. Currency mismatches represent foreign currency debt as a percentage of total debt, shown on the right-hand scale. Has the broad reduction in currency mismatches increased the scope for countercyclical monetary policy? Many EMEs cut interest rates rather sharply during the 2008–09 global recession which may have been difficult without past declines in their foreign currency liabilities. Brazil is a case in point. While the central bank raised interest rates during the 2001 global recession, it cut rates during the 2008–09 recession (Graph 4, left-hand panel). Korea is a similar case, if one compares developments during the 1998 crises with the more recent external shocks (Graph 4, right-hand panel). Indeed, our analysis suggests that monetary policy has become more countercyclical in many EMEs over the last decade. The cyclicality of monetary policy is gauged by the correlation coefficients between the cycle of the short-term interest rate around its trend and the output gap during 2000–11, in similar fashion to Vegh and Vuletin (2012). A positive correlation coefficient indicates that monetary policy is countercyclical: interest rates decline as growth slows. 10 Graph 5 shows the change in the correlation coefficients from 2000–05 to 9 Panizza (2009) discusses such trade-offs. 10 The analysis uses quarterly data for interbank interest rates and real GDP. The interest rate cycles and the output gap are constructed by extracting the cyclical component of the interest rate and real GDP series, respectively, with a conventional Hodrick-Prescott filter and smoothing parameter of 1,600. BIS Papers No 67 41 2006–11 for the various EMEs. In most economies, monetary policy has become more countercyclical over time, as the change in the correlation coefficients is positive for most economies, and highest for Malaysia and Turkey. Graph 5 Changes in countercyclicality of monetary policy from 2000–05 to 2006–11 Note: Countercyclicality of monetary policy is proxied by the correlation coefficient between short-term interest rate cycle and output gap. A positive value indicates that monetary policy became more countercyclical from 2000–05 to 2006–11. Sources: IMF; national data; BIS calculations. 5. Impact on financial stability Financial stability should have benefited from the development of domestic government bond markets described above – longer maturities, larger shares of fixed-rate issues, lower currency mismatches and greater market liquidity. It will also be affected by the two additional factors highlighted in this section, namely credit market diversification and the volatility related to greater foreign holding of domestic currency debt and derivatives markets – the former reducing risks and the latter presenting some additional risks. Diversification of credit risk A developed government yield curve allows the private sector to issue its own debt in the market, making the financial system and the broader economy more resilient to shocks. This is because domestic corporate bond markets help diversify credit risks away from banks and serve as an alternative form of intermediation to short-term credit markets. Such diversification has become increasingly apparent in EMEs. The outstanding stock of corporate bonds in most regions has increased since the mid-1990s, thus reinforcing their ability to serve as spare tyres (Graph 6). For instance, when banks in advanced economies tightened lending standards following the 2008 collapse of Lehman Brothers, non-financial corporate borrowers in EMEs turned to domestic markets for funding, taking the amount of domestic bonds outstanding to record levels. The Bank of Mexico notes in its paper that corporate domestic issuance has surged, and that mortgage-backed securities have particularly benefited from longer risk-free reference rates.11 11 See Figure 4.5 in “Banco de Mexico and recent development in domestic currency public debt”. 42 BIS Papers No 67 Notwithstanding the progress made so far, domestic corporate bond markets in EMEs remain underdeveloped. For instance, Goswami and Sharma (2011) note that, in emerging Asia, even large EMEs with sizeable corporate bond markets suffer from low trading volumes and very high transaction costs that inhibit arbitrage and active position-taking. Graph 6 Outstanding stock of domestic non-financial corporate debt securities1 As a percentage of nominal GDP Asia2 Latin America 3 Central and Eastern Europe 4 1 Simple averages across the countries listed. 2 China, Hong Kong SAR, India, Indonesia, Korea, Malaysia, the Philippines, Singapore and Thailand. 3 Argentina, Brazil, Chile, Colombia, Mexico, Peru and Venezuela. 4 Czech Republic, Hungary, Poland, Russia and Turkey. Sources: IMF, International Financial Statistics; IMF, World Economic Outlook. Foreign holdings and derivatives markets Recent episodes of market volatility have highlighted two kinds of risks associated with more developed domestic government bond markets. First, large foreign holdings may increase financial market volatility during times of stress. As suggested by the Reserve Bank of South Africa, foreign portfolio inflows may be driven by carry trade incentives against the backdrop of very low interest rates in advanced economies. Some foreign investors leave their currency risk unhedged for higher total returns on the expectation that particular EM currencies are managed at relatively weak levels and should appreciate, or in order to benefit from diversification. Such a strategy makes carry trade flows inherently sensitive to currency performance in EMEs. A second source of worry is the potential risk in the derivatives markets.12 As the Bank of Mexico points out, derivatives add liquidity and depth to domestic bond markets as they offer hedging possibilities and expand the demand for the underlying assets. However, to the extent that residents hedge their financial risk with other residents without involving foreign counterparties, exposures are shifted across balance sheets within the economy. In addition, financial stability risk would be greater if exposures ended up concentrated in a small number of residents. Derivatives products allow both residents and non-residents to take complex and leveraged positions that may be rapidly unwound in the event of market turmoil. 12 The expansion of derivatives markets in EMEs has been documented by Mihaljek and Packer (2010) and Saxena and Villar (2008). BIS Papers No 67 43 These risks have prompted many EMEs to beef up existing measures and introduce alternative instruments to limit vulnerabilities in the domestic bond markets and safeguard financial stability. As one line of defence, EMEs have accumulated precautionary official reserves that could be drawn down in times of market stress. A second and complementary line of defence in some cases has been to establish currency swap lines with the major central banks. For instance, in 2008–09, Brazil, Korea, Mexico and Singapore established currency swap lines with the Federal Reserve. These agreements, which expired in April 2009, played a pivotal role in calming markets over possible foreign currency shortages, particularly where the precautionary function of official reserves was believed to be limited, as in the case of Korea. Finally, many countries have introduced measures to help increase the resilience of their domestic financial systems to credit exposures. In some cases, measures to manage capital inflows have also been established (eg taxes on inflows, minimum holding periods and currency-specific reserve requirements). 44 BIS Papers No 67 Table A1 Outstanding stocks of domestic government debt securities1 Billions of US dollars Average annual percentage change Tradable debt2 2000 2005 2010 2000–04 2005–10 Short-term3 Long- term4 Short- term3 Long- term4 Short- term3 Long- term4 Short- term3 Long- term4 Short- term3 Long- term4 Asia 32 402 58 1010 107 2311 12 17 18 14 44 China 0 111 0 335 0 1006 0 30 0 24 37 Hong Kong SAR 9 6 9 9 18 12 1 8 62 7 22 India 4 108 18 250 27 581 33 19 13 18 38 Indonesia 0 45 0 40 3 65 0 2 43 8 56 Korea 0 62 0 231 0 331 0 27 0 12 78 Malaysia 1 27 1 50 1 124 0 14 4 18 53 Philippines 9 11 12 28 12 50 –1 18 3 13 2 Singapore 8 17 13 34 44 58 11 18 25 11 51 Thailand 1 15 5 33 2 84 49 18 13 18 59 Latin America 211 145 393 246 634 712 12 16 19 28 42 Argentina 5 28 0 48 0 31 –23 10 0 2 1 Brazil 173 46 328 88 520 429 17 3 17 40 22 Chile5 – 1 – 3 – 17 – 41 – 50 68 Colombia 12 4 34 6 66 4 18 19 19 –3 61 Mexico 19 56 28 87 46 201 11 12 14 19 42 Peru 0 4 0 6 0 14 0 6 0 20 60 Venezuela 2 6 3 8 2 16 16 23 63 37 - CEE 14 42 19 145 23 247 18 38 1 11 69 Czech Republic 5 3 5 21 6 46 10 57 –1 17 76 Hungary 3 13 7 36 8 42 23 27 0 2 67 Poland 6 26 7 88 9 159 21 31 1 13 64 Other EMEs 6 104 19 263 26 395 32 24 14 13 40 Russia 0 8 0 25 0 67 0 26 0 22 37 South Africa 3 44 6 68 20 105 22 11 28 10 60 Turkey 3 52 13 170 6 223 74 34 –12 8 24 Total of above 263 693 489 1664 790 3665 16 21 16 18 47 1 Central bank issues are excluded. Domestic debt securities in the BIS securities statistics are defined as issues by residents in the local market in local currency, targeted to resident investors. Some foreign currency issues are included in these data, but they are small. The size of the debt stock often reflects circumstances unique to the respective economies. In Mexico, the numbers include debt resulting from the rescue of the banking sector, originally issued off-balance sheet but now included in the government balance sheet. In Brazil, part of the increase in debt represents conversion of former central bank issues into government debt. 2 As a percentage share of total debt outstanding. The tradable amounts used are the end-2010 par values taken from JP Morgan Government Bond Index-EM, except for the Czech Republic, Hong Kong SAR, Korea and Singapore (Global Bond Index). 3 Bonds and notes. 4 Money market instruments. 5 For Chile, figures were taken from the Ministry of Finance: for more information please see http://www.minhda.cl/oficina-de-la-deuda-publica/estadisticas/ composicion-de-la-deuda.html. Sources: BIS securities statistics; JPMorgan Chase; national data. BIS Papers No 67 45 Table A2 Maturities of general government local currency debt1 In years Average Longest 2000 2005 2010 2000 2005 2010 Asia 3.7 4.8 5.7 15.4 16.9 24.5 Indonesia 5.0 7.0 8.1 8.6 14.7 27.2 Korea 2.6 3.9 5.0 19.0 13.3 20.0 Philippines 3.3 3.0 6.2 24.9 24.6 25.0 Singapore 2.7 3.6 3.3 10.0 15.0 20.0 Thailand 4.9 6.4 5.7 14.5 16.9 30.5 Latin America 2.1 7.9 9.6 9.1 39.2 39.7 Argentina 2.1 13.4 11.1 7.0 84.0 79.0 Brazil 2.7 2.3 3.4 – – – Chile – 17.0 14.0 – – – Colombia 2.5 3.8 5.1 11.4 14.6 13.6 Mexico 1.0 3.0 6.0 9.0 19.0 30.0 Peru – – 18.0 – – 36.0 Central Europe 2.6 4.7 5.0 18.0 18.0 31.5 Czech Republic 1.5 4.9 5.7 10.0 15.0 47.0 Hungary 3.7 4.5 4.2 26.0 21.0 16.0 Other 5.1 5.1 5.7 13.5 15.8 19.5 Israel 6.2 6.1 6.2 15.0 20.0 31.0 Saudi Arabia 4.3 4.5 2.5 10.0 10.0 7.0 South Africa 8.7 7.8 10.3 27.0 28.0 30.0 Turkey 1.2 2.1 3.6 2.0 5.0 10.0 Total 3.5 5.8 7.0 13.9 21.5 28.2 1 Data for Argentina, Colombia, Philippines, South Africa and Turkey represent central government debt. Annual data on the maturity of domestic central government debt spanning 1995–2010 are available at http://www.bis.org/statistics/qcsv/cgfs28d4 Source: Central bank responses to the BIS questionnaire. 46 B IS P apers N o 67 Table A3 Domestic bonds by instrument1 As percentage of total outstanding 2000 2005 2010 Floating rate Straight fixed rate Inflation- indexed Exchange rate- linked Floating rate Straight fixed rate Inflation- indexed Exchange rate- linked Floating rate Straight fixed rate Inflation- indexed Exchange rate- linked Asia 12 88 1 0 9 91 0 0 4 96 0 0 China 46 54 ... ... 19 81 ... ... ... ... ... ... Chinese Taipei 0 100 0 0 0 100 0 0 0 100 0 0 Hong Kong SAR 0 100 0 0 3 97 0 0 0 100 0 0 India 0 100 0 0 5 95 0 0 2 98 0 0 Indonesia 51 42 8 0 53 47 0 0 22 78 0 0 Korea 8 92 0 0 3 96 0 0 3 95 0 1 Malaysia 0 100 0 0 0 100 0 0 0 100 0 0 Philippines 8 92 0 0 4 96 0 0 2 98 0 0 Singapore 0 100 0 0 0 100 0 0 0 100 0 0 Thailand 4 96 0 0 3 97 0 0 3 97 0 0 Latin America 30 12 25 27 23 28 29 19 18 41 28 13 Argentina 12 0 0 88 1 10 68 18 18 16 38 26 Brazil 58 15 6 21 60 21 16 3 44 31 24 1 Chile 0 0 92 8 0 18 64 18 0 24 76 0 Colombia 0 50 41 7 0 70 29 1 0 75 25 0 Mexico 43 6 16 0 58 27 15 0 39 39 22 0 Peru 0 14 18 68 0 53 12 35 6 66 8 20 Venezuela 100 0 0 0 44 0 0 56 22 36 0 42 Central Europe 18 80 2 0 9 90 1 0 11 86 1 2 Czech Republic 0 95 5 0 0 100 0 0 10 90 0 0 Hungary 34 66 0 0 10 90 0 0 9 86 0 5 Poland 20 80 0 0 17 81 2 0 14 83 3 0 Other 9 62 12 16 15 61 15 9 14 65 16 2 Israel 11 17 61 11 16 26 50 6 6 26 52 6 Russia 0 37 0 63 0 76 3 22 0 96 1 3 Turkey 24 70 0 6 31 42 11 15 36 47 15 2 Saudi Arabia 9 91 0 0 17 83 0 0 15 85 0 0 South Africa 1 97 0 0 9 77 9 0 11 71 14 0 Total EMEs 17 61 10 11 14 67 12 7 11 72 12 4 1 Comprises only bonds and notes and excludes money market instruments. Regional totals based on the simple averages of the countries listed in the table. Asia and total emerging markets exclude China for all periods. Totals do not add up to 100% due to the exclusion of hybrid instruments. Ratio calculated taking the central government and all other issuers as reported in Table 2d of the Working Group questionnaire. Source: Update of CGFS Papers no 28 on local currency bond markets; Working Group survey. BIS Papers No 67 47 Table A4 Domestic exchange rate-linked bonds1 As percentage of total outstanding 2000 2005 2006 2007 2008 2009 2010 Asia 0 0 0 0 0 0 0 Korea 0 0 0 1 2 1 1 Latin America 27 19 11 18 18 13 13 Argentina 88 18 20 22 25 26 26 Brazil 21 3 1 1 1 1 1 Chile 8 18 13 2 0 0 0 Colombia 7 1 0 0 0 0 0 Peru 68 35 29 17 25 19 20 Venezuela 0 56 14 82 74 43 42 Central Europe 0 0 0 2 2 2 2 Hungary 0 0 0 5 6 5 5 Other EMEs 16 9 7 5 4 3 2 Israel 11 6 7 8 9 7 6 Russia 63 22 14 10 4 3 3 Turkey 6 15 13 9 8 5 2 Total EMEs 11 7 4 6 6 4 4 Note: For China, Chinese Taipei, Czech Republic, Hong Kong SAR, India, Indonesia, Malaysia, Mexico, Philippines, Poland, Saudi Arabia, Singapore, South Africa and Thailand percentage shares are equal to zero throughout the years shown. 1 Comprises only bonds and notes and excludes money market instruments. Regional totals based on the simple averages of countries listed in the table and the footnote. Ratio calculated taking the central government and all other issuers as reported in Table 2d of the Working Group questionnaire. Source: same as Table A3. 48 BIS Papers No 67 Table A5 Indicators of liquidity in government bond markets1 Ratio of turnover to average outstanding stocks in 2010 Most liquid (important) maturities Typical bid-ask spread on the most liquid issue2 Asia Hong Kong SAR 29.0 2, 5, 10 years 5 Korea 1.4 3 years 1 Philippines 0.7 2, 5, 7, 10 years 3 Singapore 0.5 2 years 5 Thailand 0.7 5 years 7 Latin America Argentina 1.7 20153 77 Chile 0.2 10 years 4 Colombia 0.3 9–10 years 4 Mexico 20.0 Dec 20243 2 Central and Eastern Europe Czech Republic 0.1 20243 10 Hungary 2.8 2015; 20193 40 Other emerging markets Israel 0.6 7–11 years 5 South Africa 15.0 20163 3 1 Only the maturity with the highest turnover for each country is shown. 2 In basis points. 3 Maturing in the indicated year. Source: Central bank responses to the BIS questionnaire. BIS Papers No 67 49 Table A6 Central bank bonds As percentage of GDP 2005 2010 Asia1 6.4 13.3 Hong Kong SAR 9.2 37.4 Indonesia 2.6 3.1 Korea 18.2 14.2 Thailand 8.3 24.8 Latin America1 4.6 3.0 Argentina 4.8 5.1 Brazil 0.3 0.0 Chile 16.3 9.4 Mexico 2.8 2.9 Peru 3.3 0.9 Central Europe1 11.5 15.3 Czech Republic 22.9 19.4 Hungary 0.1 11.1 Other1 4.4 8.9 Israel 14.1 17.5 Saudi Arabia 3.3 16.8 South Africa 0.3 1.2 Note: For Colombia, Philippines, Singapore and Turkey percentage shares are equal to zero for the years shown. 1 Simple averages of economies listed. Sources: Central bank questionnaires; IMF, World Economic Outlook. 50 BIS Papers No 67 References Bank for International Settlements (2002): “The development of bond markets in emerging economies”, BIS Papers, no 11, June. ______ (2007): “Financial stability and local currency bond markets”, CGFS Papers, no 28, June. Eichengreen, B, R Hausmann and U Panizza (2005): “The pain of original sin” in B Eichengreen and R Hausmann (eds), Other people’s money: debt denomination and financial instability in emerging market economies, University of Chicago Press. Goldstein, M and P Turner (2004): “Controlling currency mismatches in emerging markets”, Institute for International Economics, Washington DC, April. Goswami, M and S Sharma (2011): “The development of local debt markets in Asia”, IMF Working Paper, 11/132. JP Morgan Securities (2012): “Global fixed income strategy”, 11 January. Mihaljek, D and F Packer (2010): “Derivatives in emerging markets”, BIS Quarterly Review, December. Panizza, U (2009): “Is domestic debt the answer to debt crisis?”, in B Herman, J-A Ocampo and S Spiegel (eds), Overcoming Developing Country Debt Crises, Oxford University Press. Reinhart, C, K Rogoff and M Savastano (2003): “Debt intolerance”, NBER Working Paper Series, no 9908, August. Saxena, S and A Villar (2008): “Hedging instruments in emerging market economies”, in “Financial globalisation and emerging market capital flows”, BIS Papers, no 44. Turner, P (2012): “Weathering financial crisis: domestic bond markets in EMEs”, in BIS Papers, no 63, January. Vegh, C and G Vuletin (2012): “Overcoming the fear of free falling: monetary policy graduation in emerging markets”, mimeo. BIS Papers No 67 51 Central bank and government debt management: issues for monetary policy Andrew Filardo, Madhusudan Mohanty and Ramon Moreno1 Abstract The size and maturity structure of the government debt market have important implications for monetary policy, especially in EMEs. This paper documents the remarkable growth of the market over the past decade in terms of size, issuance and maturity structure of combined government and central bank debt, and notes that a large part of the official sector debt constitutes short-term securities issued by central banks to sterilise their foreign exchange interventions. The paper then explores what these trends imply for the yield curve and for bank lending behaviour, and highlights potential conflicts that may arise when the mandates of government debt managers differ from those of the central bankers. Keywords: Debt management, monetary policy JEL classification: H63, E52 1 The authors thank Tracy Chan, Emese Kuruc, Lillie Lam, Agne Subelyte and Alan Villegas for statistical assistance. 52 BIS Papers No 67 Introduction In most emerging market economies, both governments and central banks are active in sovereign bond markets. Governments issue debt of various maturities to finance fiscal deficits. Central banks issue their own securities to finance the acquisition of assets (particularly foreign exchange reserves). They also conduct open market operations, which involve sales and purchases of government debt. As a result, both the government and the central bank directly influence the mix of short- and long-term securities held by the public. The actions of debt managers in choosing the maturity structure of their debt could have effects that are akin to monetary policy. In principle, the maturity decisions of debt managers are not influenced by explicit macroeconomic or financial market objectives. In practice, however, debt management decisions are to some extent discretionary – and depend on an assessment of market conditions. From this perspective, decisions about the consolidated debt of the official sector (government and the central bank) will determine the size and the maturity structure of debt held by the private sector and, given imperfect substitutability of assets along the maturity spectrum, this will normally influence the shape of the yield curve. For a given path of expected future short-term interest rates, the term premia will thus be affected. This was central to Tobin’s (1963) portfolio balance model.2 A further complication is that banks’ lending behaviour may be influenced by the scale of government bond holdings on their balance sheets. A practical challenge is that debt issuances by the central bank and by the government might at times work at cross purposes. Central banks are assigned the goal of macroeconomic stabilisation (ie price stability) while debt managers are typically mandated to keep the government’s funding costs to a minimum. Thus, while the government would like to issue most of its debt in long-term paper to reduce the need to roll it over, central banks may have a strong preference for short-term bills for their day-to-day liquidity operations. This could lead to undesirable consequences for the monetary transmission mechanism through the term structure. The rest of the note is structured as follows. Section 1 provides a brief review of the developments in the size and the maturity structure of government and central bank debt. Section 2 presents estimates of consolidated public sector debt. Section 3 discusses the potential consequences of debt maturity for the monetary transmission mechanism. Section 4 concludes with some remarks on potential coordination challenges facing the central bank and the government. 1. The size and maturity structure of government and central bank debt In this section, we touch only briefly on government debt, since this topic has been extensively covered in the background paper Developments of domestic government bond markets in EMEs and their implications. We highlight the central bank issuance of bills and bonds, before presenting estimates of consolidated public debt securities. 2 Modigliani and Sutch’s (1966) “preferred habitat” hypothesis points in a similar direction (see also Friedman (1978)). The portfolio balance channel is absent in the standard New Keynesian models. BIS Papers No 67 53 The size and maturity of government debt Table 1 summarises the main facts about government debt – the amount is growing and the maturity lengthening. The outstanding stock of domestic debt securities issued by emerging market (EM) governments as a percentage of GDP has increased in all regions over the past decade. Table 1 Government debt and its maturity1 Outstanding domestic government debt securities, as a percentage of GDP2 Average maturity, in years3 2000 2005 2010 2000 2005 2010 Asia 19.0 25.8 29.4 4.3 4.7 5.7 China 9.2 14.9 17.1 8.0 6.6 8.1 Hong Kong SAR 9.0 9.9 13.7 India 23.4 33.1 37.3 Indonesia 25.1 14.2 10.3 Korea 26.2 32.9 2.6 3.9 5.0 Malaysia 34.9 48.6 Philippines 25.5 39.0 30.9 3.3 3.0 6.2 Singapore 26.5 37.4 46.1 2.7 3.6 3.3 Thailand 13.9 22.3 27.7 4.9 6.4 5.7 Latin America 11.6 26.5 26.2 2.1 7.9 9.6 Argentina 11.6 26.1 8.4 2.1 13.4 11.1 Brazil 58.0 65.2 2.7 2.3 3.4 Chile 17.0 14.0 Colombia 16.7 27.2 24.3 2.5 3.8 5.1 Mexico 11.2 13.6 23.9 1.0 3.0 6.0 Peru 6.8 7.6 9.1 18.0 Central & eastern Europe 21.8 23.6 26.5 2.5 4.2 4.9 Czech Republic 13.8 20.9 27.0 1.5 4.9 5.7 Hungary 32.7 38.9 38.4 3.5 3.6 4.0 Poland 18.7 31.5 35.9 Russia 3.3 4.6 Other 33.0 34.9 38.8 5.4 5.3 6.7 Israel 31.2 42.4 51.1 6.2 6.1 6.2 South Africa 34.8 30.2 34.5 8.7 7.8 10.3 Turkey 32.1 30.6 1.2 2.1 3.6 Total average 19.4 26.8 29.4 3.6 5.8 7.2 1 Outstanding government debt (taken from China Central Depository & Clearing Co Ltd, BIS debt securities statistics and national data) and average maturity taken from central bank questionnaire. 2 Data for Argentina, Hong Kong SAR, Peru, the Philippines, Poland and Russia represent central government debt. 3 Data for Argentina, Colombia, Indonesia, the Philippines and Turkey represent central government debt. Sources: Central bank questionnaires; China Central Depository & Clearing Co Ltd; IMF, World Economic Outlook; BIS debt securities statistics. In addition, there has been a significant rise in the average maturity of outstanding government debt, from 3.6 years in 2000 to 7.2 years in 2010. This is most striking in Latin 54 BIS Papers No 67 America and, to a lesser extent, in Asia and central and eastern Europe (CEE). Nevertheless, differences across countries are large. In 2010, the average maturity of government debt was above 10 years in Argentina, Chile, Peru and South Africa, but was much shorter in Brazil, Hungary, Korea, Singapore and Turkey (between three and five years). Taking EMEs as a whole, the proportion of government debt with maturity below one year declined from about 28% in 2000 to around 18% in 2010. The size and the maturity structure of central bank debt Table 2, based on central bank questionnaire responses, shows that many EME central banks are major issuers of their own bills and debt securities. Most central bank issuance has had short maturities, with a heavy concentration below one year. In recent years, some central banks have tried to lengthen the maturity, but low investor appetite for duration has represented a challenge. Table 2 Central bank securities Total outstanding Maturity distribution at end-2010 2000 2005 2010 Below 1 year Between 1 and 3 years Above 3 years Average remaining maturity As a percentage of GDP Percentage of the total outstanding In years Asia China1 0.0 12.2 10.3 70.3 29.7 0.0 Hong Kong SAR 8.2 9.2 37.5 91.9 4.5 3.6 0.5 Korea 11.0 17.9 13.9 63.5 36.5 0.0 0.8 Thailand n/a 8.4 23.6 68.0 26.0 6.0 1.0 Latin America Argentina 0.0 4.6 4.7 88.1 11.9 0 0.5 Brazil2 7.3 0.3 Chile 29.9 15.0 8.6 25.9 36.6 37.6 3.4 Colombia3 Mexico4 0.0 2.7 2.7 61.0 36.0 3.0 1.1 Peru5 0.7 3.4 0.8 100.0 0.0 0.0 0.3 Other EMEs Czech Republic 18.3 23.5 19.1 100.0 0.0 0.0 0.0 Hungary6 3.5 0.2 11.3 100.0 0.0 0.0 0.0 Israel 5.7 14.5 18.4 100.0 0.0 0.0 0.5 South Africa 0.3 1.0 100.0 0.0 0.0 1 www.chinabond.com.cn/Site/cb/en. 2 Pursuant to the Brazilian Fiscal Responsibility Law, since 2002 the Central Bank of Brazil has not been able to issue its own securities. 3 The central bank can issue its own debt securities, but it has not issued any yet. 4 The Bank of Mexico can issue its own debt to meet its objectives. However, the federal government issues bonds on behalf of the central bank so that the bank can undertake open market operations. At the end of 2010, debt issued by the federal government for monetary regulation purposes represented about 99% of the outstanding central bank bonds. 5 Includes all types of certificates of deposit. 6 Currently, the only debt security issued by the central bank (Magyar Nemzeti Bank, MNB) is the main liquidity-absorbing policy instrument, namely the two-week MNB bill. Source: Central bank questionnaires; China Central Depository & Clearing Co Ltd. BIS Papers No 67 55 There is significant cross-country variation, with the proportion of central bank debt to GDP ranging from lows of around 1% in Peru and South Africa to highs of 24–38% in Hong Kong SAR and Thailand. Between 2005 and 2010, when there were extended periods (with some sharp interruptions) of capital inflows, there were large increases in the amount of central bank debt issued in some jurisdictions, such as Hong Kong SAR, Hungary, Israel and Thailand. In some cases (eg Singapore and to some extent Chile), growth in the overall issuance of central bank securities has reflected an objective, shared by finance ministries, to deepen government debt markets and establish an effective benchmark yield curve. This has particularly been the case where the outstanding supply of marketable domestic government bonds has been rather limited. In most countries, however, central bank debt issuance in recent years has been a by- product of exchange rate objectives. The sterilisation of the massive central bank purchases of foreign currency assets requires tools to drain the associated increase in domestic money market liquidity. While central banks have many tools to address this need, the issuance of central bank bills has been a relatively attractive option. There are several reasons for this. • Government deposits: In principle, government deposits can play an important role in helping a central bank achieve its policy rate target.3 Indeed, many central banks have sought arrangements with governments that improve the predictability of government deposits. Nevertheless, government deposits remain volatile, reflecting variation in the timing of tax payments and government expenditures as well as portfolio allocation decisions by debt managers. • Operations with government securities: During periods of large capital inflows, the supply of bank reserves initially rises in response to foreign asset accumulation, and the demand for bank reserves will tend to fall as domestic banks and non-banks find cheaper funding from abroad. Central banks in this situation sometimes find themselves running short of government securities to drain liquidity via repo operations or outright sales and must seek alternatives. • Required reserves: Reserve requirements remain an attractive option for various reasons: (i) they may be easier to implement; (ii) they do not tend to attract capital inflows as much as higher interest rates; and (iii) they provide some financial stability benefits (for a discussion, see Moreno (2011)). However, required reserves cannot be used as flexibly by commercial banks as central bank securities, which can be pledged as collateral. Furthermore, unremunerated bank reserves impose costs on banks. Finally, the motives for issuing short-term central bank debt deserve to be noted. First, some central banks and finance ministers have agreements to segment markets by maturity so as not to compete for the same investors. Second, the short end of the yield curve is often the most liquid, which reduces funding costs and hence carrying costs and interest rate risks for central banks. Third, short-term securities provide operational flexibility as liquidity conditions change in money markets; for example, during periods of financial stress, the central bank can supply liquidity in part by not rolling over its short-term debt. 3 For example, in some countries where the government budget is in surplus, such as Singapore, government deposits with the monetary authority can contribute to the draining of liquidity. Filardo and Grenville (2011) show that in 2010 government deposits accounted for 6–10% of total central bank liabilities in China, Indonesia and Korea. In some cases (eg India and Mexico), government deposits are part of an explicit arrangement with finance ministries to assist central banks in stabilising monetary conditions; see eg the contribution of the Bank of Mexico to this meeting. 56 BIS Papers No 67 2. The consolidated official debt held by the public The actions of government debt managers and central banks jointly determine the size and maturity of sovereign debt held by the public. To illustrate this point, note that if the central bank were to purchase the debt directly from the debt managers or indirectly from markets, there would be no change in the size or maturity of outstanding stocks held by the public.4 The left-hand panel of Graph 1 shows estimates of the consolidated official debt (central bank and government) held by the public as a percentage of GDP for 2005 and 2010. The right-hand panel shows the approximate maturity distribution of that debt at the end of 2010 (based on unconsolidated debt). Two facts stand out from the graph. First, consolidated official debt held by the public relative to GDP has increased in several, but not, all economies. Consolidated official debt held by the public is generally smaller in economies where central banks do not issue their own securities (eg India) but hold a significant share of government bonds on their balance sheets. In economies where both the government and the central bank issue their own securities, consolidated official debt as a share of GDP has increased significantly, especially in Hong Kong, Hungary, Mexico, Singapore, South Africa and Thailand. Graph 1 Outstanding central bank and government debt Ratio to GDP1 Maturity in 2010 (In per cent)2 AR = Argentina; BR = Brazil; CN = China; CO = Columbia; CZ = Czech Republic; HK = Hong Kong SAR; HU = Hungary; IL = Israel; IN = India; KR = South Korea; MX = Mexico; PE = Peru; PH = Philippines; SA = Saudi Arabia; SG = Singapore; TH = Thailand; TR = Turkey; ZA = South Africa. 1 The government debt is subtracted by central bank claims on central government (IMF, International Financial Statistics line 12a). 2 Central bank bonds with a maturity of greater than three years are assumed to have a maturity of less than or equal to five years. Sources: Results taken from central bank questionnaire, complemented where necessary with information from IMF World Economic Outlook; China Central Depository & Clearing Co, Ltd; BIS debt securities statistics. Second, the share of outstanding official short-term debt is high in many countries. Debt with maturity of less than one year accounted for about 37% of total EM consolidated official debt in 2010. The ratio is much higher than the average in Hong Kong SAR and Saudi Arabia. In several countries (eg Korea, Brazil and Turkey), there is a high concentration of debt with maturity below five years. The share of debt with maturity above five years is more significant in Argentina, Israel, Peru, the Philippines and South Africa than in other countries. 4 See Appendix Table A1 for a simple stylised public sector balance sheet. BIS Papers No 67 57 Short-term debt and money One fundamental question in thinking about the consequences of the maturity structure for monetary policy is the relationship between short-term government debt and monetary conditions. Short-term government debt can be a close substitute for money. Historically, the monetary authorities have often expressed their concerns about the impact of the sovereign issuance of very short-term treasury bills on the monetary policy stance. Until the mid-1990s, for instance, the Deutsche Bundesbank took the view that the government should finance itself with medium- and long-term securities only. Several countries have imposed issuance ceilings on bills. Table 3 Composition of central bank and government debt held by the public in 2010 Central bank and government debt1 Monetary base2 Total Monetary base and short-term debt � 1year > 1 year
(a) (b) (c) a + b + c = (d) = (a + c) / d
In billions of US dollars Ratio
China 533 857 426 1816 0.53
Hong Kong SAR 96 19 58 172 0.89
Indonesia 24 36 25 85 0.58
India 23 503 180 707 0.29
Korea 126 338 29 492 0.31
Malaysia 38 99 13 150 0.34
Philippines 11 45 11 67 0.33
Thailand 64 91 24 179 0.49
Brazil 292 241 42 574 0.58
Colombia 10 65 17 91 0.29
Mexico 64 213 –13 263 0.19
Czech Republic 43 46 15 104 0.56
Hungary 22 41 6 70 0.41
Poland 35 159 31 225 0.29
South Africa 24 103 –6 122 0.15
Turkey 59 160 37 255 0.37
Memo
Japan 1,677 7,744 1,111 10,532 0.26
United
Kingdom 87 1,561 298 1,946 0.20
United States 1,679 9,243 1,650 12,572 0.26
1 Central bank debt (taken from central bank questionnaire, complemented for China with information from
China Central Depository & Clearing Co Ltd and for Indonesia, Japan, Malaysia, Poland, the United Kingdom
and the United States with BIS debt securities statistics and national data) and government debt (taken from
BIS debt securities statistics and national data, complemented for Colombia with central bank questionnaire
data, and China from China Central Depository & Clearing Co Ltd) less central bank claims on central
government (as reported in IMF IFS, line 12a, except for Hong Kong SAR and the United Kingdom for which
these data are not available). 2 Monetary base less central bank liabilities to central government (taken from
IMF IFS, lines 14 and 16d; NB: line 16d is not available for the United Kingdom) less required reserves from
national sources. For China required reserves are proxied by bank deposits with the central bank (taken from
national data) and for Colombia, Czech Republic, Hungary, Philippines, Poland and Thailand proxied by central
bank liabilities to other depository corporations (as reported in IMF IFS, line 14c).
Sources: Central bank questionnaire; IMF, International Financial Statistics; China Central Depository &
Clearing Co Ltd; national data; BIS debt securities statistics; BIS calculations.

58 BIS Papers No 67

Tobin (1963) provided a framework to study the effects of debt maturity from a monetary
policy point of view. In his view, banks consider short-term government bonds as close
substitutes for excess reserves because they are subject to little capital loss and can be
easily sold to finance new lending. Given this view, a complete analysis of monetary
conditions should focus on both short-term government debt and base money.
As an illustration, Table 3 provides estimates of the short-term official sector liabilities
(ie central bank and government debt and the monetary base). The first two columns provide
a breakdown of consolidated public debt into short- and long-term securities in US dollar
terms. The third column reports the base money after netting out government deposits with
the central bank as well as required reserves of commercial banks, which are, for all practical
purposes, liquidity lost to the banking system. The fourth column shows the total liabilities,
which are the sum of base money and total official sector debt securities. The last column
reports the ratio of the sum of short-term debt and base money to total liabilities.
It is clear from Table 3 that shares of short-term official sector liabilities are high in many
EMEs, and generally exceed those seen in industrial countries. This is not surprising given
that the average debt maturity is shorter in emerging market economies than in industrial
economies. It also suggests that, taking a wider perspective, monetary conditions in many
EMEs are more accommodative than suggested by the monetary base alone.
Focusing on Asia, by this measure about 53% of official sector liabilities in China are
short-term. In Hong Kong SAR, the ratio is close to 90%, but for a special reason. Under the
currency board arrangement, all interventions are, in principle, unsterilised: monetary base
therefore closely mirrors the Hong Kong Monetary Authority’s purchase and sale of foreign
currency. In the rest of Asia, the share of short-term official liabilities is above 65% in
Indonesia and close to 50% in Thailand. In Latin America, Brazil’s short-term financing ratio
is significantly higher than other countries. In CEE, the same is true for the Czech Republic
and Hungary.
3. Implications for the monetary transmission mechanism
How do these changes in the size and maturity of official debt held by the public affect the
monetary transmission mechanisms? While not attempting to measure the impact, in what
follows we briefly discuss the potential implications for the short end and shape of the yield
curve as well as bank credit.
The short end of the yield curve
In principle, the short end of the yield curve is pinned down by the policy rate, which
determines other borrowing and lending rates. In practice, however, in several countries an
excess supply of bank reserves appears to have driven the interbank rate below the policy
rate. A case in point is Colombia (Graph 2). From March to June 2007, the central bank
bought foreign assets in response to capital inflows in order to stem the peso’s appreciation,
resulting in increased liquidity in domestic markets. In this setting, the short-term interbank
rate persistently fell below the policy rate target. In Colombia the floor is set by a Lombard-
type facility5 at a rate 1 percentage point below the policy rate target.
To expand the analysis, Graph A1 in the appendix shows interbank spreads for a number of
other EMEs. In several cases, the interbank rates have deviated significantly from the policy

5 In Colombia, facilities that set a ceiling and a floor to movements in interbank rates around the target are both
called Lombard facilities. At other central banks, the facility that sets the floor is a deposit facility.

BIS Papers No 67 59

rate target. For example, short-term rates rose above the policy target in some countries
after the Lehman Brothers bankruptcy in mid-September 2008. Outside such crisis periods,
however, short-term rates in some EMEs have remained below the policy rate target for
extended spells. This stands in contrast to the experience of industrial countries, where the
spread between the short-term rate and the policy rate is usually positive and relatively small,
suggesting a persistent shortage of liquidity in interbank markets; the spread, nonetheless,
became negative following exceptional monetary easing in 2009 and 2010.
Graph 2
Colombian short-term interest rates

1 In per cent. 2 Minimum expansion rate. 3 Overnight rate. 4 In basis points.
Sources: Datastream; national data.
The curvature of the yield curve
Along with macroeconomic and monetary policy factors, the demand for and supply of
long-term securities can influence the curvature of the yield curve. An excess demand for
long-term securities may thus reduce the term premia, leading to a flatter yield curve;
conversely, an excess supply may increase the term premia, steepening the yield curve. In
both cases, without corrective action by the central bank, monetary conditions would deviate
from those set by the policy rate.
An oft-cited expample was the strong demand for US treasury securities by the Asian central
banks leading to a “conundrum” of low US long-term interest rate prior to the recent financial
crisis (Bernanke (1995)). Monetary authorities may also have an explicit objective in
influencing the term premia, as demonstrated by the Federal Reserve’s and the Bank of
England’s quantatative easing programmes. Recent studies generally suggest that debt
maturity can have a significant effect on long-term interest rates. For instance, D’Amico et al
(2011) note that real-term premia for US Treasuries fell following the Fed’s large-scale
purchase of medium- to long-term securities. Other studies have given similar results.6
Given a shortage of high-quality EM assets and rather illiquid EM market conditions, the
impact of any given change in debt maturity is arguably much more significant in emerging
market economies than in industrial economies. And the growing investor base for emerging
market assets makes the role of debt maturity even more important in the determination of
the yield curve. For example, banks will have to hold government bonds of different

6 Krishnamurthy and Vissing-Jorgensen (2011) and Meaning and Zhu (2011). For the spillovers of quantitative
easing on yield curves in emerging Asia, see Chen et al (2012).

60 BIS Papers No 67

maturities to satisfy new liquidity regulations. Their demand for bonds may also be
conditioned by potential exposure to market risks. In terms of the changing demands at
different points along the yield curve, foreign investors may prefer to invest in short- to
medium-term maturity debt to avoid exposure to interest rate risks. Domestic pension funds,
in contrast, may demand longer-term bonds.7 Taken together, these trends in emerging
market economies suggest that the various supply and demand forces influencing the yield
curve are getting stronger and may become more volatile over time.
Graph 3
Spread between long-term1 and short-term rates2
Four-week moving averages, in percentage points
Emerging Asia Latin America Other emerging economies

1 Ten-year government bonds; for Argentina, one-year; for Brazil, three-year; for Chile and South Africa, nine-
year; for Turkey, two-year. 2 Three-month government bonds. For Colombia, one-year; for Argentina, the Czech
Republic and Poland, money market rates. 3 Simple average of Indonesia, Malaysia, the Philippines, Singapore
and Thailand.
Sources: Bloomberg; national data.
Graph 3 shows that the spread between the yields on three-month and 10-year government
securities for several EMEs has been quite volatile in recent years. In several countries, the
term spread is now very low or even negative. As term spreads measured in this way are
significantly affected by investors’ expectations about the future stance of monetary policy
and about macroeconomic prospects,8 the behaviour of spreads suggests expectations of
low short-term rates in the future. However, it is apparent that spreads fell during 2009–10
even as many countries were tightening monetary policy. This may partly reflect the stronger
demand for government paper associated with strong capital inflows during this period. The
background paper from Korea notes the dilemma posed to the central bank by large capital
inflows. Even though the Bank of Korea has raised its policy rate several times, long-term
rates have fallen. The central bank has referred to this as a “conundrum”, akin to the one
witnessed in the United States during the first half of the 2000s.
This evidence underscores the view that the effect of the maturity structure of official debt
held by the public has implications for the conduct of monetary policy in emerging market
economies. To the extent that central banks actively manage the maturity structure, they can

7 The paper from the Bank of Mexico shows that foreigners’ holdings of Mexican local currency debt have
tripled since 2009 and that their share of the market has more than doubled. Pension system reforms have
sharply increased the net assets of these funds, thereby increasing the demand for long-term bonds.
8 A more appropriate way to measure the term spread is to take the difference between the long-term rate and
an average of expected future short rates.

BIS Papers No 67 61

shape the yield curve to better reflect their policy intentions over the medium and long terms.
Indeed, authorities in EMEs have adopted different strategies for managing the maturity
structure of their bill and bond issuance. For example, as the background paper from Chile
notes, authorities seek to minimise the impact of their actions on bond yields, implying a
relatively neutral response to bond market demand and supply conditions. Another approach
seeks to actively reshape the yield curve from what would result from the actions of the
private financial markets alone. For instance, the background paper from the Reserve Bank
of India notes the challenges the RBI faced from a significant widening of fiscal deficits in
recent years. To prevent a sharp rise in the long-term interest rate, the RBI added more
market liquidity through its daily liquidity adjustment facility and shortened the average
maturity of government debt considerably, from 14.9 years in 2007–08 to 11.2 years in
2009–10.
Short-term debt and bank credit
Another potential implication of the debt structure relates to the growth of bank credit. In the
conventional monetary transmission mechanism, bank credit is determined primarily by
demand factors, so that the issuance of short-term debt (or bank reserves) should play little
role in determining the level of financial intermediation. In this case, when banks increase
their holdings of government bonds, they may crowd out credit to the private sector (Kuttner
and Lown (1998)).
Under imperfect market conditions, however, debt maturity can affect banks’ lending
behaviour. There are several channels. The first is that banks may face financing constraints.
Short-term government and central bank bills could then act as liquidity buffers (bank
reserves in waiting), relaxing these constraints and enhancing banks’ capacity to lend.9 The
size of the likely impact depends on the sensitivity of banks’ holdings of bills and securities to
the level of interest rates. When interest rates are low, banks have a greater incentive to
seek finance by liquidating their holdings of short-term paper rather than by borrowing from
the central bank.
A second is that liquid assets provide an easy way for investors to leverage up their balance
sheets. Banks and other investors may use their bond holding to build riskier exposures (see
Borio and Zhu (2008)). Liquidity and risk-taking may interact in a mutually reinforcing way,
increasing the strength of the monetary transmission mechanism and creating a destabilising
credit boom.
Finally, the desired asset maturity may matter. Bank managers may want a certain asset
maturity structure, and if the government does not supply enough long maturity assets banks
may create them by making loans, even though such lending may not be a perfect substitute;
in other words, the shorter debt maturity of sovereign bonds may lead banks to finance a
greater amount of long-term projects.
In emerging market economies, the evidence appears to provide some support for the view
that strong growth in short-term official sector debt is associated with credit growth. The
left-hand panel of Graph 4 plots the percentage change in bank credit to the private sector
and the ratio of short-term official liabilities presented in Table 3 (as a proxy for banks’
holdings of liquid assets). Both variables are averages for 2005–10. The right-hand panel
shows the relationship between changes in bank credit to the government and that to the
private sector.

9 A number of recent studies have highlighted the potential link between short-term debt securities and bank
credit; see Mohanty and Turner (2006), Filardo and Grenville (2011) and Mehrotra (2011).

62 BIS Papers No 67

Graph 4
Bank credit and short-term sovereign liabilities1

AR = Argentina; BR = Brazil; CN = China; CO = Columbia; CZ = Czech Republic; HU = Hungary; ID = Indonesia;
IN = India; KR = Korea; MX = Mexico; MY = Malaysia; PH = Philippines; PL = Poland; TH = Thailand;
TR = Turkey; ZA = South Africa.
1 Annual averages between 2005 and 2010, in per cent. 2 Short-term central bank and government debt plus
monetary base as a percentage of central bank and government debt plus monetary base (for details, see
Table 3).
Sources: IMF, World Economic Outlook and International Financial Statistics; national data; BIS debt securities
statistics; BIS questionnaire; BIS calculations.
The graph suggests that a positive relationship exists between short-term liabilities and bank
credit. By contrast, there no evidence of the “crowding-out” view. While a more systematic
analysis is needed to fully assess the role of short-term debt, the overall conclusion is that
government debt may stimulate lending while acting more as a complement to than as a
substitute for private sector lending.
4. Conclusion and comments on policy coordination
The maturity structure of government debt has potential implications for monetary policy,
especially in emerging market economies. With EM central banks becoming major issuers of
short-term debt, the yield curve is increasingly being shaped by their actions and by the
decisions of government debt managers. Increased demand from different investor classes
for EM sovereign bonds means that different segments of the yield curve are becoming more
sensitive to supply and demand. This has important implications for monetary conditions. In
addition, the maturity profile of public debt can have important effects on banks’ lending
behaviour.
This raises the possibility of occasional conflicts of interest between debt managers and
central banks. At the most basic level, good information-sharing between agencies is
essential if adverse market reactions to issuance, redemption and purchase schedules are to
be avoided. The prospects for potential conflicts might call for a review of arguments for an
independent debt management authority (see Appendix for models of information-sharing
and institutional arrangements).
The conflict of interest most relevant for monetary policy relates to differing mandates.
Consider the case of debt managers charged with keeping financing costs to a minimum. If
the central bank is trying to steepen the yield curve by lowering the policy rate in order to

BIS Papers No 67 63

achieve its macroeconomic objectives, debt managers would enjoy very favourable pricing
when issuing short-dated securities.10 But such issuance would conflict with the efforts of the
central bank. On the other hand, when central banks issue large amounts of shorter-term
bonds, they might crowd out governments from this market, forcing them to seek additional
long-term financing, possibly at higher cost.
One way to resolve such potential conflicts is to align the objectives of the two decision-
makers. In particular, debt managers may need a mandate that extends beyond standard
debt management considerations (eg cost mitigation) and includes the orderly functioning of
financial markets and overall macroeconomic stability.
In a similar vein, central banks may need to factor in additional objectives as long as they do
not compete with core monetary policy objectives. In particular, EM central banks have
traditionally played a role in promoting the development of government bond markets. But,
for some, this objective may at times interfere with their need to use central bank securities
to mop up liquidity, especially as a result of foreign reserve intervention. The use of central
bank securities can even lead to segmentation of rather shallow domestic sovereign bond
markets. On this point, Bank Indonesia has recently indicated its willingness to scrap its SBI
programme and to rely more heavily on government bonds in conducting monetary policy.
It is important that governance structures evolve in order to accurately address potential
conflicts and to mitigate the risks associated with any misalignment in the objectives of
central banks and debt managers. Enhanced consultations may be an effective option, along
the lines of those recently set up by Turkey. What we have learned in recent years about the
governance of financial stability responsibilities by setting up cross-agency committees may
now be applicable to debt management concerns. (Arrangements for selected economies
are summarised in the Appendix (pp 15–17)).

10 See CGFS (2011) and Hoogduin et al (2010).

64 BIS Papers No 67

Appendix:
Coordinating central bank and government debt management
Information coordination
Good information flows between central banks and debt managers can prevent many
avoidable day-to-day stresses in financial markets. For example, advanced notice of
upcoming debt sales and purchases, the breakdown of paper to be issued by maturity, the
types of holders being targeted and currency denomination all helps central banks to forecast
liquidity needs in the money markets. Likewise, information about upcoming monetary policy
operations is helpful to debt managers as they plan their auction schedules. In general,
information about each other’s timetable and strategies goes a long way to preventing
operational conflicts.
Various arrangements to promote the two-way flow of information have been adopted to
reduce the potential for such conflicts. Typically, they take the form of separating the
responsibility for managing the Treasury’s financing needs from the operational issues
associated with administration and settlement of government bond purchases and sales. In
many jurisdictions, the central bank has been given responsibility for settlement and
administration. For example, in the case of Colombia, the Ministry of Finance and Public
Credit regulates the overall level of government financing but leaves the timing of the
auctions, settlement and other operational details to the central bank. In Korea, where the
debt management unit has been consolidated inside the Ministry of Finance, the Bank of
Korea oversees the issuance, sale and purchase of bonds on the government’s behalf.
This information-sharing between debt managers and central banks will continue to be
particularly valuable in periods of heavy capital outflows, ie when markets are particularly
skittish and stresses easily develop; capital flows in the regions have been getting more
volatile. Conflicts may also arise when issuance schedules for both central banks and
governments are particularly busy; therefore, EMEs with large outstanding debts and high
deficits are more likely to need a higher level of information-sharing. Likewise, EME central
banks that need to roll over a substantial portion of their portfolio of assets or need to issue
central bank bills to drain liquidity may find it difficult to avoid creating volatility in financial
markets.
Centralised debt management units in central banks and explicit prohibitions
Some potential conflicts may be too costly for information-sharing and general governance
rules to be the sole means of aligning the objectives of debt managers and central banks.
The potential for future conflicts suggests that several questions will need to be revisited,
including whether to house centralised debt management units in central banks and how to
build stronger walls between the activities of debt managers and central banks.
Embedding debt management units inside the central bank, as at the Reserve Bank of India,
is a traditional way to ensure that the activities of the central bank and the debt managers are
well coordinated. That said, there is a current proposal to shift this function outside the
central bank in India. Arguments in favour of this proposal have been that debt management
responsibilities bias a central bank towards low interest rates in order to reduce sovereign
debt costs even if this compromises the central bank’s anti-inflation stance. A similar conflict
may also distort the central bank’s open market operations.11 In central banks with strong

11 For further details, see “Central bank governance issues – some RBI perspectives” by Governor Subbarao,
May 2011.

BIS Papers No 67 65

price stability credibility, such theoretical arguments appear overstated. In contrast,
arguments based on potential debt management conflicts appear to be a more important
consideration for deciding the level of central bank coordination.
Alternatively, potential conflicts may call for the building of stronger walls between the
activities of debt managers and central banks. This has been the approach to past episodes
of fiscal dominance, ie where finance ministries have pressured central banks to finance
government operations at relatively low interest rates. As a consequence, EME central banks
have been either discouraged or prohibited from making outright purchases of government
debt in primary markets. In Turkey, for example, the central bank was barred from
purchasing bonds in the primary market after the 2001 crisis and also from granting
advances and extending credit to the Treasury or to public establishments and institutions.
Such bans are common in many jurisdictions. To help prevent fiscal dominance, the
Philippines goes one step further: the Monetary Board of the Bangko Sentral ng Pilipinas
(BSP) must by law give its approval before the government can issue bonds in domestic or
foreign currency.
There are other ways of segregating the activities of central banks and debt managers. One
is for central banks and debt managers to agree that, during periods of heavy domestic
currency bond issuance, the central bank will rely on reserve requirements or FX repos to
drain liquidity instead of issuing central bank bills or selling the government bonds on its
balance sheet. Alternatively, debt managers and central banks might strategically target
different borrowers, ie debt managers might target residents while central banks target
non-residents. Or, as in some jurisdictions, central banks might confine themselves to
operating at the shorter end of the yield curve while debt managers operate at the longer
end. While all these options are technically possible, they may require formal agreements to
be effective if debt management units are not housed in central banks.

66
B
IS
P
apers N
o 67

Debt management arrangements and role of central banks, selected economies
Arrangements
CB settlement &
admin services of
government bonds

Sovereign bonds
CB
securities
ID Government debt securities (treasury bills and government bonds) are issued by the Ministry of Finance. Bank Indonesia (BI), as
an implementing agency, stipulates and administers the regulations regarding the issuance, sale and purchase of these debt
instruments.
BI is appointed by the government as an auction agent on the primary market to facilitate issuance of government securities. BI
may engage in purchase of these securities on the primary market only in respect of treasury bills. The purchase must be non-
competitive and placed directly but not through an authorised bidder. On the secondary market, BI may be appointed as an agent
for sale and purchase of these securities. In this case, BI shall conduct the sale and purchase of these securities on the secondary
market at the request of the Minister of Finance of the Republic of Indonesia.
Bank
Indonesia
certificates
(SBI)
Settlement services:
yes; via Bank Indonesia
– Scripless Securities
Settlement System
(BI-SSSS)

Administrative services:
yes
IN Government debt securities comprise dated securities issued by the Government of India and state governments as well as
treasury bills issued by the Government of India. The Reserve Bank of India (RBI) manages and services these securities through
its Public Debt Offices (PDOs) as an agent of the government.
The PDO of the RBI acts as the registry/depository of government debt securities and deals with the issuance, interest payments
and repayment of principal at maturity. These securities are managed by the Internal Debt Management Department of the RBI,
which also regulates and supervises the primary dealer system and has the responsibility of developing the government securities
market.
n/a Settlement services: no;
via the Clearing
Corporation of India
(CCIL)

Administrative services:
yes
PH The Department of Finance, through the Bureau of the Treasury, regulates issuance of government securities, which are treasury
bills and treasury bonds. The Bureau of the Treasury auctions the government securities on the primary market through its auction
system to eligible dealers or over the counter for specific investors.
The Secretary of Finance, with the approval of the President and after consultation with the Monetary Board of the Bangko Sentral
ng Pilipinas (BSP), is authorised to issue government securities in domestic or foreign currencies.1 Foreign/foreign currency-
denominated borrowings of the public sector require approval and registration with BSP to help control the size of the country’s
obligations, to keep the debt service burden at manageable levels, to channel loan proceeds to priority purposes/projects
supportive of the country’s development objectives, and to promote the best use of the country’s foreign exchange resources. BSP
also offers short-term special deposit accounts (SDAs).
n/a Settlement services:
yes; via the Philippine
Payment and
Settlement System
(PhilPASS)

Administrative services:
no
BR The National Treasury Secretariat is an agency of the National Treasury in charge of management and administration of the
domestic and external public debts. Most of the domestic government debt is issued through auctions held by the National
Treasury, making public offerings to financial institutions. The other types of issuance are: direct issuances used to meet specific
requirements defined by law, and public offerings to individuals, through the Tesouro Direto programme, which allows individuals
to purchase public bonds directly through the internet. Regarding issuance of government debt on the international market
(external debt), the proceeds may be used for paying both domestic and external debt for which the National Treasury is liable.
The Central Bank of Brazil uses treasury bonds to implement monetary policy, through the purchase and sale of securities on the
secondary market.
n/a2 Settlement services:
yes; via the Special
System for Settlement
and Custody3 (SELIC)

Administrative services:
no

B
IS
P
apers N
o 67
67

Debt management arrangements and role of central banks, selected economies (cont)
Arrangements
CB settlement &
admin services of
government bonds

Sovereign bonds
CB
securities
CO The Ministry of Finance and Public Credit (MFPC) regulates issuance of government debt securities. The Ministry determines the
target amount of total financing and periodically announces the level of compliance with the established targets, but it does not
provide the market with a specific calendar showing the dates of the auctions. Treasury bonds have been the main source of local
financing for the central government.
The central bank acts as an agent of the central government in the issuing and administration of domestic bonds. On behalf of the
MFPC, it announces the date, size and nominal value of the auction. The central bank has been authorised to carry out its
monetary policy with government bonds since 1990. The Constitution left open the possibility of the central bank acquiring
government bonds on secondary markets.�
n/a Settlement services:
yes; via the Electronic
Negotiation System
(ENS)

Administrative services:
yes
CL The International Finance Unit of the Finance Ministry is in charge of proposing and implementing strategies regarding public debt
through the Public Debt Office. The government issues peso- and Unidad de Fomento-denominated bonds in the domestic market.
The Central Bank of Chile carries out monthly bond auctions on dates published in a calendar in the amounts established by the
Finance Ministry. Both the central bank and the Finance Ministry periodically coordinate their planned debt issues. The central
bank is prohibited from purchasing government debt in the primary market. The central bank is the largest individual bond issuer in
Chile; these bonds are used in open market operations and to determine the benchmark yield curve of the economy.
Central bank
certificates
with
maturities
ranging from
30 days to 20
years
Settlement services: no;
via the Superintendencia
de Valores y Seguros
(SVS) of Chile

Administrative services:
yes
MX The Federal Government of Mexico, through the Ministry of Finance and Public Credit, is responsible for management and
issuance of government securities.
The Bank of Mexico (BM) operates as the financial agent for the Federal Government of Mexico and undertakes primary auctions
of government securities on a weekly basis. The Federal Government, through the Ministry of Finance and Public Credit, is
responsible for the management and issuance of government securities. The BM is prohibited from purchasing government
securities on the primary market.
n/a4 Settlement services: no;
via the Central Securities
Depository (INDEVAL)

Administrative services:
yes
HU The Hungarian government issues government bonds and discount treasury bills. In the late 1990s, the responsibility for debt
management was moved out of the central bank into the Ministry of Finance as a separate public debt office. The office, which was
first created as a partly independent organisation within the Treasury and then transformed into a corporation currently owned by
the Ministry for the National Economy (Government Debt Management Agency Pte Ltd (ÁKK)), is responsible for domestic and
foreign debt management, financing the central government and liquidity management of the state’s account balance. The majority
of government securities – discount treasury bills and government bonds – are sold through public issues. Primary dealers –
eligible investment companies or credit institutions – may buy government securities directly at the auctions, while investors must
order them from primary dealers. Government securities can be obtained on the secondary market, amongst others from primary
dealers or in the branch network of the Treasury.
Two-week
MNB bill
Settlement services: no;
via ÁKK (Gov’t Debt
Management Agency
Pte Ltd)

Administrative services:
yes

68
B
IS
P
apers N
o 67

Debt management arrangements and role of central banks, selected economies (cont)
Arrangements CB settlement &
admin services of
government bonds
Sovereign bonds
CB
securities
IL The Government Debt Management Unit comes under the Ministry of Finance. The government is authorised, with the
agreement of the Bank of Israel, to obtain services from banks or financial entities, provided this is done only in order to manage
the government’s debt and fiscal activity.
The Bank of Israel is responsible for the registration, clearing and payment of debts. The central bank administers the
government’s domestic loans, both non-compulsory (negotiable or non-negotiable) and compulsory. It is also responsible for the
redemption in Israel of foreign loans (State of Israel Bonds); and records receipts from the sales of bonds and payments to bond
and securities holders, maintains the register of holders of non-negotiable state loans and securities traded on the stock
exchange, and in this regard acts as the coordinating bank for short-term government securities.
Makam (up to
one year)
Settlement services: no;
via the Development
Corporation for Israel
(DCI)

Administrative services:
yes
PL The Republic of Poland, via the Ministry of Finance on behalf of the State Treasury, issues treasury bills of up to one year and
bonds of up to 10 years to cover the budget deficit. The Treasury is currently the largest issuer of bonds in Poland.
The National Bank of Poland (NBP) plays an important role in the securities clearing and settlement systems. The NBP is the
owner and operator of two securities settlement systems: the Register of Securities and the National Depository for Securities.
The NBP can purchase treasury bonds in the secondary market only exceptionally, in the case of a severe crisis threatening the
stability of the domestic financial system.
SEBOP
central bank
bills
Settlement services: yes;
via SKARBNET of the
Register of Securities;
and the National
Depository for Securities
Admin services: yes
TR The Undersecretariat of the Treasury, which is the issuer of government bonds and treasury bills, is responsible for the method
and terms of issuance as well as debt management. On behalf of the Undersecretariat, the Central Bank of the Republic of
Turkey (CBRT) issues bonds and bills in accordance with the financial services agreement with the Treasury. The CBRT is the
central securities depository, and all securities transfers are registered with the CBRT’s TIC-ESTS system.
Turkey has a coordination committee for debt management, but it does not include central bank representatives. Nonetheless,
the Treasury meets regularly with the central bank with respect to CBRT market liquidity management, as part of its coordinated
debt management strategy. The central bank, as the fiscal agent of the Treasury, also organises the auctions by collecting bids,
and sorting and submitting the lists to the Treasury on auction day. Auctions of domestic debt securities are open to all investors
(either institutional or individual). Since April 2001, the law has prohibited the central bank from granting advances and extending
credit to the Treasury and to public establishments and institutions, and from being a purchaser, in the primary market, of the
debt instruments issued by the Treasury and public establishments and institutions.
Liquidity bills5 Settlement services: yes;
via the Electronic
Securities Transfer and
Settlement System (TIC-
ESTS)

Administrative services:
yes
This table was compiled from publicly available sources.
1 The Philippine Constitution authorises the President to incur and guarantee foreign loans on behalf of the Republic of the Philippines with prior concurrence of the Monetary Board; all foreign
borrowing proposals of the government, government agencies and financial institutions have to be submitted for approval in principle by the Monetary Board before commencement of actual
negotiations, or before a mandate of commitment is issued to foreign funders/arrangers. 2 Since May 2000, the Central Bank of Brazil has no longer been authorised to issue its own
securities. 3 The Central Bank of Brazil manages SELIC and operates it jointly with ANBIMA. 4 Brems, “Bank of Mexico Monetary Regulation Bonds”, were issued from August 2000 to
July 2006. 5 Maturity shall not exceed 91days; first issued in July 2007.

BIS Papers No 67 69

Table A1
The consolidated public sector balance sheet
A B
General government Central bank
Assets Liabilities and net worth Assets
Liabilities and net
worth
1. Deposit with the
central banks
5. Gross debt
Net worth (4–5)
1. Foreign assets 5. Currency
2. Other financial
assets
2. Government bonds 6. Bank reserves
3. Capital stock 3. Claims on other
sectors
7. Government
deposits
4. Total assets 4. Total assets 8. Central bank bonds
9. Other liabilities
10. Capital
11. Total liabilities
Net worth (4–11)
C
Consolidated balance sheets
Assets Liabilities
1. Foreign assets (B1) 5. Currency (B5)
2. Financial assets (A2 + B3) 6. Net bank reserves (B6 – B7)
3. Capital stock (A3) 7. Government bonds (A5 + B8 – B2)
4. Total assets (C1 + C2 + C3) 8. Other liabilities (B9)
9. Total liabilities (C5 + C6 + C7 + C8)
Consolidated net worth (C4 + B10 – C9)

70 BIS Papers No 67

Graph A1
Short-run interbank rate: spread to reference rate or level1
22-day moving average
Asia

Latin America

Other emerging markets Advanced economies

On right-hand side scale unless otherwise indicated; in basis points unless otherwise indicated.
1 For Argentina, seven-day interbank (BAIBOR) rate and BCRA seven-day reverse repo agreement rate
(http://www.bcra.gov.ar/pdfs/estadistica/tasser.xls); for Brazil, financing overnight SELIC rate and SELIC target
rate; for Chile, Central Bank of Chile daily midday nominal interbank rate and official monetary policy rate; for
China, one-week SHIBOR (level; in per cent); for Colombia, overnight interbank rate and minimum expansion
rate; for the Czech Republic, two-week interbank rate and two-week repo rate target; for the euro area, EONIA
rate and ECB main refinancing rate; for Hungary, two-week interbank rate and base rate; for India, Mumbai
overnight interbank rate and repo rate; for Indonesia, overnight interbank rate and BI rate; for Korea, brokered
overnight call rate and base rate; for Malaysia, Kuala Lumpur overnight interbank rate and overnight policy rate;
for Mexico, overnight bank funding rate and overnight interbank rate target; for Peru, one-month interbank
(LIMABOR) rate and reference interest rate; for the Philippines, interbank call loan rate and reverse-repo rate; for
Poland, one-week interbank rate and reference rate; for Russia, overnight interbank (MIBOR) rate and refinancing
rate; for South Africa, SABOR rate and official repo rate target; for Turkey, one-week interbank rate target
(overnight interbank rate prior to May 2010) and one-week repo lending rate (overnight borrowing rate prior to
May 2010); for Thailand, overnight interbank rate (Bangkok one-week interbank rate prior 17 January 2007) and
overnight repo rate target (14-day repo rate target prior to 17 January 2007); for the United States, federal funds
effective rate and federal funds rate target.
Sources: Bloomberg; Datastream; national data.

BIS Papers No 67 71

References
Borio, C and H Zhu (2008): “Capital regulation, risk-taking and monetary policy”, BIS Working
Papers, no 268, December.
Chen, Q, A Filardo, D He and F Zhu (2012): “The impact of central bank balance sheet
policies on the emerging economies”, BIS Papers, no 66.
Committee on the Global Financial System (2011): “Interactions of sovereign debt
management with monetary conditions and financial stability: lessons and implications for
central banks”, CGFS Papers, no 42, May.
D’Amico, S, W English, D López-Salido and E Nelson (2011): “The Federal Reserve’s large-
scale asset purchase programs: rationale and effects”, mimeo.
Filardo, A and S Grenville (2012): “Central bank balance sheets and foreign exchange rate
regimes: understanding the nexus in Asia”, BIS Papers, no 66.
Friedman, B (1978): “Crowding out or crowding in? Economic consequences of financing
government deficits”, Brookings Papers on Economic Activity, vol 1978, no 3.
Hoogduin, L, B Öztürk and P Wierts (2010): “Public debt managers’ behaviour: interactions
with macro policies”, DNB Working Papers, no 273, December.
Krishnamurthy, A and A Vissing-Jorgensen (2011): “The effects of quantitative easing on
interest rates”, Brookings Papers on Economic Activity, September.
Kuttner, K and C Lown (1998): “Government debt, the composition of bank portfolios, and
transmission of monetary policy”, in A Chrystal (ed), Government debt structure and
monetary conditions, Bank of England.
Meaning, J and F Zhu (2011): “The impact of recent central bank asset purchase
programmes”, BIS Quarterly Review, December.
Mehrotra, A (2011): “On the use of sterilisation bonds in emerging Asia”, BIS Working
Papers, no 66.
Modigliani, F and R Sutch (1966): “Innovations in interest rate policy”, American Economic
Review, vol 56, no 1.
Mohanty, M and P Turner (2006): “Foreign exchange reserve accumulation in emerging
markets: what are the domestic implications?”, BIS Quarterly Review, September.
Moreno, R (2011): “Foreign exchange market intervention in EMEs: implications for central
banks”, BIS Papers, no 57.
Tobin, J (1963): “An essay on principles of debt management”, in Essays in Economics –
Volume 1: Macroeconomics, North Holland.
Turner, P (2011): “Fiscal dominance and the long-term interest rate”, Financial Markets
Group Special Paper Series, no 199, May.
Vargas, H (2011): “Monetary policy and the exchange rate in Colombia”, BIS Papers, no 57.

BIS Papers No 67 73

Fiscal policy, public debt management and
government bond markets: issues for central banks
Miguel Angel Pesce1
Abstract
The global financial crisis showed that both authorities and markets failed to properly assess
the size and the evolution of the public debt stock in various economies. In some countries
the monetary authorities focused excessively on inflation, without taking into account other
key macroeconomic variables and ratios. That said, it is important to ask why some
macroeconomic variables were able to follow such unsustainable paths for lengthy periods.
Part of the explanation is the scenario of strong growth, with high international liquidity and
low inflation, that prevailed before the crisis. In addition, EU countries, especially the less-
developed ones, were able to reduce their financing costs after the introduction of the euro.
In this paper, we also examine the role played by economic authorities, and the inter-
relationships among them in the design and implementation of fiscal policy and debt
management in response to the crisis. Rigid central bank goals and inflexible boundaries
between the central bank and the treasury were erased, allowing the economic authorities to
behave in a pragmatic way. Finally, we discuss the role played by credit rating agencies and
regulatory frameworks.

Keywords: Monetary policy, public debt
JEL classification: E42, E52, E63

1 Deputy Governor, Central Bank of Argentina.

74 BIS Papers No 67

1. Introduction
The current global financial crisis has made it clear that economic authorities and markets in
various countries have not properly considered the effects that the size and evolution of the
stock of public debt and the government primary surplus can trigger. And some authorities
have focused their assessment view of the economy exclusively on inflation.
In this sense, it is important to analyze the reasons that could explain why some
macroeconomic variables were able to follow unsustainable medium- and long-term paths for
a long period. At the same time, it is also important to see the role played by economic
authorities, central banks among them, and their institutional arrangements for the design
and implementation of fiscal and monetary policy and debt management.
2. Some macroeconomic thoughts
Following the crisis of the late nineties that mainly affected Southeast Asian countries,
Russia and other emerging economies (EMEs), economic authorities began to implement
policies basically aimed at reducing their dependence on foreign capital and promoting fiscal
robustness. At the same time, EU countries, especially the ones with a relatively lower
degree of development, were benefited financially by the introduction of the euro as a
common currency in early 1999. This allowed those countries to effectively reduce their costs
of indebtedness.
In both cases, those changes took place within a framework characterized by ample
international financial liquidity, which in domestic financial markets was reflected in a high
growth rate of monetary aggregates and increases in asset values. Usually the
consequences of the latter were underestimated due to the importance given, in
implementing monetary policies, to domestic inflation indicators that do not incorporate asset
prices.2
On the other hand, it is important to stress that in the last 30 years the world economy has
experienced structural changes. Some of these have had deep consequences for the role
played by different variables over the inflationary process, such as the robust growth of
international trade, with the growing importance of low-cost manufactured products provided
by EMEs and the continuous decrease in tariff trade barriers that on average went down from
26% in 1986 to 8.8% in 2007. As a result, the world experienced a period of strong real
growth combined with low levels of inflation.3
In this regard, many economies had controlled domestic inflation while other macroeconomic
variables, such as debt ratios as a percentage of GDP, current account or fiscal deficits (or
both), showed disruptive trajectories. This evolution has not affected all economies to the
same extent, as the restrictions they faced were not similar. In fact, countries or areas with
currencies that are internationally accepted – used in trade or international reserves – enjoy
higher degrees of freedom than economies that do not possess such currencies.
As shown in Table No. 1, comparing the average of 2008–11 with that of 1998–2007 it is
clear that, while industrialized countries and EU members showed a rise in their levels of

2 In relation to this, Axel Leijonhufvud maintains that inflation targeting implies “a central banking doctrine that
requires an exclusive concentration on keeping consumer prices within a narrow range with no attention to
asset prices, exchange rates, credit quality or (of course) unemployment” (“Keynes and the crisis”, CEPR
Policy Insight, no. 23, May 2008). We would add to this list a lack of attention to fiscal deficits and public debt
levels.
3 M Pesce, “Monetary policy and measures of inflation”, BIS Papers, no. 49, December 2009.

BIS Papers No 67 75

public debt relative to GDP (some of them a very sharp one), the EMEs had, in general,
reduced this ratio. It can also be see in Table No. 2 that, for the first group of countries, with
the exception of Japan, while the sum of the primary fiscal balance for 1998–2007 was
positive, it became negative during 2008–11. In the case of the EMEs the evolution was in
the same direction, although in the second period primary fiscal balances still were positive
or showed small negative values. Combined with the figures on public debt interest
presented in Table No.3, we can say that the fiscal stance in EU and other developed
countries worsened, and was not much better in EMEs.
Several factors explain the evolution of macroeconomic variables, which obviously differs for
each economy. It is clear that different public deficits and debt levels have different effects on
medium- and long-term economic performance: debt and fiscal balance levels and their
paths have impacts on the stability of the economic and financial systems.
From the data presented, it is possible to say that some economies showed primary fiscal
results that were not enough to cope with rising debt levels and interest burdens. In this
sense, some economies shifted from “speculative” to “Ponzi” situations as their fiscal primary
balances were insufficient to cover the interest payments.
In this context, and as has been widely reported in recent months, credit rating agencies
have played an important role in the evolution of the crisis. In Table 4 we can see the
evolution of sovereign debt ratings assigned to a group of countries in the period December
2000–December 2011. For some countries, strong swings in sovereign ratings can be
observed during very short periods. Though they had different macroeconomic frameworks,
the positive assessments and ratings given to EU countries early in the process of
introducing the common currency can also be noted. However, when comparing these
countries’ fundamentals with those of some EMEs, it appears that the EMEs showed better
and more sustainable indicators. Yet those healthier fundamentals were not reflected in
EMEs’ credit ratings. Favorable sovereign bond ratings allowed banks in a number of
financial systems to increase their assets without imposing pressures for increased capital
integration or loan loss provisions. At the end of the day, it allowed pro-cyclical lending
behavior to develop.

76
B
IS
P
apers N
o 67
Table 1
Public Debt
percentage of GDP
Source Country 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011
2011 –
1998
FMI
United States 64.6 60.8 54.8 54.7 57.1 60.4 61.5 61.7 61.1 62.3 71.6 85.2 94.4 100.0

59.9
Euro Zone 72.9 71.9 69.3 68.3 68.1 69.3 69.7 70.3 68.6 66.4 70.1 79.7 85.8 88.6

69.5
Euro Zone
(Excluding
Germany)
78.1 76.7 73.2 72.0 71.0 71.1 70.9 70.8 68.7 66.8 71.3 81.7 86.4 90.7

71.9
Germany 60.5 61.3 60.2 59.1 60.7 64.4 66.2 68.5 67.9 65.0 66.4 74.1 84.0 82.6

63.4
Italy 114.9 113.7 109.2 108.8 105.7 104.4 103.9 105.9 106.6 103.6 106.3 116.1 119.0 121.1

107.7
France 59.5 58.9 57.4 56.9 59.0 63.2 65.0 66.7 64.0 64.2 68.2 79.0 82.3 86.8

61.5
Spain 64.1 62.4 59.3 55.5 52.6 48.7 46.2 43.0 39.6 36.1 39.8 53.3 60.1 67.4

50.7
Portugal 50.4 49.6 48.5 51.2 53.8 55.9 57.6 62.8 63.9 68.3 71.6 83.0 92.9 106.0

56.2
Greece 96.6 102.5 103.4 103.7 101.5 97.3 98.8 100.3 106.1 105.4 110.7 127.1 142.8 165.6

101.6
Ireland 53.0 48.0 37.5 35.2 31.9 30.7 29.1 27.1 24.7 24.9 44.4 65.2 94.9 109.3

34.2
United Kingdom 46.3 43.7 40.9 37.7 37.2 38.5 40.2 42.1 43.1 43.9 52.0 68.3 75.5 80.8

41.4
Japan 120.1 133.8 142.1 151.7 160.9 167.2 178.1 191.6 191.3 187.7 195.0 216.3 220.0 233.1

162.4
Brazil 43.8 55.5 66.7 70.2 79.8 74.7 70.6 69.1 66.7 65.2 63.6 68.1 66.8 65.0

66.2
Chile 12.5 13.8 13.7 15.0 15.7 13.0 10.7 7.3 5.3 4.1 5.2 6.2 9.2 10.5

11.1
Colombia 27.5 34.1 36.3 40.9 43.9 45.6 42.9 38.5 36.8 32.7 30.8 35.8 36.0 35.9

37.9
Mexico 45.4 47.4 42.6 42.0 45.7 45.6 41.4 39.8 38.4 37.8 43.1 44.7 42.9 42.9

42.6
Perú 34.4 47 42.4 41.5 43.2 47.1 44.3 37.7 33.2 30.9 25.0 27.1 24.5 21.5

40.2
Russia 69.9 99 59.9 47.6 40.3 30.4 22.3 14.2 9.0 8.5 7.9 11.0 11.7 11.7

40.1
Turkey 44.8 51 51.3 77.6 73.7 67.4 59.2 52.3 46.1 39.4 39.5 46.1 42.2 40.3

56.3
China 11.4 13.8 16.4 17.7 18.9 19.2 18.5 17.6 16.2 19.6 17.0 17.7 33.8 26.9

16.9
India 65.8 68.0 71.8 76.2 80.6 81.7 81.0 78.7 75.4 72.7 73.1 69.4 64.1 62.4

75.2
Source: IMF, IFS

B
IS
P
apers N
o 67
77

Table 2
Public Debt
percentage of GDP
Source Country 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011
Sum
Period
Annual
Average
FMI
United States 3.4 3.4 3.9 2.0 -1.8 -2.9 -2.5 -1.2 -0.1 -0.7 -4.5 -10.9 -8.4 -8.0

3.6 -31.7
Euro Zone 1.9 2.3 3.5 1.5 0.6 -0.1 -0.1 0.2 1.2 1.9 0.6 -3.8 -3.6 -1.5

12.8 -8.3
Euro Zone
(Excluding
Germany)
2.4 2.7 3.2 2.2 1.2 0.5 0.4 0.6 1.4 1.6 -0.1 -4.9 -4.4 -2.2

16.3 -11.6
Germany 0.8 1.3 4.0 -0.3 -1.2 -1.5 -1.3 -1.0 0.8 2.7 2.5 -0.8 -1.2 0.4

4.3 0.9
Italy 4.7 4.6 5.2 2.9 2.4 1.4 1.1 0.1 1.1 3.3 2.2 -1.0 -0.3 0.5

26.9 1.5
France 0.4 0.9 1.1 1.1 -0.6 -1.5 -1.0 -0.5 0.0 -0.3 -0.6 -5.3 -4.9 -3.4

-0.3 -14.3
Spain 0.7 1.9 2.0 2.1 2.0 1.9 1.5 2.5 3.3 3.0 -3.1 -9.9 -7.8 -4.4

20.9 -25.2
Portugal 0.9 1.7 1.3 0.0 1.5 2.3 2.1 -0.3 2.2 -0.4 -0.7 -7.4 -6.3 -1.9

11.4 -16.2
Greece 4.6 4.5 3.6 2.1 0.8 -0.7 -2.5 -0.7 -1.5 -2.0 -4.8 -10.3 -4.9 -1.3

8.2 -21.4
Ireland 5.6 4.9 6.6 2.3 0.8 1.5 2.4 2.7 3.9 0.8 -6.5 -12.4 -28.9 -6.8

31.4 -54.5
United Kingdom 2.3 2.9 3.2 2.1 -0.6 -1.9 -1.9 -1.8 -1.1 -1.1 -3.3 -8.5 -7.7 -5.6

2.2 -25.1
Japan -4.3 -6.1 -6.3 -5.0 -6.8 -6.8 -5.1 -4.1 -3.5 -1.9 -3.4 -9.4 -8.1 -8.9

-49.9 -29.7
JP Morgan
Brazil 0.0 2.9 3.2 3.4 3.2 3.3 3.8 3.9 3.2 3.4 3.5 2.1 3.3 3.3

30.3 12.2
Chile -0.5 -0.6 0.2 0.6 -0.8 0.2 3.8 6.1 8.8 9.4 6.1 -3.8 -0.9 -0.2

27.2 1.2
Colombia 0.8 -0.3 0.3 0.6 0.8 1.8 3.3 2.9 3.5 3.4 3.6 1.0 -0.4 0.2

17.1 4.4
Mexico 1.6 2.3 2.4 2.3 1.6 1.9 2.2 2.2 2.5 2.2 1.8 -0.1 1.3 1.4

21.2 4.4
Peru 1.2 -0.9 -0.8 -0.2 -0.1 0.4 1.0 1.6 4.0 4.9 3.7 -0.6 0.2 0.7

11.1 4.0
Russia -0.6 2.3 4.9 5.6 3.5 3.4 5.5 8.4 8.0 5.8 4.5 -5.4 -3.4 -2.2

46.8 -6.5
Turkey 4.5 2.0 4.4 5.2 3.3 4.0 4.9 6.0 5.4 4.1 3.5 0.1 0.7 1.0

43.8 5.3
China -1.1 -1.9 -2.5 -2.3 -2.6 -2.2 -1.3 -1.2 -0.8 0.6 -0.4 -2.3 -2.7 -1.8

-15.3 -7.2
India -3.5 -0.1 0.4 -1.5 -1.5 -2.9 -1.3 -0.6 0.5 0.5 -2.4 -4.7 -2.8 -1.9

-10.0 -11.8
Source: IFS, IMF and EM Debt and fiscal indicators JPM.

78
B
IS
P
apers N
o 67
Table 3
Interest on Public Debt
percentage of GDP
Source Country 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011
OECD
United States 3.1 2.7 2.5 2.2 2.0 1.8 1.8 1.8 1.8 1.9 1.9 1.6 1.7 2.0
Euro Zone 4.2 3.7 3.5 3.3 3.1 2.9 2.8 2.7 2.6 2.6 2.6 2.5 2.4 2.6
Euro Zone
(Excluding
Germany) 4.7 4.0 3.8 3.6 3.4 3.1 2.9 2.8 2.6 2.7 2.7 2.5 2.6 2.8
Germany 3.0 2.8 2.7 2.6 2.5 2.6 2.5 2.5 2.5 2.5 2.4 2.3 2.1 2.0
Italy 7.8 6.3 6.1 6.0 5.4 4.9 4.6 4.5 4.4 4.7 4.9 4.3 4.2 4.5
France 3.0 2.7 2.6 2.7 2.7 2.6 2.6 2.5 2.4 2.5 2.7 2.2 2.3 2.4
Spain 3.8 3.3 2.9 2.6 2.4 2.1 1.8 1.6 1.3 1.1 1.1 1.4 1.5 1.5
Portugal 3.1 2.9 2.9 2.9 2.8 2.7 2.6 2.4 2.7 3.0 3.1 2.9 3.0 4.2
Greece 7.3 6.4 6.6 5.9 5.2 4.7 4.6 4.4 4.5 4.5 4.8 5.0 5.7 6.9
Ireland 3.2 2.2 1.8 1.2 1.1 1.1 1.0 0.9 0.7 0.6 0.8 1.4 2.7 3.2
United Kingdom 3.0 2.5 2.4 2.0 1.7 1.7 1.7 1.8 1.7 1.8 1.8 1.6 2.6 2.6
Japan 1.5 1.5 1.5 1.4 1.4 1.3 1.2 0.8 0.6 0.6 0.9 1.1 1.4 1.6
JP Morgan
Brazil 7.9 8.2 6.6 6.7 7.6 8.4 6.6 7.3 6.7 6.1 5.4 5.4 5.5 6.0
Chile -0.9 1.5 0.9 1.1 0.4 0.6 1.7 1.4 1.1 0.6 0.8 0.8 0.6 0.6
Colombia 4.1 3.3 4.0 4.4 3.9 4.1 3.8 3.3 3.9 3.8 3.1 3.2 3.2 3.2
Mexico 2.7 3.3 3.4 3.0 2.7 2.5 2.4 2.3 2.4 2.2 1.9 2.2 3.9 3.9
Peru 2.2 2.3 2.5 2.3 2.2 2.1 2.0 1.9 1.9 1.8 1.6 1.4 1.2 1.2
Russia 5.7 3.4 3.5 2.6 2.1 1.7 1.2 0.9 0.6 0.4 0.4 0.5 0.6 0.8
Turkey 11.8 13.8 12.3 17.1 17.5 12.8 10.1 7.1 6.0 5.7 5.2 5.6 4.3 3.6
China

India 5.8 8.5 8.7 9.2 8.5 6.5 5.9 6.3 6.8 5.2 7.3 5.4 5.2 5.3
Source: IFS, IMF and EM Debt and fiscal indicators JPM. Own calculation based on Primary surplus and Fiscal Surplus.

B
IS
P
apers N
o 67
79

Table 4
Long term Sovereigns Ratings S&P
Country Dec.00 Dec.01 Dec.02 Dec.03 Dec.04 Dec.05 Dec.06 Dec.07 Dec.08 Dec.09 Dec.10 Dec.11
United States AAA AAA AAA AAA AAA AAA AAA AAA AAA AAA AAA AA+
Germany AAA AAA AAA AAA AAA AAA AAA AAA AAA AAA AAA AAA
France AAA AAA AAA AAA AAA AAA AAA AAA AAA AAA AAA AAA
Spain AA+ AA+ AA+ AA+ AAA AAA AAA AAA AAA AA+ AA AA-
Italy AA- AA- AA- AA- AA- A+ A+ A+ A+ A+ A+ A
Greece A A A+ A+ A A A A A BBB+ BB+ CC
Ireland AAA AAA AAA AAA AAA AAA AAA AAA AAA AA A BBB+
Portugal AA- AA- AA- AA- AA- AA- AA- AA- AA- A- A- BBB-
India BB+ BB+ BB+ BB+ BB+ BB+ BB+ BBB- BBB- BBB- BBB- BBB-
Turkey B+ B- B- B+ B+ BB- BB- BB- BB- BB- BB BB
Russia B- B B+ BB+ BB+ BBB BBB+ BBB+ BBB BBB BBB BBB
China BBB BBB BBB BBB BBB+ A- A A A+ A+ AA- AA-
Brazil B+ BB- B+ B+ BB- BB- BB BB+ BBB- BBB- BBB- BBB
Mexico BB+ BB+ BBB- BBB- BBB- BBB BBB BBB+ BBB+ BBB BBB BBB
Source: S&P

80
B
IS
P
apers N
o 67
Table 5
Government Securities – BCRA’s Own Portfolio – Total as of

Balances in Argentine
Pesos Balances in US Dollars Exchange Rate USD/$
(in thousands)
12/31/2007 4,798,811 1,522,949 3.1510
12/31/2008 6,216,129 1,799,846 3.4537
12/31/2009 14,242,300 3,751,231 3.7967
12/31/2010 20,167,413 5,072,542 3.9758
12/31/2011 (1) 16,519,320 3,838,846 4.3032
Source: BCRA
(1) subject to adjustments

Securities issued by the BCRA – LEBAC and NOBAC – Total as of

Outstanding
Stock as of Total Domestic Holders Foreign Holders
(in Nominal Value)
12/31/2007 51,497 44,131 7,366
12/31/2008 36,698 35,799 899
12/31/2009 46,828 46,823 5
12/31/2010 74,352 74,307 45
12/31/2011 68,725 68,725 0
Source: BCRA

B
IS
P
apers N
o 67
81

Table 6
Financial Systems Total Assets
In millions of USD
1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010
United States – – – 9,521 10,004 10,598 11,442 12,591 13,986 15,649 17,335 16,880 16,817
Euro Zone 15,457 14,246 14,266 14,719 18,219 23,098 26,872 25,338 31,043 39,105 39,174 40,312 37,090
United Kingdom
(*) 3,625 3,645 3,988 4,212 4,933 5,991 7,343 7,608 9,660 12,179 10,194 10,537 10,518
Japan – – – 9,608 10,331 11,535 11,847 10,663 10,207 10,606 13,734 13,429 15,574
Brazil – – – 462 341 476 579 788 1,037 1,578 1,394 2,094 2,593
Chile – – – 58 56 68 84 104 116 152 143 182 201
Colombia – – – 37 29 33 46 56 67 89 92 107 126
Mexico – – – 330 326 312 321 365 414 467 405 444 518
Peru 21 20 20 21 20 20 22 26 28 41 50 53 66
Russia – – 86 103 122 174 241 297 476 736 885 895 1,014
Turkey – – – 92 118 162 208 278 323 451 435 514 591
China 1,396 1,552 1,738 1,947 2,399 2,858 3,261 3,692 4,426 5,724 7,093 8,945 11,181
India 151 180 201 222 274 333 424 478 587 801 790 952 1,194
Source: own elaboration based on IFS/IMF.

82
B
IS
P
apers N
o 67
Table 7
Financial Sector Exposure to public sector (*)
In millions of USD
1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010
United States – – – 643 696 639 604 623 664 872 1,283 1,138 1,147
Euro Zone 2,274 1,952 1,695 1,681 2,051 2,608 2,893 2,660 2,753 3,171 3,081 3,601 3,650
United Kingdom
(*) 24 18 -8 1 20 11 28 25 -6 -19 -20 42 129
Japan – – – 1,581 1,865 2,471 2,677 2,439 2,469 2,527 3,459 3,854 4,702
Brazil – – – 196 135 221 275 379 462 594 458 704 812
Chile – – – 1 1 1 1 1 1 2 2 3 4
Colombia – – – 10 5 9 13 16 13 13 13 20 23
Mexico – – – 101 92 89 95 101 121 142 127 161 174
Peru 1 2 1 2 2 2 2 2 2 2 2 3 2
Russia – – 22 22 26 30 34 30 38 47 40 55 74
Turkey – – – 44 57 81 94 118 115 149 133 182 188
China 60 73 89 133 164 184 223 246 291 397 442 557 657
India 53 51 48 46 47 49 51 50 51 57 46 48 50
(*) Public sector net assets of the own jurisdiction
Source: own elaboration based on IFS/IMF.

B
IS
P
apers N
o 67
83

Table 8
Financial Sector Exposure to public sector (*)
In % of total assets
1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010
United States – – – 6.8 7.0 6.0 5.3 4.9 4.7 5.6 7.4 6.7 6.8
Euro Zone 14.7 13.7 11.9 11.4 11.3 11.3 10.8 10.5 8.9 8.1 7.9 8.9 9.8
United Kingdom
(*) 0.7 0.5 -0.2 0.0 0.4 0.2 0.4 0.3 -0.1 -0.2 -0.2 0.4 1.2
Japan – – – 16.5 18.0 21.4 22.6 22.9 24.2 23.8 25.2 28.7 30.2
Brazil – – – 42.3 39.5 46.3 47.6 48.0 44.6 37.6 32.9 33.6 31.3
Chile – – – 1.5 1.4 1.2 1.2 1.3 0.9 1.2 1.4 1.9 1.9
Colombia – – – 27.9 18.1 27.0 27.9 28.8 18.9 15.0 14.0 18.9 18.5
Mexico – – – 30.6 28.0 28.6 29.5 27.6 29.1 30.4 31.4 36.3 33.6
Peru 5.9 7.4 7.4 9.7 10.0 10.0 8.9 7.5 5.5 4.1 3.1 4.9 3.8
Russia – – 25.4 21.7 21.2 17.4 13.9 10.1 8.0 6.4 4.6 6.2 7.3
Turkey – – – 48.0 48.1 49.6 45.1 42.5 35.5 33.1 30.6 35.4 31.8
China 4.3 4.7 5.1 6.9 6.8 6.4 6.9 6.7 6.6 6.9 6.2 6.2 5.9
India 34.8 28.6 23.8 20.9 17.0 14.7 12.1 10.4 8.6 7.1 5.8 5.0 4.2
(*) Public sector net assets of the own jurisdiction
Source: own elaboration based on IFS/IMF.

84 BIS Papers No 67

It thus seems clear that it might be misleading to pay attention only to sovereign debt ratings
when assessing sovereign credit risks. That variable gives insufficient information on its own
and should be complemented with more attention to ratios such as fiscal primary
balance/GDP, financial balance/GDP and debt service/GDP and their evolution.
With the benefit of hindsight, we know that in the period before the first phase of the
international financial crisis in 2007, many central banks implemented policies that were not
able to prevent the effects of a reversal of the cycle, and, in some cases, these policies
amplified the inconsistent trajectories of some key variables and ratios. When the crisis broke
out, they deployed a set of policies to try to support liquidity conditions in the financial
systems and stabilize the market value of sovereign debt. In addition, some central banks
modified their regulatory framework regarding the valuation of assets, including sovereign
debt. All those goals were achieved through the implementation of unconventional monetary
policy measures that included special programs granting credit lines, swaps and the
extension of collateral, together with very active participation in secondary markets. In this
sense, we can say that central banks played two roles simultaneously. On the one hand,
they maintained, and in some cases recovered, their role of lender of last resort in financial
systems. On the other hand, when most investors carried out strong sales of government
bonds, they became lenders of last resort for some countries by buying public debt in
secondary markets. In addition, some central banks transferred profits to treasuries which in
many cases were originated in its purchases of sovereign securities in secondary markets.4
In the case of Argentina, starting in 2008 the central bank increased its open market
operations by purchasing different government securities in the secondary market (see
Table No. 5). This mechanism worked as an additional tool to provide liquidity beyond the
banking sector, and at the same time enabled intervention in different segments of the yield
curve, preventing market distortions. The central bank also authorized a new liquidity window
that enables financial institutions to obtain funds with different sovereign bonds and assets. It
has also decided that these securities can be used as collateral for inter-bank loans at a
minimum seven-day term. This allowed a number of institutions that did not have central
bank bills and notes – Lebac and Nobac – in their portfolios to access our liquidity provision
mechanisms, by using treasury bonds as collaterals.
3. Aspects of the regulatory framework in relation to sovereign debt
3.1 Some features of the financial system regulatory framework may also explain the
evolution of the crisis mentioned above. Prior to 2007, in a context of abundant
international liquidity, the regulatory framework included favorable incentives, in
terms of capital integration, for the voluntarily maintenance of sovereign debt as part
of the assets of financial institutions. Table No. 6 shows the strong growth of total
assets in financial systems. That rise clearly understates the increase in sovereign
bond holdings shown in Tables No. 7 and No. 8. These tables show the significant
growth in the period 2002–10 of financial systems’ exposure to the public sector.
As I said before, it is important to remember that in the context of Basel II there were
incentives for holding government securities in bank assets.

4 For instance, “The Federal Reserve in recent years has transferred net income to the US Treasury, by
preliminary unaudited results the increase was primarily due to increased earnings on securities holdings
during 2009” (Federal Reserve, press release, January 12, 2010).

BIS Papers No 67 85

In the case of the capital requirement for credit risk under the standardized method,5
even when it incorporates different risk weights for loans to sovereign states and their
central banks (Paragraph 53), in Paragraph 54 introduces a degree of discretion for
national regulators to apply lower weights to those assets – even zero for cases in
which they are denominated and financed in the domestic currency. Some countries
implemented such regulations in this way.6
As I mentioned, under the Basel II standardized approach, the calculation of credit in
risk requirements mechanically rests on the ratings issued by credit rating agencies or
export credit agencies. In the case of sovereign debt in particular, recent international
experience shows that rating agencies have not been effective in pointing out, well in
advance, credit risks that arise from economic or fiscal weaknesses (their failure in the
subprime crisis has also been evident). This fact, coupled with favorable treatment –
i.e. lower weighting of sovereign risk exposures given to domestic currency expressed
in euros – resulted in low capital requirements to cover these exposures, while at the
same time, it may have also acted as an incentive to increase sovereign debt
exposures.
In the case of Argentina, during the nineties there was favorable regulatory banking
treatment for public sector asset holdings. Until March 2000, both domestic holdings of
government securities and public sector loans did not face, in practice, minimum capital
requirements for credit risk. After March 2000, public sector asset holdings have been
subject to minimum capital requirements in terms of their modified duration, although
lower than those of the private sector assets.
Since 2003, capital requirements for credit risk – exposure to the public sector – have
been similar to those for non-financial exposure to the private sector, 8% of the capital
compliance. In addition, the central bank established two limits for assistance to the
non-financial public sector:
1. Regarding their total assets, a maximum of 40% (reduced to 35% in 2007)
2. Regarding their regulatory capital, 50% (for national bodies), 10% (for provinces)
and 3% (for municipalities), with the three levels in combination amounting to no more
than 75% of the regulatory capital.
3.2 Another macroprudential instrument implemented to deal with the dollarization and
currency mismatch that may affect financial stability was the decision to issue Lebacs
and Nobacs (Com. A 4715, September 2007) which can only be traded locally (known
as domestic Lebacs and Nobacs). This measure sought to prevent short-term foreign
investors from acquiring these securities, which tended to distort their value in the
secondary market, affecting their function as a source of liquidity.
The issue of domestic Lebacs and Nobacs had more than one aim: the one mentioned
above, as well as to indirectly help deal with the mixed blessing of short-term capital

5 Basel Committee on Banking Supervision, International convergence of capital measurements and capital
standards (Basel II), Part 2, section II: “Credit risk approach: the standardised approach”, paragraphs 53
and 54 says “Credit Risk weight A. Exposures to central governments and central banks. 1. The exposures to
the Central Government, the Bank of Spain and other central governments and central banks of other
countries of the UE, denominated and funded in the local currency of the Member State concerned, as well as
against the European Central Bank, shall be weighted at 0%” (own translation).
6 For example, in the case of Spain’s interpretation of Basel II, Circular 3/2008 of the Bank of Spain says:
“Credit Risk weight A. Exposures to central governments and central banks. 1.The exposures to the Central
Government, the Bank of Spain and other central governments and central banks of other countries of the EU,
denominated and funded in the local currency of the Member State concerned, as well as against the
European Central Bank, shall be weighted at 0%” (own translation)..

86 BIS Papers No 67

inflows. In this sense, there is consensus that when capital flows take the form of short-
term financial capital they are mostly driven by international investors’ appetite for risk
and international liquidity conditions, and tend to lead to currency appreciation, asset
price bubbles, and indebtedness levels that are not compatible with the receiving
countries’ productive capacity. These kinds of effects are more important in small open
economies with limited banking and capital markets.
When capital flows turn – sudden stops – employment and activity levels are severely
damaged, and serious financial crises could break out.7
In order to partially avoid the side effects of short-term capital inflows on monetary
equilibrium, the Argentine central bank carried out sterilized operations in the FX
market. Domestic Lebacs and Nobacs play a key role in this process. Moreover, those
instruments’ “domestic” characteristic helps prevent a problem like the one that recently
affected the Bank of Israel, where the debt instruments that the central bank was using
to sterilize were heavily demanded by foreign investors, producing, at the end of the
day, a negative feedback loop.
Regarding securities valuation, the Argentine central bank established that the
sovereign bonds launched in the debt swap of 2005 could be recorded by banks as:
(1) the book value of net-delivered instrument regularization, or (2) the value of the sum
of nominal cash flow until the final maturity of the bonds. The analysis was based on
US accountancy rules (FASB 15), which allow financial institutions to register
government securities in order not to incur accounting losses. Accounting for
government securities at technical and non-market value has the advantage of
removing volatility in banks’ income statements, helping reduce financial market panic
at times of crisis (Com A 5180, April 2011).
4. Institutional arrangements: interaction between central banks and
Treasuries during periods of crisis
In this section we discuss institutional arrangements during the crisis between national
Treasuries and central banks, focusing on the role of the latter. This is important, as was
already pointed out, considering that public debt paths have a direct impact on monetary and
financial stability conditions that ultimately must be addressed by the monetary authorities.
One of the usual analyses focuses on the formal objectives, the mandate established in
central banks’ charters. While some of them have price stability as their main goal, for
example the ECB (“the primary objective … shall be price stability”) and the Bank of England,
in others this objective is complemented with other aspects of the economy, as in the case of
the Bank of Canada, the Bank of Japan8 or even the Federal Reserve.
As regards central bank independence in formulating and implementing monetary and
financial policies, formal frameworks can differ. On the one side, there are banks that have
significant autonomy to implement monetary policy and in relation to other authorities such
as those in charge of fiscal policy and the executive or legislative branches. They have to

7 Read more about this in the Box in BCRA, Recent measures taken by Central Banks from emerging
economies in view of capital inflow, Inflation Report, Second Quarter 2011, www.bcra.gov.ar.
8 “Article 4. (Relationship with the Government): The Bank of Japan shall, taking into account the fact that
currency and monetary control is a component of the overall economic policy, always maintain close contact
with the government and frequently exchange views, so that its currency and monetary control and the basic
stance of the government’s economic policy shall be mutually compatible.”

BIS Papers No 67 87

communicate decisions and to report them, at least formally, only at a given frequency. This
is the case of the ECB9 or the Bank of England.
In the case of the ECB, in accordance with paragraph 3 of Article 284 of the Treaty on the
Functioning of the European Union, it must submit annually to the Parliament, the Council,
the Commission and the European Council a report on the activities of the ESCB (European
System of Central Banks) and the development of monetary policy in the previous and the
current year. Also, ECB authorities, usually the President, attend quarterly hearings at the
European Parliament (Committee on Economic and Monetary Affairs). In some cases, other
members of the Executive Committee may also be asked to attend to these hearings.
Furthermore, once a year, following its practices and customs, the ECB presents to the
Members of Parliament the previous year’s Annual Report. Until 2009 the report was
presented to the Committee on Economic and Monetary Affairs. Since 2010 the Annual
Report has been presented by the ECB President to the plenary of the European Parliament.
The ECB President also gives speeches on different economic issues, including fiscal
policies.10
Moreover, some central banks have in their legal frameworks special provisions for their
interaction with the Treasuries and have more formal links with legislative bodies. This group
includes the Federal Reserve and the Bank of Japan.11 For instance, the BoJ authorities,
usually the President, attend twice-yearly hearings in the Japanese parliament, in both the
House of Representatives and the House of Councilors, before the Committee on Finance.
In addition, during the crisis, some regulations were modified in order to clarify the legal
relationship between government agencies, as was the case of the Federal Reserve and the
US Treasury regarding unconventional monetary policies.12
These examples show that some central banks departed informally and to a certain extent
from their institutional arrangements, following an eclectic strategy to foster economic activity
and employment.

9 “Article 7. Independence. As set out in Article 130 of the Treaty on the Functioning of the European Union,
when exercising the powers conferred by the Treaties and this Statute and carrying out the functions and
duties, neither the ECB nor the national central banks or any member of its governing bodies shall seek or
take instructions from union institutions, bodies or agencies, from any government of a Member State or from
any other body. The institutions, bodies or agencies of the Union and the Governments of the Member States
undertake to respect this principle and not seek to influence members of the governing bodies of the ECB or
national central banks in the exercise of their functions.”
10 As an example, on December 8, 2011, Mario Draghi stated: “Turning to fiscal policies, all euro area
governments urgently need to do their utmost to support fiscal sustainability in the euro area as a whole. A
new fiscal compact, comprising a fundamental restatement of the fiscal rules together with the fiscal
commitments that euro area governments have made, is the most important precondition for restoring the
normal functioning of financial markets. Policy-makers need to correct excessive deficits and move to
balanced budgets in the coming years by specifying and implementing the necessary adjustment measures.
This will support public confidence in the soundness of policy actions and thus strengthen overall economic
sentiment To accompany fiscal consolidation, the Governing Council has repeatedly called for bold and
ambitious structural reforms. Hand in hand, fiscal consolidation and structural reforms would strengthen
confidence, growth prospects and job creation. Key reforms should be immediately carried out to help the euro
area countries improve competitiveness, increase the flexibility of their economies and enhance their
longer-term growth potential. Labour market reforms should focus on removing rigidities and enhancing wage
flexibility. Product market reforms should focus on fully opening up markets to increased competition”
(extracted from M. Draghi and V. Constâncio, introductory statement to the press conference of 8 December
2011, Frankfurt).
11 Board of Governors of the Federal Reserve System, Federal Reserve Act, Section 10.
12 Board of Governors of the Federal Reserve System, Federal Reserve Act, Section 13.

88 BIS Papers No 67

As mentioned before, taking into account recent experiences, in Argentina the central bank’s
charter was modified by Congress this year. It changed the unique goal of preserving the
value of the currency to a multiple mandate which provides that (Article 3), under policies set
by national authorities, its aims are to promote monetary stability, financial stability, and
economic development with social equity. Financial stability and monetary stability were
added, which are goals that many countries have explicitly incorporated after the devastating
effects of the financial crisis.
Through the changes introduced, the central bank can regulate credit conditions regarding
terms, interest rates, commissions, and other charges, as well as guide the granting of credit
through reserve requirements, differential reserves, and other appropriate means.
The central bank continues to enjoy autarky and is not subject to orders or instructions of the
executive branch, although it aims for greater coordination with other government policies.
The Charter also establishes that the Bank shall perform, among others, the following duties,
some of them relating to its relationship with the rest of the government. Article 4 provides
that the central bank shall:
c) act as a financial agent for the Nation, and as depository and agent for the country
before international monetary, banking and financial entities, of which the Nation is a
member
f) implement an exchange policy in accordance with such legislation as the National
Congress may lay out
h) provide for the protection of the rights of users of financial services and competition
According to Article 10, the BCRA’s president shall:
I) submit an annual report on the BCRA’s transactions to the National Congress for
consideration. In addition, the president shall attend public and joint sessions of the
Budget and Treasury Committees of both Chambers, the Economy Committee of
the Senate, and the Finance Committee of the House of Representatives at least
once during the general term or whenever any of these Committees may ask him to
attend for reporting on the scope of the monetary, exchange and financial policies
under way.
Articles 12, 26, and 29 relate to the relationship between the central bank and the Economy
Ministry. They provide that the Economy Ministry shall participate on the central bank Board
with voice but without vote, that the Bank shall inform the Economy Ministry on monetary,
financial, exchange, and credit regimes, and that the Bank shall advise the Ministry and
Congress on the exchange system, and establish the relevant general regulations.
5. Final comments
It seems to be clear that in different countries, economic authorities and markets have not
properly considered the size and evolution of the stock of public debt and the government
primary surplus required to have a sustainable path. Moreover, in some countries the
monetary authorities focused their economic assessment excessively on inflation. Under that
approach, it seems that there was some disregard of key macroeconomic variables and
ratios.
In this paper we have analyzed the reasons that could explain why some macroeconomic
variables were able to follow unsustainable medium- and long-term trends for a long period.
We also examined the role played by economic authorities, and the inter-relationships
among them in the design and implementation of fiscal policy and debt management.

BIS Papers No 67 89

In this sense, the crisis that took place during 2008–11 shook the paradigm that ruled
macroeconomic theory, and specifically monetary policy, in a way that has not been
observed since the Great Depression. Consequently, rigid central bank goals and inflexible
boundaries between the central bank and the Treasury were erased, letting economic
authorities behave in a pragmatic way.
In the paper we also discussed the role played by credit rating agencies and regulatory
frameworks. The former showed pro-cyclical behavior, producing strong swings in ratings
well after the crisis erupted. With regard to the latter, in a context of broad liquidity, regulatory
frameworks included favorable incentives, in terms of lower capital integration, for the
maintenance of sovereign debt as an important part of assets in some financial systems.

BIS Papers No 67 91

Fiscal consolidation and macroeconomic
challenges in Brazil
Carlos Hamilton Araújo, Cyntia Azevedo and Sílvio Costa1
Abstract
This paper explores two important points regarding the Brazilian fiscal framework. The first
part analyses the significant improvement of the fiscal stance in the last decade as the result
of the promotion of fiscal discipline and debt management policies. This consolidation is
argued to be one of the reasons why Brazil has not been subject to the same concerns about
debt sustainability that have become a focal point in most developed economies. The second
part studies how the coordination between monetary and fiscal policies is important in
dealing with the issues that arose in the aftermath of the recent crisis. By using models
simulated with Brazilian data, we show that the implications for inflation of implementing a
fiscal retrenchment policy crucially depend on which instrument is being used and on the
behaviour of monetary policy.

Keywords: Fiscal consolidation, fiscal policy, debt management, monetary policy,
macroeconomic stabilization
JEL classification: E52, E62, E63

1 The authors are, respectively, the Deputy Governor for Economic Policy and analysts at the Research
Department of the Central Bank of Brazil. We thank Adriana Soares Salles, Eduardo J. A. Lima and
André Minella for comments.

92 BIS Papers No 67

1. Introduction
The financial crisis has highlighted the importance of coordination between monetary,
macroprudential, and credit policies. Fiscal policy proved to be outstanding in tackling the
urgent challenges that arose following the financial bump, attenuating the depth of the crisis
and ensuring the resilience of the financial system. Indeed, governments of many developed
countries2 have used fiscal instruments to supply broad liquidity for firms, banks, and credit
markets, while traditional fiscal policies have also been employed to stimulate the economic
recovery. These actions helped reverse the recessionary process, improve financial
conditions and contributed to the upturn in market confidence (IMF, 2009). Unfortunately,
these measures are never without cost, and some countries built up such unsustainable
public account imbalances that a structural solution is dramatically needed3.
Acknowledged as the second stage of the financial crisis, the fiscal turmoil struck several
developed countries, but Brazil has not undergone the same difficulties4. That can be
explained by a virtuous combination of effective policy reactions during the crisis and
strengthened fiscal conditions in the years preceding the crisis. Countercyclical policies were
implemented immediately after the tightening of financial conditions5. Moreover, some
targeted fiscal measures were undertaken in order to stimulate the recovery of aggregate
supply6. Nevertheless, at the current phase of the international crisis, Brazil is required to
demonstrate its strong commitment to the fiscal framework by fully meeting the primary
surplus target.
The set of measures were successful in getting the economy out of the initial negative impact
on GDP in the fourth quarter of 2008. Only two periods ahead, quarterly output started
recovering towards pre-crisis growth rates7. Real interest rates dropped markedly, from a
13.75% yearly rate in September of 2008 to 8.5% in August of 2009. On the other hand, the
consumer price index ended 2009 at 4.3%, below the 4.5% inflation target. However, there
were signs of rising pressures on prices, and the interest rate started to suddenly increase in
April 2010, after remaining steady for eight months.
Managing internal pressures on aggregate demand and, at the same time, recessionary
conditions abroad leading to lower international interest rates, is complicated. Indeed,

2 In fact, the United States, the Euro Zone, Japan, the United Kingdom and Canada undertook a series of
initiatives to stabilize the financial system, such as capital injections, asset purchases, loans to financial firms,
guarantees of financial assets and bank liabilities and deposit insurance, all of them with significant effects on
public debt (IMF, 2009).
3 Between 2007 and 2010, the net public debt outstanding in terms of GDP increased sharply in many
developed countries, for example, the United States (from 42.9% to 68.3%), Germany (50.2% – 57.6%),
Japan (81.5% – 117.2%), the United Kingdom (38.2% – 67.7%), and France (59.5% – 76.6%). See IMF
(2011).
4 The Brazilian net public debt decreased from 45.1% to 40.2% in percent of GDP between 2007 and 2010.
5 For example, monetary policy easing, reduction of reserve requirement rates for banks, increase in directed
credit policies, supplying liquidity in foreign currencies, and a other policies were implemented between
September 2008 and August 2009.
6 Taxes on industrial products (IPI) like vehicles, durable consumer goods, building materials, and capital goods
were cut at the end of 2008. In turn, taxes on financial operations (IOF) for lending to households also
dropped.
7 Quarterly GDP fell by 4.20% in the last quarter of 2008, and the drop in investment (-10.18%) led the results.
In the first quarter of 2009, government consumption reacted by increasing 3.96%, in comparison to the fall
(-3.66%) in the previous quarter. The effective countercyclical measures were important to reverse local
expectations in spite of a volatile scenario abroad, and investment leapt to 5.99% in the second quarter of
2009, with high rates observed in following quarters as well. Likewise, output grew up by 2.01% in the second
quarter of 2009.

BIS Papers No 67 93

emerging markets around the world showing strong growth and financial resilience share the
same situation in which monetary policy could lead to higher interest rate spreads and attract
large global capital inflows. In order to deal with this dilemma, Brazil adopted a series of
policies that included strengthening the macroprudential framework to ensure that financial
risks are contained, allowing appreciation of the exchange rate and accumulation of external
reserves and adjusting the mix of monetary and fiscal policy in order to assure a sustained
pace of demand growth and to keep inflation under control and converging towards the
target.
Hence, there are two very interesting points to explore with regard to the Brazilian fiscal
framework. The first is the fiscal consolidation that has been ongoing since 1999, which can
explain to a large degree why Brazil has not been subject to the same concerns about debt
sustainability that are currently a focal point in most developed economies. The second issue
is the importance of coordination between fiscal and monetary policy to deal with the
challenges of the present conjuncture. In fact, simulations performed by models estimated
with Brazilian data show a tighter fiscal policy could lead to meaningful decreases in
inflationary pressures, even when the effort is short-lived, and that a long-lasting policy could
imply significant structural changes in the long run.
2. The Brazilian fiscal framework
Brazil’s recent economic policy can be described by a framework based on three main
guidelines implemented in 1999: a floating exchange rate, an inflation target regime and
fiscal austerity. In that year, targets for the fiscal surplus as a ratio of GDP (on average
above 3%) were announced, and these have been an important guideline for policy since
then. As shown in Graph 1, in the first year there was already a significant increase of the
primary surplus to 2.92% after being null in 1998. The following year, the Fiscal
Responsibility Law (FRL) was enacted to strengthen fiscal institutions and establish a broad
framework of fiscal planning, execution, and transparency at the federal, state, and municipal
levels. It reinforced the goal of promoting fiscal discipline and has helped to obtain consistent
surpluses, even during the crisis.
The fiscal consolidation process has been the result not only of the implementation of the
FRL and of meeting the primary surplus target, but also of the efforts made by the Central
Bank of Brazil (BCB) and the Treasury Department regarding the management of public
debt8. The government has been working to promote fiscal discipline meant to reduce
indebtedness and is also following a set of guidelines to enhance the debt profile9. These
include the reduction of short-term debt and lengthening of the average debt maturity,
progressive replacement of overnight rate-indexed and dollar-indexed securities by fixed rate
and inflation-indexed securities, the expansion and diversification of the investor base, and
the stimulation of the secondary market for public debt.

8 Turner (2002) argues that an important reason for fostering debt markets is that such markets can contribute
to the operation of monetary policy. The author points out how essential for the smooth transmission of policy
this market is. Besides, the long-term market also gives relevant information about expectations of likely
macroeconomic developments and about market reactions to monetary policy actions.
9 These improvements are a trend observed in most EMEs, as pointed out in the background paper
“Developments of domestic government bond markets in EMEs and their implications”.

94 BIS Papers No 67

Graph 1
Primary surplus
As a percentage of nominal GDP

Source: BCB.
The set of fiscal measures10 adopted has proven to be very effective in helping the
government move towards these goals. Graph 2 shows the notable decrease in the net debt-
to-output ratio since 2001. In 2002, in the middle of political turmoil, with significant currency
depreciation, the debt-to-GDP ratio peaked at 62.86%. Since then, it has shown a downward
trend, especially since mid-2006 when the country started accumulating external surpluses.
In September 2011, the ratio reached 36.49%, the lowest value in the observed series.
The latest quarterly Inflation Report (BCB, 2011a) presents projections for selected fiscal
variables, as in Table 1.

Table 1
Estimates of fiscal variables (% GDP)1
PSND GGGD Nominal deficit Nominal interest
2012 35.7 51.9 1.2 4.3
2013 33.8 48.8 1.1 4.2
2014 31.3 45.7 0.5 3.6
2015 28.9 43.0 0.3 3.4
2016 26.1 40.4 –0.1 3.0
1 Consider the primary surplus expected in Lei de Diretrizes Orçamentárias (Budget Guidelines Law) for 2012,
and 3.1% of GDP for the other years.
Source: Inflation Report (BCB, 2011a).

10 See Figueiredo et al. (2002) for a description of the measures implemented in order to fulfill these guidelines.

BIS Papers No 67 95

They were formulated assuming the primary surplus target is met and using market
perspectives for the main indexation indicators and projections for output presented in the
same report. The public sector net debt (PSND) and general government gross debt (GGGD)
are expected to continue their descending trajectory until 2016. The same is expected to
happen to the nominal deficit and interest payments.
Graph 2
Public sector net debt
As a percentage of nominal GDP

Source: BCB.
Besides promoting debt reduction, fiscal consolidation has also achieved very positive results
regarding the guidelines established to improve the domestic debt profile.
With regard to the composition of domestic debt outstanding, the guideline to increase the
share of fixed rate securities and simultaneously reduce the share of dollar- and overnight
rate-indexed securities has been clearly followed in the period, as shown in Graph 3.
Although still high, the proportion of securities indexed to the overnight interest rate (Selic)
dropped from 54.4% in December 2001 to 26.2% in December 2011. This was an important
step towards fostering the efficacy of monetary policy, since this sort of security exacerbates
the wealth effect generated by increases in the nominal interest rate.
Over the same period, the share of indexed securities dropped from 29.5% to 0.5%, reducing
the exposure of domestic debt to exchange rate volatility. Besides the reduction in new
issuance, the appreciation of the exchange rate11 has also contributed to the reduction of the
share of this type of security. Meanwhile, fixed rate and inflation-indexed securities
significantly increased their share, from, respectively, 8.1% and 7.2% in December 2001 to

11 It dropped from a peak of R$3.89/US$ in September 2002 to R$1.74/US$ in December 2011.

96 BIS Papers No 67

32.6% and 25.2% in December 2011. One aspect of economic policy that certainly
contributed to the attractiveness of fixed rate securities was the downward trend of the
nominal interest rate observed in recent years12.

Graph 3
Composition of public debt outstanding
As a percentage of total debt

Source: BCB.

The average maturity and proportion of debt expiring in less than 12 months in total debt
outstanding are presented in Graph 4. The average maturity (left axis) decreased until 2005,
reaching a trough of 27 months in November. Since then, it has lengthened significantly, to
42 months in December 2011. At the same time, moving in the opposite direction, the
proportion of short-term debt (right axis) was very volatile early in the period, but since mid-
2004 has significantly improved, staying on average below 25% in 2011.
Another improvement was the expansion and diversification of the investor base, as shown
in Graph 6. Data are not available prior to January 2007, but from then a prominent increase
in holdings by foreign residents is observed. They jumped from 1.6% at the beginning of the
series to 11.3% in December 2011. One aspect that certainly contributed to attracting foreign
investment in domestic debt was the “investment grade” rating granted by Standard & Poor’s
in April 2008, followed by Fitch, which granted the same rating to Brazil the following month.
In 2009, Moody’s also raised Brazil’s rating to the “investment grade” category. In 2011, all
three agencies increased the rating by one more level.

12 In December 2001 the Selic rate was at 19%; it reached a peak of 26.5% in June 2003, but has been falling
since then, reaching 11% in December 2011.

BIS Papers No 67 97

Graph 4
Average maturity and proportion of short-term debt
in public debt outstanding

Source: BCB.

Graph 5
Short-term debt by indexation
As a percentage of total debt by indexation

98 BIS Papers No 67

Source: BCB.
Graph 6
Holdings of general government debt
As a percentage of total debt

Source: Tesouro Nacional.
Obs.: Data for holdings by the government are available only after January 2011. Prior to that, the
stocks were allocated between financial institutions and investment funds.
Changes were also observed regarding fostering the trading of domestic securities on the
secondary market. The volume of operations (Graph 7, left axis) presented a steep upward
trend over the past decade, with a significant boost after the implementation of the new
payment system in 2002. The ratio between the volume of operations and outstanding debt
jumped from an average of 19% in 2000 to 27% in 2011. The average maturity of
operations13 (Graph 7, right axis) also increased following the improvement of liquidity in the
secondary market.
This outlook shows how Brazil’s debt management policy advanced in recent years, working
towards the fulfillment of the guidelines established in the early 2000s. Since 2001, the
Treasury Department has been publishing an annual borrowing plan (“Plano Anual de
Financiamento” – PAF) for debt management, and the criteria analyzed above have been
maintained as the main goals for domestic debt policy. The mission established in the plan is
to ensure that the government’s financial needs and payment obligations are met at the
minimum possible cost in the medium and long term, while keeping risks at a prudent level,
and to contribute to better operation of the debt market (Tesouro Nacional, 2011).

13 Data only available after January 2004.

BIS Papers No 67 99

Graph7
Volume and average maturity of operations in the
secondary market for domestic securities

Source: BCB.
3. Effects of fiscal policy on inflation and output
Beyond the structural efforts of fiscal consolidation, Brazil has also faced challenges in terms
of fiscal policy. Macroeconomic conditions have created a tension between foreign capital
inflows and domestic factors, like high domestic growth. Specifically, capital inflows are
quickly increasing the foreign capital share in bank funding sources, enabling small and
medium-sized banks to scale their credit supply, and posing important issues about the
stability of the financial system.
Besides the medium-term concerns about capital inflows, the rapid expansion of credit in
Brazil also has significant effects on inflation today. Indeed, excess credit supply has driven
lending rates down and lengthened the maturities of credit contracts, despite the lack of
reasonable improvements in borrowers’ profiles. In other words, current credit expansion
generated by large capital inflows has boosted aggregate demand and amplified pressures
on inflation.
For that reason, the BCB has implemented several macroprudential policies to contain the
unsustainable credit expansion. But the current general scenario requires a vast and
coordinated series of measures to decrease inflationary pressures and, at the same time,
sustain the economic growth and prevent the formation of asset price bubbles. In this
context, fiscal policy has a lot to contribute.
In order to analyze the effects of fiscal policy on inflation and output in quantitative terms, the
BCB has been performing various simulations using some of the analytical models available.
Two recent editions of the quarterly Inflation Report – March and December, 2011 – came up
with noteworthy results.

100 BIS Papers No 67

In March’s edition (BCB, 2011b), simulations were performed by running a medium-size
semi-structural model14 estimated with Brazilian data. The fiscal tightening is exogenous,
sized as equivalent to 1% of GDP, and lasts for four consecutive quarters. Two scenarios for
monetary policy were considered: non-accommodative, in which the interest rate regularly
follows the estimated Taylor rule reactions, and accommodative, in which the interest rate
remains constant for four quarters but reacts according to the Taylor rule afterwards.
The results show that a contractionary fiscal impact can cause sudden, significant, and
longstanding effects on inflation. The lack of action of an accommodative monetary policy
can notably amplify the results since the nominal interest rate is unable to respond to the
deflationary pressures. The transmission mechanism considered by the model is essentially
the direct aggregate demand reduction, which is amplified if the interest rate is not allowed to
go down to stabilize the economy.
A more structural analysis regarding the transmission channels can be addressed by
performing simulations in medium-sized dynamic stochastic general equilibrium (DSGE)
models. The standard model used by the BCB for forecasting and policy analysis is known
as SAMBA15. Besides enabling the same scenarios, the structural model allows us to
distinguish clearly between government consumption and public tax revenue changes,
although the impact on the primary surplus is the same. The December Inflation Report
(BCB, 2011c) gave those results.
In line with the results of the semi-structural model, in SAMBA a tightening of fiscal policy
causes an initial and consistent drop in inflation whose effects last for several periods. The
non-accommodative monetary policy also shows lower impacts in comparison to the reactive
interest rate rule scenario. Contraction of public spending is interpreted as a temporary
movement of the primary surplus target, which leads to an immediate reduction in demand
for consumer goods and a direct fall in aggregate demand. So, second-order effects in the
adjustment dynamic take place following the contraction in demand, such as downfalls in
labor, wages, and rule-of-thumb16 households’ consumption. Reduction of production inputs
leads to lower marginal cost, which explains much of the fall in inflation.
By raising taxes equivalently instead of cutting public spending, the fiscal authority will face
other challenges in terms of timing and magnitude because the transmission channels
involved are quite different. For instance, an increase in the rule-of-thumb households’ tax
rate produces an immediate reduction in demand, but the products purchased by households
consist of both domestic and foreign goods, whereas government consumption is based on
domestic goods only. The initial contractionary impacts on consumption17 thus partially
spread abroad, which reduces the second-order effects passing through the supply side. Not
surprisingly, marginal cost and inflation fall less than when the fiscal policy is built on
spending cuts.
Graph 8 compares the magnitudes of the effects on inflation of those shocks on both sides of
government balance sheet. Although the timing of transmission seems quite similar in all
scenarios, with maximum effects taking place between the fourth and fifth quarters after the

14 See Minella and Souza-Sobrinho (2009) for further details.
15 Stochastic Analytical Model with a Bayesian Approach; see Castro et al. (2011). The model was developed
and estimated by the BCB.
16 Rule-of-thumb households (equivalently, non-Ricardian or hand-to-mouth households) are agents which face
technological restrictions to transfer resources from one period to the next. Models usually have a certain
fraction of non-Ricardian households as a kind of abstraction to simulate actual constraints on a fraction of
consumers. In SAMBA, this group consumes all disposable income each period.
17 The current version of SAMBA does not have as comprehensive an approach for taxation as some macro models
focused on fiscal issues. Therefore, tax rate increases will primarily impact only household consumption.

BIS Papers No 67 101

impact, the fall in inflation caused by public spending cuts is more intense than when
increases in tax rates are used as a fiscal instrument. For purposes of comparison, the
overall effect on inflation of cuts in government spending is, on average, 1.6 larger than an
equally sized impulse driven by increasing public revenue.

Graph 8
Effects on inflation of a tightening in fiscal policy equivalent
to 1% of GPD and lasting for 4 quarters

Source: BCB.
The lower effects of incentives on taxation should be considered carefully. Tax revenues are
collected by wage taxes that directly impact only non-Ricardian households’ consumption.
Traditional distorting mechanisms like taxation on investment and production are not factored
into the model, nor are public investment and other aspects of government spending. In
theory, this could explain why the effects of tax increases are smaller than expected.
Effects on output were also addressed. The public spending multiplier, measured by SAMBA,
is around 1.2 under a non-accommodative monetary policy, and about 0.9 if calculated with
the semi-structural model. In turn, the tax revenues multiplier is 0.9.
Results yielded by both the semi-structural and DSGE models used by the BCB are closely
in line with the theoretical literature and the practical experience in central banks. Coenen et
al. (2010) study fiscal multipliers and effects on inflation of several fiscal instruments by
performing simulations in DSGE models used by seven international institutions, such as the
Federal Reserve Board, European Central Bank, Bank of Canada, European Commission,
OECD, and IMF. The effects on inflation depend to a great extent on what policy instrument
is used. However, it is clear that a non-accommodative monetary policy always generates
lower responses, because accommodative monetary policy allows real interest rates to fall
further, leading to greater responses in consumption and investment.

102 BIS Papers No 67

4. Final remarks
Fiscal policy has an important role in the policy balance. Indeed, both public revenue and
government spending can be effectively used to tighten aggregate demand, although their
different transmission channels and total effects in the economy must be taken into account.
The government’s significant direct participation in the credit markets and historical role as
an investment catalyst in Brazil are two factors that demonstrate the importance of consistent
use of fiscal instruments as tools to help achieve macroeconomic stability.
Brazil’s fiscal consolidation has demonstrated the effectiveness of the long-term framework,
as seen during the financial crisis. Notwithstanding, a more sustained strengthening of fiscal
conditions is needed for the Brazilian economy to reach a new baseline for monetary and
fiscal frameworks in the medium term, by enhancing the sustainability of the public debt, the
investment-savings dynamic, and the broad mechanisms of price setting.
References
Banco Central Do Brasil (2011a). “Perfil e Projeções do Endividamento Público.” Inflation
Report, v. 13, n. 4, December, 2011. Available at http://www.bcb.gov.br/?RELINF.
Banco Central Do Brasil (2011b). “Multiplicador Fiscal, Produto e Inflação.” Inflation Report,
v. 13, n. 1, March, 2011. Available at http://www.bcb.gov.br/?RELINF.
Banco Central Do Brasil (2011c). “Efeitos na Inflação e no Produto de Variação de Gastos e
Impostos.” Inflation Report, v. 13, n. 4, December, 2011. Available at
http://www.bcb.gov.br/?RELINF.
Castro, Marcos R. de, Gouvea, Solange N., Minella, Andre, Santos, Rafael C., Souza-
Sobrinho, Nelson F. (2011). “SAMBA: Stochastic Analytical Model with a Bayesian
Approach.” Banco Central do Brasil, Working Paper Series, n. 239. April, 2011. Available at
http://www.bcb.gov.br/?workingpapers.
Coenen, G., Erceg, C., Freedman, C., Furceri, D., Kumhof, M., Lalonde, R., Laxton, D.,
Lindé, J., Mourougane, A., Muir, D., Mursula, S., de Resende, C., Roberts, J., Roeger, W.,
Snudden, S., Trabandt, M., E in’t Veld, J. (2010). “Effects of Fiscal Stimulus in Structural
Models.” IMF Working Paper, n. 10/73.
Figueiredo, Luiz F., Fachada, Pedro, Goldstein, Sergio (2002). “Public Debt Management
and Open Market Operations in Brazil”. Bank for International Settlements, BIS Papers,
n. 11.
International Monetary Fund (2009). “Unconventional Choices for Unconventional Times:
Credit and Quantitative Easing in Advanced Economies.” IMF Staff Position Note,
November 4th, 2009.
International Monetary Fund (2011). Fiscal Monitor. Addressing Fiscal Challenges to Reduce
Economic Risks. September, 2011.
Minella, Andre, Souza-Sobrinho, Nelson F. (2009). “Monetary Channels in Brazil through the
Lens of a Semi-Structural Model.” Banco Central do Brazil, Working Paper Series, n. 181.
April, 2009. Available at http://www.bcb.gov.br/?workingpapers.
Tesouro Nacional (2011). Annual Borrowing Plan – PAF. Available at
http://www.tesouro.fazenda.gov.br/ english/public_debt/annual_borrowing_plan.asp.
Turner, P. (2002). “Bond markets in emerging economies: an overview of policy issues”.
Bank for International Settlements, BIS Papers, n. 11.

BIS Papers No 67 103

Macro policies and public debt in Chile
Sebastián Claro and Claudio Soto1
Abstract
This note characterises the evolution of Chile’s public debt, and discusses its implications for
the management of the country’s monetary policy. Historically, the main issuer of public debt
in Chile was the central bank. The government, in turn, has recently started to engage in a
more active debt policy, with the aim of deepening the market for risk-free securities and
diversifying its funding sources. In general, the soundness and predictability of fiscal policy
and the high degree of coordination between the government and the central bank has
meant that the debt policy of the fiscal authority has posed no major challenge for the
conduct of monetary policy. Moreover, the government’s positive net asset position
government has played an important role, allowing the central bank to fulfil its price and
financial stability objectives in spite of a negative equity position.

Keywords: Monetary policy, fiscal policy, public debt
JEL classification: E52, E62, H6

1 Board Member of the Central Bank of Chile and Head of Macro Analysis, respectively. The views expressed in
this note do not necessarily represent those of the central bank. We thank Luis Oscar Herrera, Matías Bernier
and Sergio Salas for their contribution to an earlier version.

104 BIS Papers No 67

I. Introduction
This note outlines the evolution of Chile’s public debt and discusses its implications for
monetary policymaking. Historically, the main issuer of the country’s public debt has been the
Central Bank of Chile (CBC). In the 1980s, a large amount of debt was issued by the
monetary authority to finance operations resulting from the 1982–83 banking crisis. Later, in
the 1990s, debt was issued to sterilise large reserves accumulations. The last two
administrations, in turn, have engaged in a more active debt policy and the share of public
debt issued by the treasury has increased.
In the last few years, two factors have shaped the evolution of Chile’s public debt. On the
one hand, the government has issued debt mainly with a view to deepening the market for
risk-free securities. It has needed to issue only relatively small amounts of public debt to
finance its operations and roll over existing debt. On the other hand, the CBC has issued
long-term bonds to finance two massive reserve accumulation programmes, one in 2008 and
the other in 2011.
Overall, the soundness and predictability of fiscal policy and the high degree of coordination
between the government and the CBC have meant that the conduct of monetary policy has
faced no major challenges. In this context, the country’s positive net asset position has
played an important role: it has allowed the central bank to fulfil its price and financial stability
objectives in spite of a negative equity position.
II. The evolution of public debt in Chile
As mentioned, a large share of Chile’s current stock of public debt was issued by the central
bank (Figure 1). This debt reflects in part the operations implemented after the financial crisis
of 1982, when the monetary authority issued large amounts of bonds in order to finance the
rescue of the Chilean financial system. In this sense, although registered on the CBC’S
balance sheet, this debt has a fiscal origin. More precisely, the central bank acted on behalf
of the fiscal authority in order to provide liquidity and to recapitalise the banking system at a
time when the monetary authority lacked autonomy. Later, when the central bank was
granted autonomy at the end of the 1980s and the size of its capital was defined, the value of
the liabilities arising from the financial rescue were not fully recognised in its balance sheet.
Another significant share of the public debt issued by the CBC is related to the sterilisation of
its massive accumulation of international reserves throughout the 1990s, just prior to the
Asian crisis. This build-up occurred in the context of large capital inflows to the country,
during a period in which the CBC maintained a target zone for the exchange rate.
Since the start of the last decade, total public debt as a share of GDP began to fall, declining
from about 37% of GDP in 2001 to 27% of GDP in 2011. The main reason was the end of the
CBC’s massive accumulation of reserves. Without the large capital inflows of the 1990s, and
after the target zone for the exchange rate band had been abandoned, the monetary
authority stopped systematically intervening in the exchange market, obviating the need to
sterilise such interventions. Nominal debt stabilised and, thanks to sustained growth, started
to fall as a share of GDP. In 2011, due to an extraordinary programme of reserves
accumulation and the corresponding sterilisation, the CBC’s debt went from around 10% of
GDP in 2010 to 16% of GDP by the end of 2011.
Government debt has remained more or less stable over the last 15 years, oscillating
between 5% and 10% of GDP. It has increased in recent years due to explicit policy
objectives to establish benchmarks for risk-free securities prices, and also to diversify the
currency composition of public funding. In the next section, we discuss Chile’s fiscal policy in
greater depth and how it has determined the evolution of the government’s debt. Then, we
analyse the case of monetary policy.

BIS Papers No 67 105

Figure 1
Chile’s public debt owed to the private sector
% of GDP

Sources: Central Bank of Chile and DIPRES (Budget Office).
III. Fiscal policy and public debt
Over the past several decades, the Chilean government has managed its fiscal policy rather
conservatively, with positive balances in 14 out of the last 21 years. As a result, large stocks
of public assets have been accumulated and the government has needed to issue only
relatively small amounts of public debt to finance its operations and roll over existing debt.
Chile’s prudent approach to the conduct of public finance was institutionalised in 2006 by the
Fiscal Responsibility Act. This legislation obliges the government to adopt an explicit fiscal
target for its structural budget so as to keep fiscal expenditure in line with long-term or
structural fiscal revenues. A fiscal rule had previously been established, in 2001, albeit
without any binding commitment. Since the adoption of this rule – and against the backdrop
of a large terms-of-trade windfall – the government’s net position has improved considerably,
moving from a net debtor to a net creditor position. However, since 2003, the Treasury has
made some significant public debt issuances, both domestically and internationally, with the
aim of adding liquidity to the domestic bond market, helping to establish benchmark prices
and diversify public financing sources. In addition, these domestic currency issuances have
obviated the need for the government to immediately liquidate all the revenues it receives in
foreign currency.2

2 The Chilean government has a structural currency mismatch between its revenues and its expenditures. A
significant part of its revenues are denominated in foreign currency while most of its expenditures are in local
currency.

106 BIS Papers No 67

Since 2001, Chile’s fiscal policy has been guided by a fiscal rule based on the concept of the
structural balance. In Chile, government revenues are influenced not only by the level of
economic activity but also by the price of copper and other minerals.3 With the main goal of
smoothing fiscal policy over a medium-term horizon, government expenditure is set each
year at a level compatible with a target for the structural balance, where revenues are
adjusted according to the cyclical position of the economy and copper price deviations from
the metal’s long-run value.4
Shortly after the implementation of the structural budget fiscal rule in 2001, the government
mandated two independent committees to provide the key variables on which the structural
balance is computed: trend GDP and the reference or long-run price of copper. In 2006, the
Fiscal Responsibility Act was enacted, further strengthening the fiscal rule by requiring each
incoming government to (i) announce its target for the structural balance during its term;
(ii) estimate the expected outcome of its fiscal policy on the structural balance for the public
budget law; and (iii) annually report the actual outcome of the structural balance.
Between 2001 and 2006, the target for the structural balance was set at 1% of GDP. Then,
for 2007 and 2008, the target was reduced to 0.5% of GDP and in 2009 it was originally set
at 0% of GDP. During this period, expenditures moved in tandem with structural revenues
and the government successfully met its target, except in 2008. In 2009, an escape clause
was put in place so that a countercyclical fiscal policy could be implemented that would
support aggregate demand in response to the global financial crisis. During that year, the
structural deficit amounted to 3% of GDP. In 2010 and 2011, it was reduced but remains
sizeable (Table 1). The present administration has announced that it expects to bring the
structural deficit back to 1% of GDP by 2014.
Table 1
Public balance
% of GDP
Expenditure Revenues Effective balance
Structural
revenues
Structural
balance
2001 22.3 21.8 –0.5 23.4 1.1
2002 22.3 21.1 –1.2 23.2 0.8
2003 21.1 20.7 –0.5 22.0 0.8
2004 19.9 22.0 2.1 21.0 1.0
2005 19.3 23.8 4.6 20.4 1.1
2006 18.1 25.8 7.7 19.5 1.4
2007 18.7 26.9 8.2 19.8 1.1
2008 21.2 25.5 4.3 20.4 –0.8
2009 24.8 20.4 –4.5 21.9 –3.0
2010 23.4 22.9 –0.4 21.3 –2.0
2011 (*) 23.3 24.5 1.2 21.7 –1.6
(*) Estimated. Source: DIPRES (2011).

3 Income from copper and other minerals greatly influences revenues accrued through the profits of Codelco,
the state-owned mining company, and from taxes on private mining companies, which started to make
significant profits in 2005.
4 The current fiscal rule evolved from previous attempts to insulate fiscal policy from copper price fluctuations. A
Copper Stabilisation Fund was set up as early as 1985 in order to help smooth public revenues.

BIS Papers No 67 107

Despite the fiscal expansion of 2009 and the subsequent public deficit in 2010, the
government’s net position is still positive. During the 1990s, net liabilities were reduced
systematically until the Asian crisis hit. After that episode, the government posted deficits
until 2003 with a consequent increase in net liabilities. Then, from 2005 onwards, the
government’s net position turned positive. Currently, its net assets amount to slightly less
than 10% of GDP (Figure 2).
When the net position of the government became positive, two sovereign funds were
created: the Pension Reserve Fund (PRF) established at the end of 2006 to fund fiscal
pension obligations, and the Economic and Social Stabilisation Fund (ESSF), set up in early
2007 to help cover fiscal deficits and/or repay public debt. Given its experience in managing
Chile’s international reserves, the operation of the two funds was delegated to the Central
Bank of Chile.5 In addition, an independent Financial Committee was set up to advise on
investment policy. By March 2012, the two funds held aggregate assets of slightly more than
USD 19 billion.
Figure 2
Total and net liabilities of the central government
% GDP

Source: DIPRES and Chile’s Ministry of Finance.
As mentioned above, the government has regularly issued domestic bonds since 2003
regardless of its net creditor position. Two types of bonds have been sold: nominal bonds
(Bono Tesorería en Pesos, BTP) and inflation-indexed bonds (Bono Tesorería UF, BTU).
The stated aim of these issuances is to enhance bond market liquidity in Chile by setting

5 The central bank acts as the government’s fiscal agent for the placement of bonds in the local market and
their administration.

108 BIS Papers No 67

benchmarks to complete the yield curve. Most of the issuances have been at the long end of
the yield curve.6 As a result, domestic public debt held by the private sector has risen from
about 1% of GDP in 2004 to almost 9% of GDP at the end of 2011 (Figure 3).
Figure 3
Government domestic and external debt
% GDP

Source: DIPRES.
The market has considered these bonds to be risk-free instruments, similar to those of the
central bank. Turnover has been sizeable, in particular for the nominal bonds or BTP (see
Figure 4).

6 In 2003, the Ministry of Finance issued a BTU with a 20-year maturity. Then, together with the 20-year
maturity bonds, it issued a BTP and a BTU with 10-year maturities. In 2008 and 2010, BTUs of seven- and
30–year maturities were issued. In 2011, a seven-year BTP was issued.

BIS Papers No 67 109

Figure 4
Monthly turnover of Treasury bonds
Times

Source: Central Bank of Chile and Ministry of Finance.
IV. Monetary policy and public debt
The central bank was granted full autonomy in 1989. This means that the institution has its
own legal status, independent of the government. The new constitutional law – Law 18 840
of October 1989 – provides for the central bank’s independence in technical and financial
terms, and defines its objectives as follows: to ensure the currency’s stability and the normal
functioning of domestic and foreign payments. One of the key aspects of the new
constitutional law is that it prohibits any form of financing of the government by the monetary
authority, except in extreme circumstances such as war or national emergency.
In this context, coordination between monetary and fiscal policy is key for the proper
functioning of the economy. One benefit of Chile’s fiscal rule is that it simplifies the task of
policy coordination between the fiscal and monetary authorities. The central bank follows a
flexible inflation targeting approach in conducting its monetary policy so that expected
inflation plays a central role in defining monetary policy. For its part, the fiscal rule provides a
predictable fiscal stance over a medium-term horizon. Thus, in making the projections on
which monetary policy is based, the central bank can take as given the path for public
expenditure that is consistent with the fiscal rule. Another dimension of the coordination
between fiscal and monetary authorities is related to their respective debt policies. The
Ministry of Finance and the central bank annually coordinate their debt issuance for the year,
bearing in mind, among other aims, the need to promote the development of the local
financial market.
To meet its inflation target, the central bank uses the overnight nominal interest rate as its
main policy instrument. It sets a notional level for the monetary policy rate (MPR), and then
adjusts market liquidity to bring the overnight interbank interest rate to around that level. It
offers permanent overnight borrowing and lending liquidity facilities to commercial banks with
a view to keeping the interbank lending rate close to the MPR. The central bank favours a
policy of non-intervention in the interbank market. However, when liquidity pressures cause
the overnight interbank interest rate to move significantly away from the MPR, the central

110 BIS Papers No 67

bank uses a range of instruments to accommodate liquidity. The main tools are traditional
repo operations, where the central bank purchases its own securities with a buy-back clause
for the next working day. These securities are discount promissory notes (PDBC) due in
30 to 360 days, nominal bonds with maturities of two, five and 10 years (BCP2, BCP5 and
BCP10 respectively), and inflation-indexed bonds with maturities of five and 10 years (BCU5
and BCU10).
In addition to traditional repo operations, the central bank regularly performs open market
operations, issuing short-term securities with the aim of supporting the smooth functioning of
the money markets. The issuance of these securities is determined and pre-announced on a
monthly basis. Other mechanisms to provide peso and dollar liquidity are also available.
Some of them were used extensively during the 2008–09 financial crisis (see Annex A).
The issuance of the long-term securities that represent the bulk of central bank liabilities is
done in such a way as to accommodate the increase in demand for monetary base. To
define the size and composition of long-term debt issuance, an annual calendar sets out the
securities falling due in more than one year and money demand increases are estimated.
Hence, the net increase in money demand roughly corresponds to the net value of securities
that are not re-issued. This programme interacts with the short-term security issuance
programme in order to adequately manage liquidity at different maturities. This programme is
pre-announced annually, but the central bank explicitly states that it may be subject to
modification if conditions change.
Naturally, commercial banks account for a large part of demand for debt securities issued by
the central bank. But a significant share of demand also comes from pension funds and
insurance companies, which are typically “buy and hold” investors. This affects the turnover
in secondary markets, so that bid-to-cover ratios vary from around 2.2 for nominal bonds to
2.7 for inflation-indexed bonds.
A long history of inflation in Chile explains the prevalence of indexation in most long-term
issuance. However, since 2000, efforts have been made to issue nominal bonds. One reason
to issue nominal bonds was to provide benchmarks with the aim of deepening the market
(Table 2).
Table 2
Evolution of the debt stock by currency and term

One implication of the large stock of public debt issued by the central bank is that its balance
sheet is weak. As mentioned above, part of the debt was issued to finance the rescue of the
Dec’02 Dec’03 Dec’04 Dec’05 Dec’06 Dec’07 Dec’08 Dec’09 Mar’10 Jun’10 Sep’10 Dec’10 Jun’11
Short debt (under 1 year)
Total CBDN 6,307 4,264 3,457 4,096 3,648 2,178 2,289 7,213 8,122 4,770 6,140 3,355 6,140
Long debt (over 1 year)
CBP 772 3,148 3,711 3,457 3,680 4,090 4,432 3,751 3,921 4,479 4,584 4,543 6,018
CBU 606 2,458 3,276 4,008 4,770 5,203 11,015 9,773 9,414 4,479 7,572 8,209 12,656
Other inflation-linked 15,551 12,504 10,283 8,142 6,447 5,875 3,454 2,749 2,477 2,321 2,161 1,777 1,572
USD 5,868 6,270 5,519 3,288 790 435
Total bonds 22,797 24,380 22,789 18,895 15,687 15,603 18,901 16,273 15,812 11,278 14,317 14,529 20,247
Peso long debt (% ) 3% 13% 16% 18% 23% 26% 23% 23% 25% 40% 32% 31% 30%
Inflation-linked long debt (%) 71% 61% 59% 64% 72% 71% 77% 77% 75% 60% 68% 69% 70%
USD long debt (% ) 26% 26% 24% 17% 5% 3%
TOTAL DEBT CBC 29,103 28,644 26,246 22,992 19,335 17,780 21,189 23,486 23,933 16,048 20,457 17,884 26,387
% Short debt 22% 15% 13% 18% 19% 12% 11% 31% 34% 30% 30% 19% 23%
% Long debt 78% 85% 87% 82% 81% 88% 89% 69% 66% 70% 70% 81% 77%
Peso total (% ) 24% 26% 27% 33% 38% 35% 32% 47% 50% 58% 52% 44% 46%
Inflation-linked total (%) 56% 52% 52% 53% 58% 62% 68% 53% 50% 42% 48% 56% 54%
USD total (%) 20% 22% 21% 14% 4% 2%
Remaining maturity (years)
Short debt 0.35 0.17 0.09 0.10 0.10 0.08 0.24 0.10 0.06 0.13 0.18 0.22 0.08
Long debt 3.66 3.42 3.23 3.47 3.59 3.28 4.76 4.10 3.97 3.88 4.14 4.05 5.48
Total CBC debt 2.94 2.94 2.82 2.86 2.93 3.28 4.27 2.87 2.64 2.77 2.95 3.33 4.22
Calculated with parities $/USD 500 and $/UF21,000.

BIS Papers No 67 111

financial system at the end of the 1990s. But the return on the assets the bank received in
exchange has been on average lower than the return on the bonds issued. In the case of the
debt issued to sterilise the reserve accumulation, the return on international reserves has
been also lower than the debt issued. As a result, the central bank’s net worth has usually
been negative since the mid-1990s. At the end of 2011, the bank’s negative net worth
position was equivalent to an estimated 3.5% of GDP.
However, this has not proved to be a problem for the conduct of monetary policy. The central
bank has successfully issued domestic debt to manage liquidity in order to keep the
monetary policy rate at its target level. Moreover, the debt issued by the central bank is
considered to be a relatively risk-free asset. In fact, the monthly turnover of central bank’s
securities is relatively high, reflecting their liquidity (Figure 5).
The reasons for this are manyfold: first the government – which is perceived by the market
as the guarantor of the central bank – has a strong financial position and is expected to
remain that way in part due to the fiscal rule. Second, the adoption of the pay-as-you-go
principle for the pension system has deepened the financial system and increased the
demand for safe assets, including central bank securities. Third, banking and financial sector
regulation and supervision have improved considerably since the banking reforms in the mid-
1980s. This has helped to keep the risk of financial crises relatively low. Finally, despite its
negative net worth position, the central bank has run a primary surplus due to gains from the
inflation tax, low non-financial costs and the absence of quasi-fiscal commitments. Moreover,
in the long run, thanks to the expansion of the monetary base, it is expected that the central
bank’s net worth will eventually become positive (see Restrepo et al (2009)).7 In sum, the
government’s solvency along with financial institutional developments has allowed the central
bank’s debt to be rolled over without difficulty.
Figure 5
Monthly turnover of central bank bonds
Times

Source: Central Bank of Chile.

7 J Restrepo, L Salomó and R Valdés, “Macroeconomía, Política Monetaria y Patrimonio del Banco Central de
Chile”, Revista Economía Chilena, 12(1), April 2009. This paper’s baseline forecast calls for the central bank’s
net worth to return to the positive domain in about 25 years.

112 BIS Papers No 67

Annex A:
Policy actions taken by the Central Bank of Chile
during the 2008–09 crises.
����� ������� ������������������� ����������������
������������ ������������� ������������������
������������� ������������������ ��� �����
����������� ������������� ��� �����
������������� ������������������ � �������
������������������ ������������� � �������
�����������������������������������
�����������������������������
�����
�������� �����
������������
���� ����������������������������
������� �����
�����������������������������
�������� �����������������������
��������
�������� �����
� �����������
����������������������� ������������
���������������������
�����������������������������
�������� �����������������������
��������
������������������� ���
����
������������
�������������������������
� �����������

����� ������� ������� ���������������� ����
������������ �� ����������
����� �������������������� ��������������
� ���������� ������
��������
�����
� ����������� �� ����������
����� �������������������� ��������������
� ���������� ������
��� ����
��������
�����
� �����������
�����������������
�������
��� ����������� �����
� �����������
�� �������������
����
����� �������������������� ��������������
� ���������� ������
��� �����
��������
�����
������������
�� ��������������
��������
����� ������
������������
�� ��������������
��������
� �������
� �����������
�����������������
�������
� ��������������
������������������������������

BIS Papers No 67 113

Monetary policy, fiscal policy and
public debt management
People’s Bank of China
Abstract
This paper touches on the interaction between monetary policy, fiscal policy and public debt
management. The first part looks at public debt sustainability and monetary policy. When
measuring the fiscal stance, data such as current fiscal income and expenditure, the scale of
public debt, and the coverage of budget revenue and expenditure should be properly
monitored. In addition, factors that could influence the mid-term fiscal stance should be taken
into consideration. Central bank assets may not be used to offset public debt, and pension
funds are not in practice used to offset gross government debt in most economies. To some
extent, oil and commodity stability funds held by resource-abundant economies may be used
to offset gross government debt. Monetary policy is influenced by the financing of the fiscal
deficit by way of bond issuance. The second part is about the development of money
markets, the maturity and yield curves of domestic government bonds, and the deepening of
domestic financial markets and financial stability in China. The concluding part concerns
the central bank and public debt management. There is no need for the central bank in
economies with a well developed treasury bond market to issue debt of its own. In emerging
economies, a regular rollover issuance of central bank debt may help to form a consecutive
short-term risk-free yield curve, serving as a benchmark for pricing in money and bond
markets. Central banks involved in public debt management need to coordinate closely with
the debt management agency on the policy objectives for various macro control instruments
associated with fiscal policy, monetary policy and public debt management.

Keywords: Monetary policy, fiscal policy, public debt management, China
JEL classification: E52, H63

114 BIS Papers No 67

I. Public debt sustainability and monetary policy
(1) When measuring the fiscal stance, the following factors should be taken into
consideration:
• The following data should be monitored:
1. Current fiscal income and expenditure. In recent years, China’s deficit has
remained low, compared with those of the major developed economies. In 2009 and
2010, the fiscal deficit amounted to RMB 950 billion and RMB 1050 billion
respectively, slightly less than 3% of GDP.
2. Public debt. By the end of 2010, China’s outstanding government debt stood at
around RMB 7 trillion, or 20% of the GDP, a relatively safe level. However, some
local government liabilities are not included in the statistics and thus need to be
monitored closely as to their scale and potential impact on China’s fiscal condition.
3. Coverage of budget revenue and expenditure. In many economies for
statistical reasons, not all public revenues and expenditures are included in the
national budget. The revenues and outgoings of local governments and state-owned
enterprises are a case in point. These should be taken fully into account when
measuring the fiscal stance. China’s government budget covers the revenue and
expenditure of both the central and local governments but not all such items. Not
included are some special funds (eg pension funds and land transfer funds etc), as
well as the profits and losses of state-owned enterprises.
• Factors that could influence the mid-term fiscal stance, especially those that could
increase contingent liabilities.
Among these factors is the ageing problem. It is estimated that, by 2030, China’s
elderly population will reach twice today’s level. This calls for more fiscal input into
related pension and healthcare services. This ageing trend will also offset the
“demographic dividend” effect and thus indirectly influence future fiscal income.
Moreover, the potential losses of government-led construction projects take a long
time to show up in the balance sheet. This may increase government’s mid-term
contingent liabilities to a certain extent.
• Characteristics of emerging market economies.
Compared with developed countries, emerging markets enjoy faster growth but face
higher inflation and more volatility, which may have significant negative implications
for their mid-term fiscal stance.
(2) Public sector assets as offset to gross debt
Central bank assets cannot be used to offset public debts. Central bank assets are acquired
by injecting base money, meaning that any increase in the central bank’s assets implies a
simultaneous and equal increase in its liabilities. Nor is it the practice in most economies for
pension funds to be used to offset gross government debts. In reality, a large pension
funding gap is a problem for most developed and emerging markets. Pension funds cannot
be used as an offset to gross government debt, and the funding gap may even increase the
government’s contingent liabilities. To some extent, oil and commodity stability funds
held by resource-abundant economies may be used to offset gross government debt.
However, authorities should take a prudent approach if they choose to follow that course,
and they should take into account that such funds may become contingent liabilities and
could cause maturity and structure mismatches in the balance sheet. Thus, these funds
should be duly discounted when used as offsets.

BIS Papers No 67 115

(3) The influence of fiscal deficit financing on monetary policy
Governments usually finance their deficits by levying taxes or issuing bonds. Taxation has
basically no effect on monetary policy, whereas bond issuance has either a direct or indirect
influence on monetary policy. Excessive government bond issuance will worsen the
government’s future fiscal condition, which could result in the central bank being forced to
acquire government bonds and thus compromise the independence of monetary policy.
II. Domestic currency-denominated public debt in China’s domestic
market
(1) Money market development
The volume of interbank lending in China has steadily and rapidly increased since the
beginning of this year. From January to October, interbank lending totalled RMB 26.6 trillion,
up 18.3% year on year. From January to October, the total borrowing of the state-owned
commercial banks — the main borrowers in the interbank market, reached RMB 16.2 trillion,
up 35.3%, accounting for 60.9% of the total borrowing in the market. On the other hand, the
total lending of the main lenders, namely joint-stock commercial banks, city commercial
banks and state-owned commercial banks, registered RMB 19.4 trillion, up 2.20%,
accounting for 72.9% of the total lending in the market. The lending of foreign-funded
financial institutions totalled RMB 2.80 trillion, up 68.30%, and the lending of other financial
institutions reached RMB 4.4 trillion, up 99.20%.
(2) Maturity and yield curves of domestic government bonds
The Chinese government started to issue treasury bonds with maturities of more than
10 years after the Asian financial crisis in 1997. Maturities of 15 years and 20 years were
issued in 2001, of 30 years in 2002 and of 50 years in 2009. At the end of October 2011,
148 batches of bonds were outstanding in the interbank bond market, with a volume of
RMB 5.84 trillion and maturities ranging from three months to 50 years.
The People’s Bank of China has issued a series of rules and guidelines since 2001 on
standardising yield calculations in the interbank bond market. In June 2011, the People’s
Bank of China and the Ministry of Finance issued a joint notice, with detailed guidelines for
market-makers on market-making for treasury bonds with key maturities. This has improved
liquidity as well as the quality of treasury bond quotations, and it has further extended the
yield curve.
(3) Deepening domestic financial markets and financial stability
In recent years, China’s bond market has continued to expand, with enhanced market
liquidity and more foreign investors involved. The deepening of the bond market has
contributed to the stability of China’s financial system. First, by accelerating the
transformation from indirect financing to direct financing, thus diversifying the risks in the
banking system and lowering financing costs in the real economy. Second, financial
institutions have been provided with tools to replenish Tier 2 capital and the resilience of
financial institutions has been improved. Third, market participants have adopted investment
and risk management tools, so that the risks faced by credit entities have been diversified.
Fourth, with a diversified base of bond market investors and increased bond trading volumes,
the bond pricing mechanism has become more market-oriented, which has improved capital
allocation and promoted financial market stability.

116 BIS Papers No 67

III. Central bank and public debt management
(1) Whether the central bank can issue its own debt paper, and its coordination
with the Ministry of Finance
Whether a central bank can issue its own debt paper depends on the development of
domestic treasury bond market and its monetary management policies. In economies with
well developed treasury bond markets, the issuance of the bonds may perform two functions:
as a financing tool for the central government, and as an instrument for regulating and
managing financial markets. There is no need for the central banks in these economies to
issue debt paper of their own. However, in emerging economies with less developed bond
markets, bond issuance serves solely as a financing tool for government activities. In this
case, long-term bonds account for most of the total volume, whereas short-term bond
issuance is inadequate. In these economies, a regular rollover issuance of central bank debt
may help form a consecutive short-term risk-free yield curve, serving as a benchmark for the
pricing in the money market and bond market. Moreover, in economies with excess liquidity,
central bank bond issuance can effectively absorb any excess short-term liquidity in the
banking sector, and create a favourable environment for economic growth and price stability.
In this case, central bank debt issuance could itself be regarded as an important measure in
strengthening the coordination between fiscal and monetary policies.
(2) Involvement of central banks in public debt management
Central banks may involve themselves in public debt management based on each country’s
specific circumstances. Central banks tend to engage more deeply in the management of
short-term rather than long-term debt, because short-term debt is more useful as a policy
instrument for adjusting the base money supply. Central banks can adjust market liquidity
and the money market interest rate by purchasing short-term government bonds via open
market operations or by issuing central bank bills. Sound communication and cooperation
between the ministry of finance and the central bank should be maintained in managing
foreign currency-denominated debt. If financial resources raised by issuing foreign currency-
denominated bond are invested domestically, the debt-service risk arising from exchange
rate movements should be fully taken into account. In this case, government plans for
managing its foreign debt could have significant implications for the coordination of local and
foreign currency policy by central banks.
Quasi-fiscal operations and unconventional monetary policy will expand central bank balance
sheets and are likely to reduce, to some extent, the quality of central banks’ assets and
therefore their profits. Apart from direct financial losses, the excessive issuance of money as
a consequence of quasi-fiscal operations is likely to drive up market expectations for inflation
and therefore undermine the credibility of central banks in maintaining the value of the
currency. In addition, the independence of central banks will be undermined if they take on
responsibility for fiscal functions. As a consequence, central banks should attach great
importance to the hidden threat of inflation, whether or not quasi-fiscal operations or
unconventional monetary policy will directly affect their financial condition.
(3) Institutional arrangements for coordinating monetary policy and public debt
management
First of all, the debt management agency and the central bank should reach a consensus on
the policy objectives of the various macro control instruments associated with fiscal policy,
monetary policy and public debt management. Principles should be established for public
debt management that reinforce the coordination between public debt management and
monetary policy. Second, choices about the scale, variety and pace of public debt issuance
should be made with great consistency, and they should be fully coordinated with the central

BIS Papers No 67 117

bank so that the latter can carry out liquidity management effectively. Third, central banks
and governments should step up their efforts to improve the bond market environment in
order to enhance the liquidity of secondary markets for government bonds. This will provide a
solid foundation for the formation of a reliable yield curve for risk-free market interest rates,
thus enhancing the efficiency of the monetary policy transmission mechanism. Fourth, public
debt issuance should not undermine the implementation of monetary policy and the stability
of the currency’s value. And if the government launches a plan for large-scale debt issuance
to combat an economic and financial crisis, a fiscal consolidation programme should also be
in place with a view to protecting the central bank’s independence.

BIS Papers No 67 119

Macroeconomic effects of structural fiscal
policy changes in Colombia
Hernando Vargas, Andrés González and Ignacio Lozano1
Abstract
In the past decade the Colombian economic authorities undertook a series of measures that
reduced the structural fiscal deficit, decreased the government currency mismatch and
deepened the local fixed-rate public bond market. This paper presents some evidence
suggesting that these improvements had important effects on the behavior of the
macroeconomy. They seem to have permanently reduced the sovereign risk premium,
increased the reaction of output to government expenditure shocks and strengthened the
response of market interest rates to monetary policy shocks.

Keywords: Monetary policy, fiscal policy changes, public debt management, government
expenditure, market interest rates, monetary policy shocks
JEL classification: E44, E6, E62

1 Deputy Governor, Director of Macroeconomic Modeling and Senior Researcher of Banco de la República
(Central Bank of Colombia), respectively. The authors are grateful to Juan P. Zárate, José D. Uribe, Jorge
Ramos and Franz Hamman for helpful comments, and to Pamela Cardozo, Juan Manuel Julio, Karen Leiton,
Enrique Montes, José D. Pulido and Sebastián Rojas for useful suggestions about some variables and
indicators used in this paper.

120 BIS Papers No 67

1. Introduction
Over the last decade the Colombian government and congress undertook a series of
measures and reforms that significantly shifted the trend of public debt, reduced the financial
fragility of the government and deepened the domestic public bond market. First, starting
from a rising, unsustainable debt path, several structural fiscal reforms were instrumental in
the decline of the public debt to GDP ratio between 2003 and 2008, and its more recent
stability. Second, an explicit policy of reducing the currency mismatch of public finances
decreased their vulnerability in the face of a sharp depreciation following an adverse external
shock. Third, there has been an effort to shift the composition of public debt toward fixed-
rate, peso denominated bonds and to lengthen its maturity.
One would expect that this set of prudent policies had important effects on the behavior of
the macroeconomy both in the long term and in response to exogenous shocks. After briefly
highlighting some aspects of fiscal policy and public debt management in the past ten years,
this paper assesses some of those effects. Specifically, the influence of fiscal policy changes
on the country’s sovereign risk premium, the short-run response of output to a fiscal shock
and the transmission of monetary policy shocks to market interest rates are evaluated.
2. Fiscal policy in Colombia
The adoption of a new constitution in 1991 entailed a strong expansion of the size of
government in Colombia. Increased demand for public spending in health, education and
justice drove central government primary expenditure from 7.2% of GDP in 1990 to 12.4% of
GDP in 2000. At the same time, the Constitution of 1991 and the law extended fiscal
decentralization and imposed a regime in which an increasing fraction of central government
current revenues was transferred to local governments. The tax increases adopted to pay for
the additional expenditure were not sufficient and had to be shared with local governments,
which, in turn, increased their spending. In addition, the intertemporal solvency of the pay-as-
you-go national pensions system was in doubt, given its prevailing parameters and the
co-existence of a defined-contribution private pension fund system.
By the end of the nineties, fiscal sustainability in Colombia was uncertain. The central
government debt to GDP ratio was rising fast and several local governments were over-
indebted. The external shocks of that period (especially the Russian crisis) triggered the
largest output drop in Colombia since the Great Depression and a financial crisis. The cost of
the latter had to be absorbed by the government, thus worsening an already weak fiscal
situation.
Starting in the early 2000s, an adjustment had to be implemented that included four tax
reforms, two reforms to the transfers to sub-national governments and other measures that
substantially reduced the non-financial public sector (NFPS) deficit from 4.9% of GDP in
1999 to a balanced position in 2008. During this period, the deficit of the central government
was reduced from 6% to 2.3% of GDP, while the remaining NFPS recorded surplus
balances. As a result, the central government debt to GDP ratio declined throughout the
2000s and has been stable in recent years (Graph 1). Moreover, a reform to the general
pension regime in 2003 made progress toward ensuring the sustainability of the pay-as-you-
go system.
Since 2003, Colombia has been implementing its fiscal policy through a qualitative rule:
Law 819 on transparency and fiscal responsibility. Under this mandate, the central
government must prepare a Medium Term Fiscal Framework (Marco Fiscal de Mediano
Plazo, MFMP) every year as its main tool for financial programming. The MFMP sets a
numerical target for the primary balance of the NFPS for the following year as well as some
indicative targets for the subsequent ten years, so that public indebtedness remains in line

BIS Papers No 67 121

with a sustainable path. Among other aspects, the MFMP includes an assessment of the
contingent liabilities of the public sector, the cost of tax benefits, and some sections on the
fiscal programming of sub-national governments. Fiscal forecasts are made based on
macroeconomic assumptions jointly formulated by the Ministry of Finance, the central bank
and the National Planning Department.
Even though the MFMP is a valuable tool for fiscal stance programming, it has some
constraints from a macroeconomic perspective. On the one hand, the multi-annual primary
balance targets are adjusted repeatedly for diverse reasons, thus lessening the initial
commitments of the government. On the other hand, it does not assess explicitly the effects
of the business cycle on tax revenues and expenditures, which increases the risk of
procyclicality in fiscal policy. In fact, some studies have found some evidence of procyclicality
of fiscal policy in Colombia and other emerging economies (Cárdenas et al., 2006, Lozano,
2011 and Ilzetzki and Vegh, 2008).
To overcome the MFMP limitations, Law 1473, by which the central government adopted a
quantitative fiscal rule, was passed in mid-2011. In addition to ensuring the sustainability of
public debt and promoting a countercyclical fiscal policy stance, it is expected to alleviate the
effects of exchange rate volatility on the economy’s tradable activities, for it should foster
better management of the resources generated by the mining and energy sectors.
Furthermore, the framework of fiscal policy in Colombia was supplemented with a royalty law
for the exploitation of natural resources, approved in 2011. This law aims at distributing
royalty funds more equitably among the country’s several regions and at saving their
transitory component.
3. Public debt management in Colombia
Along with fiscal consolidation, in the last decade the Colombian authorities have sought to
improve the composition of public debt in order to reduce the financial fragility of the
government and to encourage the development of capital markets in the country. To that
end, steps were taken to decrease the currency mismatch of the public sector, by shifting the
composition of its debt from foreign currency denominated bonds and loans (mostly external
debt) toward local currency denominated bonds (mostly internally issued). As a result, a
substantial drop in a currency mismatch indicator was achieved for the central government
(Graph 2)2.
In turn, an effort has been made to change the composition of domestic debt from inflation or
dollar indexed bonds toward fixed-rate peso denominated bonds (Graph 3). This process
began in the late nineties with the inception of a market makers program, but was greatly
enhanced by fiscal consolidation, the achievement of single digit inflation and a consistent
convergence toward the long term inflation target (3%) in the 2000s. In September 2011 the
stock of local, fixed-interest, peso denominated bonds (TES) accounted for 51.4% of total
central government debt and represented 18.3% of GDP.
Besides increasing the participation of these instruments in total debt, government policy has
successfully extended the maturity of the new issues throughout the last decade (Graph 4), a

2 The indicator, inspired by Goldstein and Turner (2004) and Rojas-Suárez and Montoro (2011), attempts to
capture the ability of the central government to serve its foreign currency-linked debt on the basis of its foreign
currency-linked revenues. It is constructed as the ratio: (FCD/TD) / (FCR/TR) for the central government.
FCD = Foreign currency debt. TD = Total debt. FCR = Foreign currency-linked revenue, which includes
external VAT, import tariffs, Ecopetrol (the state oil company) dividends, income taxes paid by mining
companies and other exporting firms, and income derived from external assets. TR = Total revenue. Data
sources: Banco de la República, DANE, DIAN, Ecopetrol, Supersociedades and Hamann et al. (2011).

122 BIS Papers No 67

sign of credibility in both fiscal and monetary policy (Hamann and González, 2011). The
share of the outstanding stock of bonds with less than one year residual maturity has
declined in the past ten years in favor of issues with maturity greater than five years, while
the share of issues with residual maturity between three and five years has remained stable
(Graph 5). Today the longest maturity in the TES market is fifteen years. This was important
for the development of a fixed-rate mortgage loan market in the 2000s (Galindo and
Hofstetter, 2008, and Hamann et al., 2010), and may have influenced the transmission of
monetary policy shocks to other financial system interest rates, as will be discussed below.
4. The macroeconomic effects of the fiscal policy changes
The aforementioned improvements in fiscal and public debt management policy were large
enough to have an impact on the behavior of the macroeconomy both in the long term and in
response to exogenous shocks. This section explores some of those effects.
a. Effects on the sovereign risk premium
Among the most important goals of the structural adjustment process undertaken since the
early 2000s were ensuring the sustainability of the public debt and strengthening the
resilience of the economy in the face of external shocks. Specifically, the correction of
structural imbalances and the shift in the trend of the public debt to GDP ratio reduced the
probability of default of the Colombian government and the government’s vulnerability to
shocks impacting its revenues and expenses. Further, the fall of its currency mismatch
reinforced the ability of the government to withstand a depreciation shock. At a more
aggregate level, the decline in the government currency mismatch was part of a general
trend that also included the private sector and allowed greater scope for exchange rate
flexibility and the possibility of a countercyclical monetary policy response to external shocks.
This, in turn, moderated the effect of those shocks on output and fiscal revenues.
Overall, the reduction in the public debt to GDP ratio and government currency mismatch
decreased the credit risk of the government and the country. Hence, they contributed to a
permanent drop in the sovereign risk premium and to a decline in its sensitivity to global risk
aversion shocks.
To test the first implication, we estimated a model for the Colombian sovereign risk premium,
measured by the EMBI Colombia, based on the following specification:
embict = a o + a 1 grat + a 2 (d/y)t + a 3 cmt + et
embic is the EMBI Colombia, gra is a measure of global risk aversion, d/y is the central
government debt to GDP ratio and cm is the currency mismatch indicator calculated above.
As measures of global risk aversion, the VIX and the 5-year high yield spread were used. All
variables were expressed in logs and were non-stationary in the sample 1999.Q2-2011.Q4
(quarterly data). Cointegration was found for these systems based on the Hansen test
(Hansen, 1992).
The long run relationships presented in Table 1 confirm the importance of local fiscal
variables in the determination of the EMBI Colombia, beyond the effect of global risk
aversion. In both specifications (with the VIX and the high yield spread as measures of global
risk aversion) the government currency mismatch appears significant and with the expected
positive sign. The debt to GDP ratio is also significant and with the expected positive sign in
the specification that uses the VIX as the global risk aversion variable (Table 1, upper panel).

BIS Papers No 67 123

It is positive, but not significant in the specification that includes the high yield spread as the
measure of global risk aversion (Table 1, lower panel)3.
The second implication, changing sensitivity of the sovereign risk premium to global risk
aversion as a result of improved fiscal policy, is tested by Julio et al. (2012). Following
Favero and Giavazzi (2004), these authors estimate a model in which the response of the
EMBI Colombia to the spread between US BAA corporate bonds and 10-year US Treasury
Bonds depends on the difference between the observed government primary surplus and the
value of the primary surplus that would stabilize the debt to GDP ratio at each point in time.
They posit a non-linear relationship in which large observed primary surpluses relative to
their debt ratio-stabilizing values drive the sensitivity of the EMBI Colombia to global risk
aversion toward zero, while the opposite situation increases that sensitivity.
Working on a monthly sample between 1998 and 2010, Julio et al. (2012) find that the
sensitivity of the EMBI Colombia to their measure of global risk aversion does depend
significantly on their fiscal health indicator. Furthermore, they find a structural break in the
sensitivity function around mid-2006. After this period, there seems to be a substantial
reduction of the sensitivity function, which the authors associate both with a permanent
improvement in the Colombian fiscal health indicators and with the deterioration of public
debt ratios in advanced economies.
In sum, the evidence presented in this section and in Julio et al. (2012) supports the
hypothesis that the abovementioned improvements in fiscal policy and public debt
management did permanently reduce the sovereign risk premium in Colombia and its
sensitivity to global risk aversion shocks. The macroeconomic implications of this result are
important.
First, it means that, ceteris paribus, the long term level of the real interest rate is lower today
than a decade ago4. Based on the long run relationship presented in Table 1 (upper panel),
on average, local factors (the decline in the government currency mismatches and the debt
to GDP ratio) would imply roughly a 60% decrease in the EMBI Colombia between
2002.Q1-2006.Q4 and 2007.Q1-2011.Q45.
Also, a permanent decrease in the risk premium entails a permanent adjustment in the long
run level of the real exchange rate. Hence, it could be argued that part of the real
appreciation of the Colombian peso in the past decade could be attributed to better fiscal
policy. The permanent movement of the long run level of both the real interest rate and the
real exchange rate has important consequences for the design and operation of monetary
policy. It implies that the mean value of the natural interest rate is lower than ten years ago
and that indicators of trend real exchange rates that give large weights to values from the
early 2000s are probably biased.
Second, the empirical results suggest that the economy is generally less vulnerable to global
risk aversion shocks because of the reduced sensitivity of the risk premium to them. This
implies lower responses of the exchange rate and capital flows to those shocks, and,
consequently, lower pressure on inflation, output and monetary policy.

3 Other factors that promote higher rates of long term growth may have also reduced the sovereign risk
premium. In Colombia there were improvements in security, findings of large mineral and oil reserves, specific
policies aimed at fostering investment and a permanent decrease in inflation throughout the decade.
4 Interestingly, the external real interest rate decreased in the same period, reinforcing the effect of a lower
sovereign risk premium on domestic real interest rates. Also, the reduction in inflation volatility may have
contributed to a decline in domestic long term real interest rates through smaller inflation risk premia.
5 We computed the changes in the logarithm of the average government currency mismatch indicator and the
debt to GDP ratio between 2002.Q1-2006.Q4 and 2007.Q1-2011.Q4, and multiplied them by the
corresponding elasticities from Table 1. We then added the calculated impacts.

124 BIS Papers No 67

b. Effects on the short-run response of output to government expenditure
shocks
It is likely that the perception of households, firms and investors about the sustainability of
the public debt and the financial fragility of the government influences their reaction to fiscal
policy shocks. An unexpected increase in public expenditure may prompt an expectation of
higher taxes in the short run in a dire financial situation of the government, thereby offsetting
its possibly expansionary effect on output. Moreover, a similar shock in a small, open
economy may sharply raise the sovereign risk premium, bringing about a tightening response
of the monetary authority to curb currency depreciation and inflation, or a contraction of
external finance and credit (Ilzetzki et al., 2009). When public debt sustainability is more
certain or government currency or liquidity mismatches are low, the expansionary effects of a
public expenditure shock may be greater.
To explore this hypothesis, the empirical strategy must carefully consider the problems of
identification of a fiscal shock (finding the movement of fiscal variables that are not
contemporaneous responses to output) and the anticipation of fiscal policy by the private
sector. The first issue is crucial to avoid a bias in the estimation of the response of output to
an exogenous fiscal shock and requires isolating the part of the movement in the fiscal
variables that are purely discretionary, non-output related changes. The second issue is
important because an anticipated fiscal policy shift may induce an anticipated response by
the private sector consumption or output, so that the estimated response after the realization
of the shift could be biased (Perotti, 2007).
SVAR models have been widely used in the literature to identify fiscal shocks6. Another
technique, the so called “narrative approach”, uses dummy variables to measure the effects
of fiscal policy shocks that are not related to movements of output (e.g. wars, “ideological”
policy shifts, output-independent cross sectional effects, etc.)7. In Colombia, SVAR models
used to estimate the effect of fiscal policy shocks on output have rendered results that range
from negligible impacts (Restrepo and Rincón, 2006) to positive expenditure multipliers
between 1.1 and 1.2 (Lozano and Rodríguez, 2011). However, these studies include a
relatively long sub-period in which the exchange rate was not as flexible as after 1999
(crawling peg or target zone regimes). Consequently, their estimated impacts may be
affected by a structural break related to the adoption of a floating exchange rate regime8.
Our approach differs from the previous work in three important dimensions. First, our sample
covers only the floating exchange rate period (1999-2011). Second, we are interested in
capturing a possibly changing effect of public expenditure shocks, as fiscal policy became
sounder throughout the 2000s. This implies the use of a non-linear technique that allows for
a smooth transition between regimes that are defined according to indicators of fiscal health.
Third, since we do not estimate a SVAR, we identify the government expenditure shock
based on innovations on the public spending announcements for the central government9.

6 See, for example, Blanchard and Perotti (2002) for the US, and Perotti (2004), and Caldara and Kamps (2008)
for the OECD countries.
7 See Perotti (2007) and Romer (2011).
8 Standard Mundell-Fleming theory suggests that the exchange rate regime makes a difference regarding the
effect of fiscal policy shocks in a small, open economy. See Ilzetzki et al. (2009) for some evidence about the
differences of output responses to fiscal shocks in economies with flexible and pegged exchange rates.
9 We do not study the effects of tax shocks due to the difficulties involved in their identification and the problems
derived from the sensitivity of the theoretical results to the time profile of distortionary tax responses (Perotti,
2007).

BIS Papers No 67 125

Following Auerbach and Gorodnichenko (2012), instead of estimating a SVAR and deriving
standard impulse response functions, we approximate the non-linear impulse response
function by the following linear projection:
Yt+h = G(zt) ( Y 1
h Ft + L 1(L) Yt-1) + (1-G(zt)) ( Y 2
h Ft + L 2(L) Yt-1) + et
The impulse response function of output (Yt+h) to an unexpected government expenditure
shock (Ft) is estimated directly by G(zt) Y 1h + (1-G(zt)) Y 2h , where Y 1h and Y 2h are estimated
by least squares (for details see Jordà, 2005).
Notice that the impulse response function depends on the value of the variable zt. In our
case, zt is a fiscal health indicator. At a given point in time the impulse response function may
be understood as a combination or “average” of the functions corresponding to the extreme
states of the fiscal health indicators (e.g. “High Debt” vs. “Low Debt”, or “High Currency
Mismatch” vs. “Low Currency Mismatch”). The weight of each extreme state will be given by
the transition function G(zt) = e-gzt/(1+e-gzt), which measures how close the fiscal health
indicator of the moment is to one extreme state or to the other.
The above technique requires the definition of an exogenous government spending shock,
Ft, outside the model that meets the criteria of no anticipation and no contemporaneous
correlation with output. To do so, we define the shock as the difference between the actual
central government primary expenditures (overall spending without interest payments on
public debt) and the forecast made of this variable10. For the OECD countries, these
predictions are typically taken from professional forecasting surveys. Since this type of
information is not available for Colombia, we derived it from the Ministry of Finance’s
announced Financial Plans, as explained in Appendix 1. The fiscal shocks thus computed
are not anticipated by construction, nor are they correlated with current output because of the
lag with which output and other real activity data are available, and the lag with which
expenditure decisions are executed11.
As fiscal health variables, zt, we used the central government debt to GDP ratio, the
government currency mismatch and the difference between the observed government
primary surplus and the value of the primary surplus that would stabilize the debt to GDP
ratio at each point in time (Graph 6)12. The impulse response functions of output to a
government expenditure shock are estimated using quarterly data for the 1999-2011 sample.
The results in Graphs 7 and 8 suggest that there were important changes in the response of
output to the fiscal shock throughout the decade, as fiscal health indicators improved
markedly13. The responses in the beginning of the decade were, when positive, small and
short-lived; in other cases, they were negative on impact and non-significant afterwards.
When the debt to GDP ratio stopped rising or the primary surplus deviation from its debt-
stabilizing level increased (2002-2003), output responses turned positive and remained
significantly different from zero for several periods. Interestingly, the positive reactions seem

10 Due to data availability, we use central government primary expenditure, which corresponds roughly to two
thirds of total general government primary expenditure.
11 A potential drawback of our measure of expenditure shock is that we cannot separate public consumption,
investment, transfers and subsidies expenses, since the government Financial Plans do not disaggregate the
outlays in these categories. We are then capturing the effects of a shock to aggregate central government
expenditure. This may be a problem if the macroeconomic effects of public consumption, investment and
transfer shocks are very different, and if the composition of the aggregate shocks changes significantly from
year to year.
12 See Julio et al. (2012) for details on the construction of this series.
13 The technique used allows us to estimate the impulse-response functions with confidence intervals for each
quarter in the sample. The results presented in Graphs 8 to 10 correspond to the average responses for each
year with the confidence interval calculated appropriately. We used four lags of GDP in the estimation.

126 BIS Papers No 67

to be clearer and larger when the primary surplus is higher (2007-2008) (Graph 8), although
in no case the estimated conditional government expenditure multipliers exceed one.
Similarly, the output responses related to low government currency mismatches (2005-2011)
were in general significantly positive for several quarters, unlike the responses observed in
years of high currency mismatches (1999-2004) (Graph 9)14.
Hence, the stronger the financial position of the government, the greater the power of fiscal
(expenditure) policy to affect output. The implication of this result for the assessment of the
convenience of countercyclical fiscal policy is apparent; i.e., a sound public finance situation
not only has benefits in terms of permanently lower real interest rates and lower vulnerability
of the economy to global risk aversion shocks, but also seems to enhance the effectiveness
of countercyclical fiscal policy.
c. Effects on the transmission of monetary policy shocks to market interest rates
As the fiscal situation improved structurally and monetary policy gained credibility throughout
the 2000s (Hamann and González, 2011), the transmission of monetary policy shifts to
financial market interest rates may have been strengthened. To begin, under a more credible
monetary policy regime, a movement in overnight policy rates is more likely to be
incorporated in longer term public bonds and financial system interest rates because the
policy change will most probably be perceived by market participants as a persistent signal
on the policy stance, instead of a noisy policy error to be undone in the near future.
Furthermore, as mentioned above, the enhanced credibility of a low and stable inflation rate
as well as a stronger perception of public debt sustainability permitted the extension of the
maturity of fixed-rate public bonds. Consequently, the depth and liquidity of longer term
public bond markets may have been increased, thereby making their prices a better guide for
interest rate setters in the financial system and allowing them to better filter the news from a
monetary policy shock.
To explore the relevance of these hypotheses, we use the same non-linear model from the
previous section to test whether the transmission of monetary policy shocks to public bond
interest rates (TES) and deposit or loan rates changed as the maturity of the government
fixed income market was expanded throughout the 2000s. Specifically, we estimate the
following monthly models for TES and market interest rates:
itest+h = H(zt) ( P 1
h Mt +G1(L) itest-1 +S r t ) + (1-H(zt)) ( P 2
h Mt + G2(L) itest-1 S r t ) + et
imt+h = J(zt) ( F 1
h Mt + B1(L) imt-1+ K1(L) itest-1) + (1-J(zt)) ( F 2
h Mt + B2(L) imt-1+ K2(L) itest-1) + et
The response of TES rates, itest+h, to an unanticipated monetary shock, Mt, is approximated
directly by H(zt) P 1h + (1-H(zt)) P 2h in a linear projection estimated by least squares (Jordà,
2005)15. Notice that this response is allowed to change as a function of the maturity of the
new issues of fixed-rate TES (zt = long term component of the average maturity of new
issues) (Graph 4). A similar model is estimated for the response of market (deposit or loan)

14 When interpreting the impulse-response functions presented in Graphs 8 to 10, it must be recalled that they
are conditional on the state of the fiscal variable used to define the regime. For example, in 2004 the
responses of output to the fiscal shock were generally positive when the fiscal variable regime was measured
by the difference between the primary surplus and its debt-stabilizing level, but essentially zero when the fiscal
variable regime was measured by the government currency mismatch. This means that the response of output
conditional on the surplus variable of that year was significantly positive, but the response conditional on the
currency mismatch observed in the same year was non-significant. Overall, it may be concluded that the
probability of a positive impact of a fiscal shock on output increased in 2004 with respect to previous years in
which all conditional responses were non-significant, but was smaller than in later years, when all conditional
responses were statistically positive.
15 The equation for the TES rates controls for the influence of the EMBI Colombia, ȡt.

BIS Papers No 67 127

interest rates, imt+h, to an unanticipated monetary shock, Mt, but the controls include lagged
values of both market and TES rates with similar maturities.
The definition of monetary shock is crucial to minimize the bias of the estimated impulse
response functions. If a change in the policy interest rate is anticipated by market
participants, then it would be incorporated in longer term TES or financial system interest
rates before it happens. When the change occurs, the reaction of longer interest rates will be
null, leading to an estimated negligible transmission of monetary policy. Therefore, the
estimated monetary policy shock must be unanticipated and, so, orthogonal to all information
that might be relevant to predict the policy rate at each point in time. Appendix 2 provides
some details on the estimation of the monetary policy shock that is used in our estimations.
The results for the transmission of policy rates to TES interest rates are shown in Graphs 10
to 1316. There seem to be two clearly different regimes: one between 2002 and 2003, the
other between 2005 and 2011, and a transition year in 2004. Between 2002 and 2003 there
were negative monetary shocks (Graph 29), meaning that the market expected policy rate
increases that did not happen. According to Graphs 10 to 13, 0-5 year TES rates increased
and the zero coupon curve steepened up to the sixth month after the shock. TES rates for
maturities greater than five years slightly declined on impact, but rose sharply afterwards17. In
contrast, between 2005 and 2011, the monetary shock took both positive and negative
values and its volatility was substantially smaller (Graph 29). In this period all TES rates rose
with a positive monetary shock, while the zero coupon curve generally flattened afterwards,
as can be seen by comparing the impacts across time and maturity.
A possible interpretation of these results is that the monetary policy response to the risk
aversion shock, the peso depreciation and rising core inflation observed between 2002 and
200318 was deemed insufficient by the market, so it was judged as a policy mistake that
would require a correction over the short term (hence the response of the 0-3 year bond
prices) or would risk a future rise of inflation (hence the response of the bonds with maturity
greater than 3 years). Alternatively, there may be omitted variables that account for the
negative response of the TES rates to the monetary policy shock, even though the
econometric model controls for the effects of the contemporaneous sovereign risk premium
shock19. After 2004 monetary policy shocks are smaller and the curve seems to shift upward
and flatten after a positive shock, a plausible sign of greater credibility of monetary policy20.
With respect to the transmission of monetary policy shocks to market interest rates, there is
also evidence of a structural change linked to the average maturity of new issues of TES.
The main findings in this regard may be summarized as follows:
• For all loan and deposit rates considered there are two regimes: In the first, between
2002 and 2003, a positive monetary shock produces non-significant or, in few

16 The technique used allows us to estimate the impulse-response functions with confidence intervals for each
month in the sample. The results presented in Graphs 11 to 14 correspond to the average responses for each
year with the confidence interval calculated as before. We used one lag of TES rates in the estimation.
17 Given the units of the TES rates and the monetary shock, an impulse-response value of 100 corresponds to a
one-to-one transmission of the monetary shock.
18 Following a sharp increase in the EMBI the second semester of 2002, the COP depreciated by 23.3%
between June 2002 and March 2003, while annual CPI (without food) inflation rose from 5.5% on average in
the first semester of 2002 to 6.6% on average in the first semester of 2003.
19 In particular, during those years there was a strong disturbance in the TES market after a sovereign risk
aversion shock because banks cut funding to brokers that had leveraged to invest in these securities. It is
possible then that, due to fire-sales of TES, their prices fell beyond what could be explained by fundamentals.
20 This response implies that the monetary surprise is expected to persist and is therefore transmitted to longer
rates (i.e. is not considered a policy mistake).

128 BIS Papers No 67

cases, negative responses of market rates. In the other, between 2005 and 2011,
there are generally positive, significant responses of market rates to a monetary
shock. As in the case of the TES rate responses, 2004 seems to have been a
transition year (Graphs 10-27).
• The response of commercial loan rates after 2004 is monotonically increasing,
reaching values that indicate a reaction greater than one-to-one after one year. This
contrasts with the responses of the TES rates at similar maturities and suggests that
corporate credit risk premia may rise after a positive monetary shock.
• The response of consumer loan rates with maturity less than one year after 2004 is
initially negative, but positive six months after the monetary shock and less than
one-to-one. For longer maturities, the response is very small for the first five or six
months after the shock, but increases afterwards, reaching values that indicate a
reaction greater than one-to-one after one year.
• Deposit (CD) interest rates with maturities less than one year increase with the
monetary shock, reaching values that indicate a reaction close to one-to-one. CD
interest rates with maturity greater than one year show a response larger than one-
to-one after one year.
The contrast between the responses before and after 2004 may be a sign of rising credibility
of monetary policy throughout the decade, as in the case of the TES rate responses. The
lengthening of the maturity of TES could serve as a proxy for this increased credibility.
However, it is indicative that, unlike the TES rate reaction in 2002-2003, several market rates
did not display a negative, significant response to the monetary shock in the same years.
Thus, other phenomena could have influenced the estimated change in the transmission.
The extension of the maturity of new TES issues and the TES stock may have enhanced to
role of the public debt market in the determination of financial system interest rates, by
providing liquid, reliable “risk-free” benchmarks at more maturities than before. In turn, this
may have reinforced the transmission of monetary shocks to lending and deposit rates.
Without reliable “risk-free” benchmarks, interest rate setters had to produce an individual
forecast of the future path of short term policy rates in order to determine longer term deposit
or loan interest rates. Such a forecast could be compared with other agents’ forecasts only
with lags and noise, through the examination of competitors’ interest rates. In these
circumstances, future policy forecasts may be rather inaccurate, and a policy shock may be
more frequently associated with a forecast error than with a signal of a changing policy
stance. Hence, transmission could be low.
In the presence of a liquid TES market, interest rate setters could have an immediate,
centralized source of information regarding others’ views on future monetary policy. As a
consequence, the forecasts of future policy rates may have become more precise and a
monetary policy shock could be more frequently interpreted as a signal of changing policy
stance than as simple forecast error noise. Given that monetary policy shifts have some
persistence (they are rarely undone in the short term), the surprise involved in the shock is
informative of a path of future central bank interest rates that is likely to be higher or lower
than previously expected. Hence, transmission could be greater.
5. Conclusion
In the past decade the Colombian authorities undertook a series of measures that reduced
the structural fiscal deficit, corrected a possibly unsustainable public debt path, decreased
the government currency mismatch and deepened the local fixed-rate public bond market.
The evidence in this paper suggests that these improvements had profound effects on the
behavior of the macroeconomy. More specifically, they permanently reduced the sovereign

BIS Papers No 67 129

risk premium (with the ensuing effects on the real interest and exchange rates), increased
the reaction of output to (unexpected) government expenditure shocks (but still with
multipliers lower than one) and may have strengthened the response of market interest rates
to (unanticipated) monetary policy interest rate shocks. As a corollary, increased soundness
of fiscal policy may not only result in permanently lower costs of funding for all agents in the
economy, but may also enhance the power of fiscal and monetary policy to act
countercyclically.

130 BIS Papers No 67

Graph 1
Central government debt to GDP ratio

Graph 2
Currency mismatch indicator for central government

15
20
25
30
35
40
45
50
98 99 00 01 02 03 04 05 06 07 08 09 10 11
%
Quarters
0.5
1.0
1.5
2.0
2.5
3.0
3.5
98 99 00 01 02 03 04 05 06 07 08 09 10 11
%
Quarters

BIS Papers No 67 131

Graph 3
Composition of the domestic public debt

Graph 4
Average maturity of new issues of TES

.0
.1
.2
.3
.4
.5
.6
.7
.8
.9
96 97 98 99 00 01 02 03 04 05 06 07 08 09 10 11
Exchange rate linked (or issued locally in foreign currency)
Inflation indexed
Straight fixed rate
Other
Year
%
0
2
4
6
8
10
12
14
16
2002 2003 2004 2005 2006 2007 2008 2009 2010 2011
Polynomial of time of power 3
Average Maturity

132 BIS Papers No 67

Graph 5
Maturity composition of the fixed-rate TES stock

Table 1
Determination of the EMBI Colombia: long run relationships

Dependent variable: EMBI Colombia
Variable Coefficient Std. Error t-Statistic Prob.
gra: LVIX 0.6266 0.1700 3.6847 0.0006
log(d/y) 0.8529 0.3850 2.2153 0.0321
log(cm) 1.2614 0.1669 7.5569 0.0000
C 0.4002 1.6093 0.2487 0.8048
Cointegration Test Hansen (1992)
LM= 0.392339 p-value >0.20

Dependent variable: EMBI Colombia
Variable Coefficient Std. Error t-Statistic Prob.
gra: LSPREAD 0.5565 0.1229 4.5281 0.0000
log(d/y) 0.5061 0.3247 1.5586 0.1264
log(cm) 1.3208 0.1446 9.1328 0.0000
C 2.5258 1.2213 2.0681 0.0447
Cointegration Test Hansen (1992)
LM=0.474112 p-value >0.20

.0
.1
.2
.3
.4
.5
.6
.7
2002 2003 2004 2005 2006 2007 2008 2009 2010 2011
>5 1-5 0-1
year
%

BIS Papers No 67 133

Graph 6
Difference between actual and debt-stabilizing primary balances
(% of GDP)

-.010
-.008
-.006
-.004
-.002
.000
.002
.004
.006
98 99 00 01 02 03 04 05 06 07 08 09 10
Quarters
%
P
IB

134 BIS Papers No 67

Graph 7
Fiscal policy shock: Output responses conditional
on the debt to GDP ratio

-.8
-.6
-.4
-.2
.0
.2
.4
.6
1 2 3 4 5 6 7 8 9 10 11
1999
Quarters
-.6
-.4
-.2
.0
.2
.4
.6
1 2 3 4 5 6 7 8 9 10 11
2000
Quarters
-.6
-.4
-.2
.0
.2
.4
.6
1 2 3 4 5 6 7 8 9 10 11
2001
Quarters
-.2
.0
.2
.4
.6
1 2 3 4 5 6 7 8 9 10 11
2002
Quarters
-.2
.0
.2
.4
.6
.8
1 2 3 4 5 6 7 8 9 10 11
2003
Quarters
-.2
.0
.2
.4
.6
.8
1 2 3 4 5 6 7 8 9 10 11
2004
Quarters
-.2
.0
.2
.4
.6
.8
1 2 3 4 5 6 7 8 9 10 11
2005
Quarters
-.2
.0
.2
.4
.6
.8
1 2 3 4 5 6 7 8 9 10 11
2006
Quarters
-.2
.0
.2
.4
.6
1 2 3 4 5 6 7 8 9 10 11
2007
Quarters
-.2
.0
.2
.4
.6
1 2 3 4 5 6 7 8 9 10 11
2008
Quarters
-.2
.0
.2
.4
.6
1 2 3 4 5 6 7 8 9 10 11
2009
Quarters
-.2
.0
.2
.4
.6
1 2 3 4 5 6 7 8 9 10 11
2010
Quarters
-.2
.0
.2
.4
.6
1 2 3 4 5 6 7 8 9 10 11
2011
Quarters

BIS Papers No 67 135

Graph 8
Fiscal policy shock: Output responses conditional
on the difference between actual primary balance
and its debt-stabilizing level

-.4
-.2
.0
.2
.4
.6
.8
1 2 3 4 5 6 7 8 9 10 11
1999
Quarters
-.6
-.4
-.2
.0
.2
.4
.6
1 2 3 4 5 6 7 8 9 10 11
2000
Quarters
-.4
-.2
.0
.2
.4
.6
.8
1 2 3 4 5 6 7 8 9 10 11
2001
Quarters
-.4
-.2
.0
.2
.4
.6
1 2 3 4 5 6 7 8 9 10 11
2002
Quarters
-.2
.0
.2
.4
.6
.8
1 2 3 4 5 6 7 8 9 10 11
2003
Quarters
-.2
.0
.2
.4
.6
.8
1 2 3 4 5 6 7 8 9 10 11
2004
Quarters
-.2
.0
.2
.4
.6
.8
1 2 3 4 5 6 7 8 9 10 11
2005
Quarters
-.2
.0
.2
.4
.6
.8
1 2 3 4 5 6 7 8 9 10 11
2006
Quarters
-0.2
0.0
0.2
0.4
0.6
0.8
1.0
1.2
1 2 3 4 5 6 7 8 9 10 11
2007
Quarters
-0.2
0.0
0.2
0.4
0.6
0.8
1.0
1.2
1 2 3 4 5 6 7 8 9 10 11
2008
Quarters
-0.4
-0.2
0.0
0.2
0.4
0.6
0.8
1.0
1 2 3 4 5 6 7 8 9 10 11
2009
Quarters
-.2
.0
.2
.4
.6
.8
1 2 3 4 5 6 7 8 9 10 11
2010
Quarters
-.2
.0
.2
.4
.6
.8
1 2 3 4 5 6 7 8 9 10 11
2011
Quarters

136 BIS Papers No 67

Graph 9
Fiscal policy shock: Output responses conditional
on the currency mismatch indicator

-.6
-.4
-.2
.0
.2
.4
1 2 3 4 5 6 7 8 9 10 11
1999
Quarters
-.8
-.6
-.4
-.2
.0
.2
.4
.6
1 2 3 4 5 6 7 8 9 10 11
2000
Quarters
-.6
-.4
-.2
.0
.2
.4
.6
1 2 3 4 5 6 7 8 9 10 11
2001
Quarters
-.4
-.2
.0
.2
.4
.6
1 2 3 4 5 6 7 8 9 10 11
2002
Quarters
-.6
-.4
-.2
.0
.2
.4
.6
1 2 3 4 5 6 7 8 9 10 11
2003
Quarters
-.6
-.4
-.2
.0
.2
.4
.6
1 2 3 4 5 6 7 8 9 10 11
2004
Quarters
-.4
-.2
.0
.2
.4
.6
1 2 3 4 5 6 7 8 9 10 11
2005
Quarters
-0.2
0.0
0.2
0.4
0.6
0.8
1.0
1 2 3 4 5 6 7 8 9 10 11
2006
Quarters
-0.2
0.0
0.2
0.4
0.6
0.8
1.0
1 2 3 4 5 6 7 8 9 10 11
2007
Quarters
-0.2
0.0
0.2
0.4
0.6
0.8
1.0
1 2 3 4 5 6 7 8 9 10 11
2008
Quarters
-0.2
0.0
0.2
0.4
0.6
0.8
1.0
1 2 3 4 5 6 7 8 9 10 11
2009
Quarters
-0.2
0.0
0.2
0.4
0.6
0.8
1.0
1 2 3 4 5 6 7 8 9 10 11
2010
Quarters
-0.2
0.0
0.2
0.4
0.6
0.8
1.0
1 2 3 4 5 6 7 8 9 10 11
2011
Quarters

BIS Papers No 67 137

Graph 10
Monetary policy shock: Response of TES with maturity
less than one year conditional on the average maturity
of new issues of fixed-rate TES

138 BIS Papers No 67

Graph 11
Monetary policy shock: Response of TES with maturity
between one and three years conditional on the average maturity
of new issues of fixed-rate TES

BIS Papers No 67 139

Graph 12
Monetary policy shock: Response of TES with maturity
between three and five years conditional on the average maturity
of new issues of fixed-rate TES

140 BIS Papers No 67

Graph 13
Monetary policy shock: Response of TES with maturity
greater than five years conditional on the average maturity
of new issues of fixed-rate TES

BIS Papers No 67 141

Graph 14
Monetary policy shock: Response of commercial loan rate
with maturity less than one year conditional on the
average maturity of new issues of fixed-rate TES

– 4 0
– 3 0
– 2 0
– 1 0
0
1 0
2 0
1 2 3 4 5 6 7 8 9 10 11
2 0 0 2
Q uarters
– 4 0
– 3 0
– 2 0
– 1 0
0
1 0
2 0
1 2 3 4 5 6 7 8 9 10 11
2 0 0 3
Q uarters
– 1 0
0
1 0
2 0
3 0
4 0
1 2 3 4 5 6 7 8 9 10 11
2 0 0 4
Q uarters
– 2 0
0
2 0
4 0
6 0
8 0
1 0 0
1 2 0
1 2 3 4 5 6 7 8 9 10 11
2 0 0 5
Q uarters
– 4 0
0
4 0
8 0
1 2 0
1 6 0
1 2 3 4 5 6 7 8 9 10 11
2 0 0 6
Q uarters
– 4 0
0
4 0
8 0
1 2 0
1 6 0
2 0 0
1 2 3 4 5 6 7 8 9 10 11
2 0 0 7
Q uarters
– 4 0
0
4 0
8 0
1 2 0
1 6 0
2 0 0
1 2 3 4 5 6 7 8 9 10 11
2 0 0 8
Q uarters
– 5 0
0
5 0
1 0 0
1 5 0
2 0 0
1 2 3 4 5 6 7 8 9 10 11
2 0 0 9
Q uarters
– 5 0
0
5 0
1 0 0
1 5 0
2 0 0
1 2 3 4 5 6 7 8 9 10 11
2 0 1 0
Q uarters
– 5 0
0
5 0
1 0 0
1 5 0
2 0 0
1 2 3 4 5 6 7 8 9 10 11
2 0 1 1
Q uarters

142 BIS Papers No 67

Graph 15
Monetary policy shock: Response of commercial loan rate
with maturity between one and three years conditional on the
average maturity of new issues of fixed-rate TES

– 6 0
– 4 0
– 2 0
0
2 0
4 0
1 2 3 4 5 6 7 8 9 10 11
2 0 0 2
Q uarters
– 6 0
– 4 0
– 2 0
0
2 0
4 0
1 2 3 4 5 6 7 8 9 10 11
2 0 0 3
Q uarters
– 2 0
0
2 0
4 0
6 0
8 0
1 2 3 4 5 6 7 8 9 10 11
2 0 0 4
Q uarters
0
4 0
8 0
1 2 0
1 6 0
2 0 0
1 2 3 4 5 6 7 8 9 10 11
2 0 0 5
Q uarters
0
5 0
1 0 0
1 5 0
2 0 0
2 5 0
1 2 3 4 5 6 7 8 9 10 11
2 0 0 6
Q uarters
0
5 0
1 0 0
1 5 0
2 0 0
2 5 0
1 2 3 4 5 6 7 8 9 10 11
2 0 0 7
Q uarters
0
5 0
1 0 0
1 5 0
2 0 0
2 5 0
1 2 3 4 5 6 7 8 9 10 11
2 0 0 8
Q uarters
0
5 0
1 0 0
1 5 0
2 0 0
2 5 0
1 2 3 4 5 6 7 8 9 10 11
2 0 0 9
Q uarters
0
4 0
8 0
1 2 0
1 6 0
2 0 0
2 4 0
2 8 0
1 2 3 4 5 6 7 8 9 10 11
2 0 1 0
Q uarters
0
5 0
1 0 0
1 5 0
2 0 0
2 5 0
3 0 0
1 2 3 4 5 6 7 8 9 10 11
2 0 1 1
Q uarters

BIS Papers No 67 143

Graph 16
Monetary policy shock: Response of commercial loan rate
with maturity between three and five years conditional on the
average maturity of new issues of fixed-rate TES

– 8 0
– 6 0
– 4 0
– 2 0
0
2 0
4 0
1 2 3 4 5 6 7 8 9 10 11
2 0 0 2
Q uarters
– 6 0
– 4 0
– 2 0
0
2 0
4 0
1 2 3 4 5 6 7 8 9 10 11
2 0 0 3
Q uarters
– 2 0
0
2 0
4 0
6 0
8 0
1 2 3 4 5 6 7 8 9 10 11
2 0 0 4
Q uarters
– 5 0
0
5 0
1 0 0
1 5 0
2 0 0
2 5 0
1 2 3 4 5 6 7 8 9 10 11
2 0 0 5
Q uarters
0
1 0 0
2 0 0
3 0 0
4 0 0
1 2 3 4 5 6 7 8 9 10 11
2 0 0 6
Q uarters
0
1 0 0
2 0 0
3 0 0
4 0 0
1 2 3 4 5 6 7 8 9 10 11
2 0 0 7
Q uarters
0
1 0 0
2 0 0
3 0 0
4 0 0
1 2 3 4 5 6 7 8 9 10 11
2 0 0 8
Q uarters
0
1 0 0
2 0 0
3 0 0
4 0 0
1 2 3 4 5 6 7 8 9 10 11
2 0 0 9
Q uarters
0
1 0 0
2 0 0
3 0 0
4 0 0
1 2 3 4 5 6 7 8 9 10 11
2 0 1 0
Q uarters
0
1 0 0
2 0 0
3 0 0
4 0 0
1 2 3 4 5 6 7 8 9 10 11
2 0 1 1
Q uarters

144 BIS Papers No 67

Graph 17
Monetary policy shock: Response of commercial loan rate
with maturity greater than five years conditional on the
average maturity of new issues of fixed-rate TES

– 6 0
– 4 0
– 2 0
0
2 0
4 0
1 2 3 4 5 6 7 8 9 10 11
2 0 0 2
Q uarters
– 6 0
– 4 0
– 2 0
0
2 0
1 2 3 4 5 6 7 8 9 10 11
2 0 0 3
Q uarters
– 4 0
– 2 0
0
2 0
4 0
6 0
1 2 3 4 5 6 7 8 9 10 11
2 0 0 4
Q uarters
– 4 0
0
4 0
8 0
1 2 0
1 6 0
1 2 3 4 5 6 7 8 9 10 11
2 0 0 5
Q uarters
– 5 0
0
5 0
1 0 0
1 5 0
2 0 0
1 2 3 4 5 6 7 8 9 10 11
2 0 0 6
Q uarters
– 5 0
0
5 0
1 0 0
1 5 0
2 0 0
2 5 0
1 2 3 4 5 6 7 8 9 10 11
2 0 0 7
Q uarters
– 5 0
0
5 0
1 0 0
1 5 0
2 0 0
2 5 0
1 2 3 4 5 6 7 8 9 10 11
2 0 0 8
Q uarters
– 5 0
0
5 0
1 0 0
1 5 0
2 0 0
2 5 0
1 2 3 4 5 6 7 8 9 10 11
2 0 0 9
Q uarters
– 5 0
0
5 0
1 0 0
1 5 0
2 0 0
2 5 0
1 2 3 4 5 6 7 8 9 10 11
2 0 1 0
Q uarters
– 5 0
0
5 0
1 0 0
1 5 0
2 0 0
2 5 0
1 2 3 4 5 6 7 8 9 10 11
2 0 1 1
Q uarters

BIS Papers No 67 145

Graph 18
Monetary policy shock: Response of the consumer loan rate
with maturity less than one year conditional on the
average maturity of new issues of fixed-rate TES

146 BIS Papers No 67

Graph 19
Monetary policy shock: Response of the consumer loan rate
with maturity between one and three years conditional on the
average maturity of new issues of fixed-rate TES

– 6 0
– 4 0
– 2 0
0
2 0
1 2 3 4 5 6 7 8 9 10 11
2 0 0 2
Q uarters
– 5 0
– 4 0
– 3 0
– 2 0
– 1 0
0
1 0
1 2 3 4 5 6 7 8 9 10 11
2 0 0 3
Q uarters
– 3 0
– 2 0
– 1 0
0
1 0
2 0
1 2 3 4 5 6 7 8 9 10 11
2 0 0 4
Q uarters
– 4 0
– 2 0
0
2 0
4 0
6 0
8 0
1 2 3 4 5 6 7 8 9 10 11
2 0 0 5
Q uarters
– 4 0
0
4 0
8 0
1 2 0
1 2 3 4 5 6 7 8 9 10 11
2 0 0 6
Q uarters
– 4 0
0
4 0
8 0
1 2 0
1 2 3 4 5 6 7 8 9 10 11
2 0 0 7
Q uarters
– 4 0
0
4 0
8 0
1 2 0
1 2 3 4 5 6 7 8 9 10 11
2 0 0 8
Q uarters
– 4 0
0
4 0
8 0
1 2 0
1 2 3 4 5 6 7 8 9 10 11
2 0 0 9
Q uarters
– 4 0
0
4 0
8 0
1 2 0
1 6 0
1 2 3 4 5 6 7 8 9 10 11
2 0 1 0
Q uarters
– 4 0
0
4 0
8 0
1 2 0
1 6 0
1 2 3 4 5 6 7 8 9 10 11
2 0 1 1
Q uarters

BIS Papers No 67 147

Graph 20
Monetary policy shock: Response of the consumer loan rate
with maturity between three and five years conditional on the
average maturity of new issues of fixed-rate TES

148 BIS Papers No 67

Graph 21
Monetary policy shock: Response of the consumer loan rate
with maturity greater than five years conditional on the
average maturity of new issues of fixed-rate TES

– 8 0
– 6 0
– 4 0
– 2 0
0
2 0
1 2 3 4 5 6 7 8 9 10 11
2 0 0 2
Q uarters
– 6 0
– 4 0
– 2 0
0
2 0
1 2 3 4 5 6 7 8 9 10 11
2 0 0 3
Q uarters
– 3 0
– 2 0
– 1 0
0
1 0
2 0
3 0
1 2 3 4 5 6 7 8 9 10 11
2 0 0 4
Q uarters
– 2 0
0
2 0
4 0
6 0
8 0
1 0 0
1 2 3 4 5 6 7 8 9 10 11
2 0 0 5
Q uarters
– 4 0
0
4 0
8 0
1 2 0
1 6 0
1 2 3 4 5 6 7 8 9 10 11
2 0 0 6
Q uarters
– 4 0
0
4 0
8 0
1 2 0
1 6 0
1 2 3 4 5 6 7 8 9 10 11
2 0 0 7
Q uarters
– 4 0
0
4 0
8 0
1 2 0
1 6 0
1 2 3 4 5 6 7 8 9 10 11
2 0 0 8
Q uarters
– 4 0
0
4 0
8 0
1 2 0
1 6 0
2 0 0
1 2 3 4 5 6 7 8 9 10 11
2 0 0 9
Q uarters
– 4 0
0
4 0
8 0
1 2 0
1 6 0
2 0 0
1 2 3 4 5 6 7 8 9 10 11
2 0 1 0
Q uarters
– 5 0
0
5 0
1 0 0
1 5 0
2 0 0
1 2 3 4 5 6 7 8 9 10 11
2 0 1 1
Q uarters

BIS Papers No 67 149

Graph 22
Monetary policy shock: Response of the CDT rate
with maturity less than 90 days conditional on the
average maturity of new issues of fixed-rate TES

– 3 0
– 2 0
– 1 0
0
1 0
1 2 3 4 5 6 7 8 9 10 11
2 0 0 2
Q uarters
– 2 0
– 1 0
0
1 0
1 2 3 4 5 6 7 8 9 10 11
2 0 0 3
Q uarters
– 1 0
0
1 0
2 0
3 0
4 0
1 2 3 4 5 6 7 8 9 10 11
2 0 0 4
Q uarters
0
2 0
4 0
6 0
8 0
1 0 0
1 2 3 4 5 6 7 8 9 10 11
2 0 0 5
Q uarters
0
2 0
4 0
6 0
8 0
1 0 0
1 2 0
1 2 3 4 5 6 7 8 9 10 11
2 0 0 6
Q uarters
0
4 0
8 0
1 2 0
1 6 0
1 2 3 4 5 6 7 8 9 10 11
2 0 0 7
Q uarters
0
4 0
8 0
1 2 0
1 6 0
1 2 3 4 5 6 7 8 9 10 11
2 0 0 8
Q uarters
0
4 0
8 0
1 2 0
1 6 0
1 2 3 4 5 6 7 8 9 10 11
2 0 0 9
Q uarters
0
4 0
8 0
1 2 0
1 6 0
1 2 3 4 5 6 7 8 9 10 11
2 0 1 0
Q uarters
0
2 5
5 0
7 5
1 0 0
1 2 5
1 5 0
1 2 3 4 5 6 7 8 9 10 11
2 0 1 1
Q uarters

150 BIS Papers No 67

Graph 23
Monetary policy shock: Response of the CDT rate
with maturity of 90 days conditional on the average
maturity of new issues of fixed-rate TES

BIS Papers No 67 151

Graph 24
Monetary policy shock: Response of the CDT rate
with maturity between 91 and 170 days conditional on the
average maturity of new issues of fixed-rate TES

– 3 0
– 2 0
– 1 0
0
1 0
2 0
1 2 3 4 5 6 7 8 9 10 11
2 0 0 2
Q uarters
– 2 0
– 1 0
0
1 0
2 0
1 2 3 4 5 6 7 8 9 10 11
2 0 0 3
Q uarters
0
1 0
2 0
3 0
4 0
1 2 3 4 5 6 7 8 9 10 11
2 0 0 4
Q uarters
0
2 0
4 0
6 0
8 0
1 0 0
1 2 3 4 5 6 7 8 9 10 11
2 0 0 5
Q uarters
0
4 0
8 0
1 2 0
1 6 0
1 2 3 4 5 6 7 8 9 10 11
2 0 0 6
Q uarters
0
4 0
8 0
1 2 0
1 6 0
1 2 3 4 5 6 7 8 9 10 11
2 0 0 7
Q uarters
0
4 0
8 0
1 2 0
1 6 0
1 2 3 4 5 6 7 8 9 10 11
2 0 0 8
Q uarters
0
4 0
8 0
1 2 0
1 6 0
1 2 3 4 5 6 7 8 9 10 11
2 0 0 9
Q uarters
0
2 5
5 0
7 5
1 0 0
1 2 5
1 5 0
1 2 3 4 5 6 7 8 9 10 11
2 0 1 0
Q uarters
0
4 0
8 0
1 2 0
1 6 0
1 2 3 4 5 6 7 8 9 10 11
2 0 1 1
Q uarters

152 BIS Papers No 67

Graph 25
Monetary policy shock: Response of the CDT rate
with maturity of 180 days conditional on the average
maturity of new issues of fixed-rate TES

– 1 5
– 1 0
– 5
0
5
1 0
1 5
1 2 3 4 5 6 7 8 9 10 11
2 0 0 2
Q uarters
– 1 0
– 5
0
5
1 0
1 5
1 2 3 4 5 6 7 8 9 10 11
2 0 0 3
Q uarters
0
1 0
2 0
3 0
4 0
1 2 3 4 5 6 7 8 9 10 11
2 0 0 4
Q uarters
0
2 0
4 0
6 0
8 0
1 2 3 4 5 6 7 8 9 10 11
2 0 0 5
Q uarters
0
2 0
4 0
6 0
8 0
1 0 0
1 2 0
1 2 3 4 5 6 7 8 9 10 11
2 0 0 6
Q uarters
0
2 0
4 0
6 0
8 0
1 0 0
1 2 0
1 2 3 4 5 6 7 8 9 10 11
2 0 0 7
Q uarters
0
2 0
4 0
6 0
8 0
1 0 0
1 2 0
1 2 3 4 5 6 7 8 9 10 11
2 0 0 8
Q uarters
0
4 0
8 0
1 2 0
1 6 0
1 2 3 4 5 6 7 8 9 10 11
2 0 0 9
Q uarters
0
4 0
8 0
1 2 0
1 6 0
1 2 3 4 5 6 7 8 9 10 11
2 0 1 0
Q uarters
0
4 0
8 0
1 2 0
1 6 0
1 2 3 4 5 6 7 8 9 10 11
2 0 1 1
Q uarters

BIS Papers No 67 153

Graph 26
Monetary policy shock: Response of the CDT rate
with maturity between 181 and 360 days conditional on the
average maturity of new issues of fixed-rate TES

– 2 0
– 1 0
0
1 0
2 0
1 2 3 4 5 6 7 8 9 10 11
2 0 0 2
Q uarters
– 1 5
– 1 0
– 5
0
5
1 0
1 5
1 2 3 4 5 6 7 8 9 10 11
2 0 0 3
Q uarters
– 1 0
0
1 0
2 0
3 0
4 0
1 2 3 4 5 6 7 8 9 10 11
2 0 0 4
Q uarters
0
2 0
4 0
6 0
8 0
1 0 0
1 2 3 4 5 6 7 8 9 10 11
2 0 0 5
Q uarters
0
2 0
4 0
6 0
8 0
1 0 0
1 2 0
1 2 3 4 5 6 7 8 9 10 11
2 0 0 6
Q uarters
0
4 0
8 0
1 2 0
1 6 0
1 2 3 4 5 6 7 8 9 10 11
2 0 0 7
Q uarters
0
4 0
8 0
1 2 0
1 6 0
1 2 3 4 5 6 7 8 9 10 11
2 0 0 8
Q uarters
0
4 0
8 0
1 2 0
1 6 0
1 2 3 4 5 6 7 8 9 10 11
2 0 0 9
Q uarters
0
4 0
8 0
1 2 0
1 6 0
1 2 3 4 5 6 7 8 9 10 11
2 0 1 0
Q uarters
0
2 5
5 0
7 5
1 0 0
1 2 5
1 5 0
1 2 3 4 5 6 7 8 9 10 11
2 0 1 1
Q uarters

154 BIS Papers No 67

Graph 27
Monetary policy shock: Response of the CDT rate
with maturity greater than 360 days conditional on the
average maturity of new issues of fixed-rate TES

– 2 0
– 1 0
0
1 0
2 0
3 0
1 2 3 4 5 6 7 8 9 10 11
2 0 0 2
Q uarters
– 2 0
– 1 0
0
1 0
2 0
3 0
1 2 3 4 5 6 7 8 9 10 11
2 0 0 3
Q uarters
– 2 0
0
2 0
4 0
6 0
8 0
1 2 3 4 5 6 7 8 9 10 11
2 0 0 4
Q uarters
0
2 5
5 0
7 5
1 0 0
1 2 5
1 5 0
1 2 3 4 5 6 7 8 9 10 11
2 0 0 5
Q uarters
0
5 0
1 0 0
1 5 0
2 0 0
1 2 3 4 5 6 7 8 9 10 11
2 0 0 6
Q uarters
0
5 0
1 0 0
1 5 0
2 0 0
2 5 0
1 2 3 4 5 6 7 8 9 10 11
2 0 0 7
Q uarters
0
5 0
1 0 0
1 5 0
2 0 0
2 5 0
1 2 3 4 5 6 7 8 9 10 11
2 0 0 8
Q uarters
0
5 0
1 0 0
1 5 0
2 0 0
2 5 0
1 2 3 4 5 6 7 8 9 10 11
2 0 0 9
Q uarters
0
5 0
1 0 0
1 5 0
2 0 0
2 5 0
1 2 3 4 5 6 7 8 9 10 11
2 0 1 0
Q uarters
0
5 0
1 0 0
1 5 0
2 0 0
2 5 0
1 2 3 4 5 6 7 8 9 10 11
2 0 1 1
Q uarters

BIS Papers No 67 155

Graph 28
Fiscal shock

Graph 29
Monetary policy shock

-6,000
-4,000
-2,000
0
2,000
4,000
6,000
99 00 01 02 03 04 05 06 07 08 09 10 11
B
ill
io
ns
o
f
C
ol
om
bi
an
P
es
os

D
ec
em
be
r
of
2
01
0
Quarters
-.04
-.03
-.02
-.01
.00
.01
.02
2002 2003 2004 2005 2006 2007 2008 2009 2010 2011
bp
Quarters

156 BIS Papers No 67

Appendix 1:
Calculation of the government expenditure shocks
To construct the spending forecast of the central government we followed these steps:
a. The budget execution rate for each quarter in a year was obtained from the annual
and quarterly historical data on actual expenditures.
b. The annual spending announcements made by the government at the beginning of
each year in its Financial Plans are considered as the annual spending forecast.
c. Based on (a) and (b), we predict the government spending for the four quarters of
each year by multiplying the corresponding budget execution rate (using a moving
average of fourth order) by the annual spending announcements.
d. By the end of the second quarter, information on the actual first quarter expenditure
is available. Thus, we add an adjustment to the forecast of the third and fourth
quarters that results from the assumptions that the annual expenditure plan will be
fulfilled and that the first quarter forecast error is uniformly distributed between the
second, third and fourth quarters.
e. By the end of the third quarter, information on the actual second quarter expenditure
is available. Thus, we add an adjustment to the forecast of the fourth quarter that
results from the assumptions that the annual expenditure plan will be fulfilled and
that the second quarter forecast error is uniformly distributed between the third and
fourth quarters.
f. The series of forecast errors (calculated with respect to the adjusted forecasts in the
case of the third and fourth quarters) is the expenditure shock for each quarter.
Graph 28 shows the fiscal shock (measured in 2010 COP billions).

BIS Papers No 67 157

Appendix 2:
Estimation of the monetary policy shock
Similar to what is usually done in the VAR literature, we define monetary policy shock as an
unexpected movement of the policy rate. That is, we suppose that there is a policy rule that
relates the state of the economy with the actions of the monetary authorities and
consequently a monetary policy shock will be a movement in the policy rate not explained by
the rule. For example, under the assumption that the central bank follows a standard Taylor
rule, a movement in the policy rate not explained by the observed behavior of inflation and
output will be a monetary shock. However, if the central bank follows an expectations-based
rule, that is, a rule in which the expected value of inflation and output are important, then it
is natural to include within an estimated Taylor rule not just current inflation and output but
also any other variables that can be useful indicators about the future behavior of these
variables.
Notice also that under the VAR recursive identification, a monetary policy shock is not only
an unexpected movement of the policy rate but is also orthogonal to the information set of
the central bank. In other words, it is assumed that a variable that is observed by the
central bank cannot react contemporaneously to the policy shock. With this in mind, it is
possible to see that a forecast error can serve as proxy of a policy shock. In fact, we
defined the policy shock through the forecast error: it+1 – E[it+1|Wt], where it+1 is the actual
policy rate at time t+1 and E[it+1|Wt] is its expected value given the information set at time t
denoted by Wt.
Our definition of the policy shocks is consistent with the definition of the policy shock in a
VAR model for two reasons. First, it captures unexpected movements in the policy rate and,
second, by definition it is orthogonal the information set. However, given our definition of a
policy shock, we can capture policy shocks that are policy errors or changes in the policy
stance not necessarily expected at time t. In the first case, the policy rate is, unintentionally,
too low or too high with respect to what is dictated by a policy rule, whereas in the second
case, the policy shock signals a change in the monetary policy stance. The source of the
policy shocks can have very different effects on the economy.
To make this definition of the policy shock operational one needs to be particularly carefully
about the definition of the information set Wt and the way E[it+1|Wt] is estimated. Empirically,
the main concern with Wt is not to include variables that are not observed at time t. In our
exercise, the information set contains information on inflation, output, credit, the exchange
rate, etc. However, some of these variables are observed with delay and consequently their
current values cannot be in Wt.
We approximate E[it+1|Wt] with linear projections. That is, E[it+1|Wt] = a o + a 1 xt , where xt is an
element of Wt. a o and a 1 are estimated by OLS. We select the elements in xt by minimizing
the AIC criterion.
Finally, to construct a sequence of monetary policy shocks we carry out a rolling exercise
where we forecast it+1 at time t and compare it with the actual value of it+1. At each t the
information set is updated and the elements of xt are selected by minimizing the AIC
criterion. The initial sample of the rolling experiment is 1999m9-2000m12 and is expanded
until 2011m12.

158 BIS Papers No 67

The policy shocks are constructed using monthly data on the interbank rate, the Colombian
inflation target, the growth rate of the index industrial production, the growth rate of credit, the
index of capacity utilization, the nominal average unit labor cost, the nominal depreciation of
the Colombian peso, the Index of Consumer Confidence (ICC) and the US inflation rate21.
The shocks are shown in Graph 29.

21 All growth rates are annual, the index of capacity utilization, and the nominal average unitary labor cost are
included in annual changes. Data are seasonally adjusted using TRAMO-SEATS in Eviews). All these
variables are in general available with a delay of one month; however, the Index of Industrial Production, the
Unitary Labor cost and the ICC are observed with a delay of two months.

BIS Papers No 67 159

References
Auerbach, A. and Gorodnichenko, Y. (2012): “Fiscal Multipliers in Recession and
Expansion”. Paper prepared for the NBER conference Fiscal Policy after the Financial Crisis,
Milan, December 2011.
Blanchard, O. and Perotti, R. (2002): “An Empirical Characterization of the Dynamic Effects
of Changes in Government Spending and Taxes on Output”, Quarterly Journal of Economics.
Caldara, D. and Kamps C. (2008): “What are the effects of fiscal shocks? A VAR-based
comparative analysis”, Working Paper Series 877, European Central Bank.
Cárdenas, M., Mejía, C., and Olivera, M. (2006): “La economía política del proceso
presupuestal en Colombia”, Working Paper, no. 31, Fedesarrollo.
Favero, C. and Giavazzi, F. (2004): “Inflation Targeting and Debt: Lessons from Brazil”.
NBER Working Paper 10390. NBER.
Galindo, A. and Hofstetter, M. (2008): “Mortgage Interest Rates, Country Risk and Maturity
Matching in Colombia” Documento CEDE 2. Universidad de los Andes.
Goldstein, M. and Turner, P. (2004): Controlling Currency Mismatches in Emerging Markets.
Peterson Institute for International Economics. April.
Gonzalez, A. and Hamann, F., (2011). “Lack of Credibility, Inflation Persistence and
Disinflation in Colombia,” Revista Desarrollo y Sociedad, Universidad de Los Andes-Cede,
Vol. 67 p. 61.
Hamman, F., Lozano, I., and Mejía, L. F. (2011): “Sobre el Impacto Macroeconómico de los
Beneficios Tributarios al Capital”, in Borradores de Economía, N°668, Banco de la
República.
Hamann, F., Vargas, H., and González, A. (2010): “Efectos de la política monetaria sobre las
tasas de interés de los créditos hipotecarios en Colombia”. Revista Desarrollo y Sociedad
No. 66. Universidad de los Andes.
Hansen, B. E. (1992): “Tests for Parameter Instability in Regressions with I(1) Processes,”
Journal of Business and Economic Statistics, 10, 321-335.
Ilzetzki, E. and Vegh, C. (2008): “Procyclical fiscal policy in developing countries: truth or
fiction?”. NBER Working Paper, 14191, July.
Ilzetzki, E., Mendoza, E., and Vegh, C. (2009): “How big are fiscal multipliers?”. Policy Insight
No. 39. October. CEPR.
Jordà, Ò. (2005): “Estimation and Inference of Impulse Responses by Local Projections” in
The American Economic Review, Vol. 95 N°. 1, March.
Julio, J.M., Lozano, I., and Melo, L. (2012): “Quiebre Estructural de la Relación entre la
Política Fiscal y el Riesgo Soberano en los Países Emergentes: El caso colombiano”.
Borradores de Economía, N°668, Banco de la República.
Lozano, I. (2011): “Fiscal Policy in Colombia and a Prospective Analysis after the 2008
Global Financial Crisis”. Chap 17 in Fiscal Policy: Lessons from the Crisis, ed. by Banca
d’Italia – Eurosistema.
Lozano, I. and Rodríguez, K., (2011): “Assessing the Macroeconomic Effects of Fiscal Policy
in Colombia”, in Journal of Financial Economic Policy, Vol. 3, N°3, pp.206.
Perotti, R. (2004): “Estimating the Effects of Fiscal Policy in OECD Countries”, mimeo,
Bocconi University.
Perotti, R. (2007): “In Search of the Transmission Mechanism of Fiscla Policy”. NBER
Working Paper 13143. NBER.

160 BIS Papers No 67

Restrepo, J. and Rincón, H. (2006): “Identifying Fiscal Policy Shocks In Chile And Colombia”,
in Borradores de Economía, No. 397, Banco de la República.
Rojas-Suárez, L. and Montoro, C. (2011): “Latin America’s Real Credit Growth During the
Global Financial Crisis: Lessons on Coping with External Financial Shocks”. Mimeo.
Romer, C. D. (2011): “What do We Know about the Effects of Fiscal Policy? Separating
Evidence from Ideology”. Speech Hamilton College. November 7.

BIS Papers No 67 161

Some insights into monetary and fiscal policy
interactions in the Czech Republic
Vladimír Tomšík1
Abstract
The global financial and debt crisis highlights the need for a better understanding of how
fiscal and monetary policies interact. This article examines three aspects of these
interactions, as seen from the perspective of the Czech National Bank. It first looks at the
effects of fiscal policy on the interest rate channel in the Czech Republic, where long-term
government bond yields are an important determinant of market interest rates. Second, it
reviews alternative methods for the cyclical adjustment of the fiscal balance, which might
provide different assessments of the fiscal policy stance. Finally, it describes how fiscal
policy is included in the Czech National Bank forecast.

Keywords: Monetary policy, fiscal policy, interactions, transmission, Czech Republic
JEL classification: E52, E58, E62, E63

1 Vice Governor of the Czech National Bank. I thank the CNB’s Monetary Policy and Fiscal Analysis Division,
and especially -DQ�)LOiþHN��’DQD�+iMNRYi��3HWU�.UiO��3DYOD�1HWXãLORYi��$QFD�3RGSLHUD�DQG�%UDQLVODY�6D[D�IRU�
help in the preparation of this paper. Valuable comments were also provided by Tomáš Holub, head of the
Monetary and Statistics Department.

162 BIS Papers No 67

1. Introduction
In recent years, policymakers have responded aggressively with monetary and fiscal
measures to counteract the macroeconomic consequences of the economic and financial
crisis. Central banks have massively reduced interest rates. In many cases, they have
resorted to unconventional monetary policies. Fiscal authorities have run up high budget
deficits as revenues have declined and expenditures have been kept high in efforts to
safeguard the financial sector and to stimulate the demand side of the economy.
The Czech National Bank (CNB) started to cut interest rates in summer 2008, reaching a
record low level (a two-week repo rate of 0.75%) in May 2010. However, unlike many other
central banks that were pushed by unfavourable circumstances into using unconventional
monetary policy tools, the CNB was able to rely largely on standard policy instruments. It is
worth noting in this context that a marked depreciation of the koruna exchange rate in late
2008 also helped the bank substantially in terms of monetary loosening. Thus, the floating
exchange rate regime proved to be an efficient adjustment mechanism. In parallel, the Czech
government in 2009 approved an anti-crisis package of fiscal measures, which were aimed
at cushioning the impact of the crisis on Czech households and businesses.
The energetic response of central banks and governments around the globe has focused the
interest of researchers on the topic of the mutual interactions between monetary and fiscal
policy and the impact on the economy. The issue of monetary and fiscal policy interaction
has been examined in a vast number of research papers, the seminal ones being
Barro (1979), Lucas and Stokey (1983), Chari et al (1991). The more recent literature
includes Schmitt-Grohe and Uribe (2004) and Davig and Leeper (2011), and is based on
micro-based analysis of jointly optimal monetary and fiscal policies in economies featuring
nominal inertia, taxation and imperfect competition.
A consensus seems to have emerged among researchers (Kirsanova et al 2009) that
monetary policy should normally focus on business cycle stabilisation and inflation control,
and that fiscal policy should focus on the control of government debt or deficits. However, if
monetary policy is constrained in some way – either by design (eg for a monetary union
member subject to asymmetric shocks) or by circumstance (interest rates hit the zero lower
bound), fiscal policy should be used for business cycle stabilisation and inflation control. This
is in line with empirically observed patterns in recent years, when fiscal policy has gained in
importance for smoothening the business cycle and avoiding deflation, notably in the
countries hitting the zero lower bound for interest rates. From this perspective, purchases of
government securities conducted by some central banks might be considered as a monetary
policy tool that supports the fiscal stimulus with a view to stabilising output and inflation. The
other view could be, however, that such operations are a kind of monetary financing and that
these central banks have subjected themselves to fiscal dominance.
Monetary and fiscal policies interact in many ways, both nationally and internationally. The
vast number of interactions in the economy make it difficult to determine the specific
influence of each policy with any certainty. Research on monetary and fiscal policy
interactions can be divided into three strands. The first strand of research (for example,
Blinder (1982) and Tabellini (1986)) focuses on the effect of interaction of fiscal and
monetary policies using a formal game-theoretical approach. The second strand (notably
Lucas and Stokey (1983), Chari et al (1991) and Woodford (2003)) has analysed the
interactions using sophisticated macroeconomic models and has attempted to derive optimal
monetary and fiscal policy strategies. Finally, the third strand of literature is more data-driven;
using various econometric techniques (usually VaR), it investigates the impact of policy
interactions on the transmission mechanism.
This article falls into the third strand of research and outlines several issues in fiscal and
monetary policy interaction using the experience of the Czech Republic. In the second
chapter, we discuss the effects of fiscal policy on the transmission of monetary policy. The

BIS Papers No 67 163

third chapter presents two alternative methods for the cyclical adjustment of the fiscal
balance used for obtaining economically meaningful estimates of the fiscal stance. Based on
these estimates, we analyse the cyclicality of the Czech fiscal policy. The methods by which
fiscal policy is forecast and included in the CNB macroeconomic predictions is described in
the fourth chapter. The fifth chapter concludes.
2. The effects of fiscal policy on monetary policy transmission
Fiscal policy is an important determinant of economic developments and, as such, it affects
monetary policy through several channels. Some fiscal measures (such as introducing or
changing a consumption tax or value added tax) have a direct effect on inflation. Other fiscal
measures have indirect effects on inflation through their impact on aggregate demand.
Furthermore, fiscal policy influences other economic variables that are important in monetary
policy transmission, notably interest rates, interest rate spreads and exchange rates. In the
extreme case known as fiscal dominance, monetary policy might even become subordinate
to fiscal policy. As Sargent and Wallace (1981) first pointed out, this situation might emerge if
the fiscal authority sets its budget independently of public sector liabilities so that the fiscal
expansion eventually needs to be monetised, giving rise to high inflation and inflation
expectations.
In this chapter, we focus on the impact of fiscal policy on financial market interest rates and
consequently on commercial interest rates. A description of other channels through which
fiscal policy influences the monetary policy transmission is provided, for example, in
Zoli (2005).
A consensus exists that under most circumstances an expansionary fiscal policy is
associated with higher medium-term and long-term interest rates, ie that it crowds out private
investment. This is supported by many empirical studies. Most recently, López et al (2011),
using panel data of the long-term interest rate for the period 1990–2009 in 54 emerging and
developed countries, find that when the fiscal deficit expands by 1% long-term interest rates
rise between 10 and 12 basis points.
Similarly, a broad agreement exists on the role of public debt in determining long-term
interest rates. Higher public indebtedness increases the risk of default on sovereign debt,
which ultimately translates into higher spreads on government bonds (see, eg, Ferrucci
(2003)). The current situation in the European market is a clear reminder about the
importance of this mechanism.
Focusing on the Czech Republic, several studies estimate the effects of fiscal policy on long-
term interest rates. Alexopoulou et al (2009) assess the role of fundamentals in driving long-
term sovereign bond spreads in the new EU countries, including the Czech Republic, over
the period 2001–08. They find, inter alia, that an adverse 10% shock to external
indebtedness ratio shifts the long-run equilibrium spreads by 5 basis points in the Czech
Republic. Baldacci et al (2008) on a panel of 30 emerging market economies conclude that
an improvement in the primary budget balance by 1% of GDP helps to reduce spreads by
about 30–���EDVLV�SRLQWV��’XPLþLþ�DQG�5LG]DN��������XVH�SDQHO�GDWD�IRU�HLJKW�FHQWUDO�DQG�
eastern European countries and find that, if general government debt-to-GDP ratio increases
by 5 percentage points, spreads increase by 19 basis points.
Wider spreads consequently lead to higher yields on government bonds and to higher
commercial interest rates. The analysis of client interest rates on loans and deposits in the
Czech Republic in the period between January 2004 and December 2009 shows that
10-year government bond yields might be used as a benchmark rate for client long-term
interest rates (with maturity of more than one year in the case of loans and more than two
years in the case of deposits). The sensitivity of selected client interest rates to 10-year
government bonds as estimated by an error correction model is shown in Table 1.

164 BIS Papers No 67

Table 1
Transmission of changes in 10-year government bond yields into client rates
Loans to small corporations, fixed for
more than 1Y
0.33(0.27) Ļ 1.19*** (0.15) -0.4***(0.11) 3m
Loans to large corporations, fixed for
more than 1Y
0.67 (0.99) 0.83***(0.11) -0.64***(0.13) Ļ
2m
Loans to households – mortgage loans -0.09(0.07) 0.91***(0.04) Ĺ -0.28***(0.03) 3m
Deposits with maturity longer than 2Y -0.04(0.26) 0.73***( 0.07) Ļ -0.47***(0.09) Ļ 2m
Speed of
DGMXVWPHQW�ȕ0
Adjustment
speed in months
�ȕ1�Į0��ȕ0
Immediate pass-
WKURXJK�Į0
Final pass-
WKURXJK�ȕ1

Note: estimated equation
itittii
q
l
p
k
ktikiltliti mrbrbrmrbr embbaa +–+D+D=D —
= =
–� � )( 111,,0
0 1
,,
, where itbr
denotes the i-th bank interest rate at time t,
tmr represents the (bond) market rate and m is a constant that
quantifies the spread of bank interest rates vis-à-vis the market rates. Symbols ***, ** and * denote statistical
significance of parameters at 1%, 5% a 10% significance level. Standard errors are in parentheses. The symbol
denotes parameters that are statistically significantly lower (in absolute value at the 10% significance level)
than estimated parameters in the 2004–08 data sample. The symbol denotes the parameter that is statistically
significantly higher (at the 10% significance level) than the parameter estimated in the 2004–08 sample. Other
parameters are not statistically significantly different. The speed of adjustment was rounded to entire months.
Although the immediate (ie within one month) pass-through of government bond yields into
client rates is not statistically significant, the long-run pass-through is significant for all
interest rates shown in the table. A complete long-run pass-through is observed for loans to
small corporations with terms of more than one year and for mortgages.
During the financial crisis, the transmission of government bond yields into client interest
rates has slowed in the case of loans to large corporations and deposits. On the contrary, the
relationship between mortgage interest rates and the yield on long-term government bonds
has strengthened in the crisis period.
3. Two alternative methods for cyclical adjustment of fiscal balance
When assessing the fiscal policy stance, it is necessary to adjust the fiscal balance for its
cyclical component. The fiscal balance is affected by fluctuations in the economy, as an
expanding economy raises tax revenues and lowers social transfers (and vice versa). Rather
than looking at the overall balance, it is thus more appropriate to disregard this cyclicality and
focus on the structural balance. A structural, cyclically adjusted balance is defined as the
excess of public spending over revenues (or vice versa) that would persist if the economy
were near its potential.
The estimates of cyclically adjusted budget balances are routinely used by the European
Commission, the ECB, IMF, OECD and other institutions. The cyclically adjusted budget
balance (CAB) is one of the key indicators for the analysis and conduct of fiscal policy in the
EU fiscal surveillance framework. In this framework, the structural balance abstracts away

BIS Papers No 67 165

from cyclicality as well as from one-off and other temporary measures.2 The long-term EU
fiscal targets to be met by Member States under the provisions of the Stability and Growth
Pact are expressed and assessed net of cyclical conditions and one-off and other temporary
measures.
Several methods can be used to derive the cyclically adjusted balance. In this chapter, we
discuss two alternative methods as applied by the European Commission (EC) and the
European System of Central Banks (ESCB). Both these methods are calculated routinely at
the CNB. In the above-mentioned EU fiscal surveillance framework, cyclically adjusted
balances are estimated using the EC method and published in Stability and Convergence
Reports. The ESCB method is used by the ECB and other national banks within the ESCB
as an additional analysis and presented in two internal documents – the Public Finance
Report and the Autumn Fiscal Policy Note. The main difference between the two methods is
that, while the EC method is based on output gap calculations, the ESCB approach to
cyclical adjustment takes into account the composition effects originating from the different
cyclical behaviour of macroeconomic bases for the main revenue and expenditure
categories. Both methods abstract away from one-off and other temporary measures:
ttttt OGBBCCBBCAB �-=-= e
where tBB is the nominal budget balance in year t, tCC the cyclical component in the year t,
e the budgetary sensitivity parameter and tOG the output gap in the year t. The output gap
represents an economy’s cyclical position (difference between actual and potential output3).
The overall sensitivity parameter e is obtained by aggregating the elasticities of individual
cyclically sensitive budgetary items4. The individual revenue elasticities ( iR,h ) are aggregated
to an overall revenue elasticity using as weights the share of each revenue category in the
total current taxes ( RRi / ):
R
Ri
i
iRR �
=
=
4
1
,hh
As for the expenditure elasticity ( Gh ), it can be expressed as:
G
GU
UGG ,hh =
where UG,h is the elasticity of unemployment benefits and GGU / is their share in the current
primary expenditure.
The two elasticities UR hh , are then transformed into the overall sensitivity parameter of the
budget balance ( e ) used in the equation defining CAB as follows:
Y
G
Y
R
GRGR hheee -=-= ,

2 One-off and temporary measures are measures having a transitory budgetary effect that does not lead to a
sustained change in the intertemporal budgetary position (eg short-term costs emerging from natural
disasters, sales of non-financial assets).
3 Potential output is calculated on the basis of the Cobb-Douglas production function.
4 There are four tax categories (personal and corporate income tax, indirect taxes, and social contributions) and
one expenditure category (unemployment benefits).

166 BIS Papers No 67

where YR / is the share of current taxes in GDP, and YG / is the share of primary
expenditure in GDP. For a more detailed discussion of the EC method of cyclical adjustment
see Larch and Turrini (2009).
The ESCB has elaborated a different method used for estimating the cyclical component
( tCC ). In the ESCB method, the revenue and expenditure categories are adjusted individually
based on the deviation from trend5 of their relevant macroeconomic bases in real terms. The
following main budgetary items are adjusted (with corresponding macroeconomic bases in
brackets): direct taxes paid by households (average compensation of employees and
employment in the private sector), direct taxes paid by corporations (operating surplus),
social contributions paid in the private sector (average compensation of employees and
employment in the private sector), indirect taxes (private consumption) and unemployment-
related expenditure (number of unemployed persons).
The individual cyclical component of each budgetary category is calculated by applying a
constant elasticity to the trend deviation and then the CAB is calculated as follows:
)_____( ttttttttt CXUCRICRFCRSPCRHPBBCCBBCAB -+++-=-=
where tCRHP _ is the cyclical component of direct taxes paid by households, tCRSP _ the
cyclical component of direct taxes paid by corporations, tCRF _ the cyclical component of
social contribution paid in the private sector, tCRI _ the cyclical component of indirect taxes
and tCXU _ the cyclical component of unemployment-related expenditure. More details of
the ESCB method can be found in Bouthevillain et al (2001).
In Chart 1, the outcomes of these two methods are compared for the Czech Republic. One
can see from this chart that both these methods provide very similar estimations of the
cyclical component. Nevertheless, two noticeable exceptions are the years 2003 and 2009.
In 2003, the ESCB method estimates a slightly positive cyclical component due to positive
wage developments in that year, while the EC method takes into account a negative output
gap. In 2009, the more marked decline of the cyclical component in the EC method reflects
the immediate impact of the global crisis on the GDP growth, whereas the impact on wages
and private consumption was somewhat delayed and hence the ESCB method shows an
almost neutral cyclical position.
It is also evident from the chart that, before the recent crisis, structural deficits were
notoriously high as a result of a loose fiscal policy conducted in the years of prosperity
between 2003 and 2007. This unfavourable starting fiscal situation was subsequently
aggravated after the economic crisis hit the Czech Republic, when the cyclical position of the
economy sharply turned negative and automatic stabilisers came into effect.6 In addition, the
government approved an anti-crisis package of fiscal measures aimed at cushioning the
impact of the crisis on Czech households and businesses. These measures led to a further
deterioration in the public deficit.

5 Trends are estimated using the Hodrick-Prescott filter.
6 It is worth mentioning that the economic and financial crisis hit the Czech economy solely via the foreign
demand channel, very negatively affecting export and production performance of the Czech manufacturing
sector. That said, the financial impact of the crisis on the Czech economy was modest, thanks to the resilience
of the Czech banking sector, which had virtually no exposure to foreign toxic assets.

BIS Papers No 67 167

Chart 1
Cyclical decomposition of general government balance
In % of GDP
-7
-6
-5
-4
-3
-2
-1
0
1
2
3
1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013
Cyclical balance – EC Cyclical balance – ESCB
Structural balance – EC Structural balance – ESCB
Total balance

Source: CNB: the CNB forecast is shown in the grey zone.
As a result, the deficit-to-GDP ratio increased by approximately 5 percentage points between
2007 and 2009. The Excessive Deficit Procedure (EDP) was opened for the Czech Republic
in late 2009 (for the second time during its EU membership) with the deadline for correction
of the deficit below 3% of GDP being set at 2013.7 This situation called for an instant and
decisive response by the Czech authorities if the deficits were to be brought back under
control. Such an action came relatively soon; in late 2009 the government approved an
“austerity fiscal package”.
An advantage of the ESCB method of measuring the cyclical balance is that it estimates the
cyclical position separately for each revenue/expenditure item. As a result, the individual
cyclical components of revenues may in this method go in both positive and negative
directions within a single time period (of one year). By contrast, in the EC method, all
revenue/expenditure items always move in the same direction (positive or negative). This is
because, in this method, the output gap derived from a production function is a single
measure of the position of the economy within the business cycle and hence the only driver
of the cyclical part of the revenues/expenditures.
Chart 2 shows a decomposition of the Czech Republic’s cyclical balance into five major
parts. Despite the different methodology, both approaches present roughly the same picture,
indicating that the economic cycle influences the government budget balance mainly via
revenues from corporate income tax and social security contributions. During the crisis, a
sudden shortfall of budgeted revenues occurred especially in 2009 (with a more pronounced
decline estimated by the EC method), but also in 2010.

7 The EDP for the Czech Republic was opened just after the country became an EU member in 2004, and then
abrogated in 2008. It is also worth noting that the Czech Republic has as yet never reached its Medium-Term
Objective (MTO), which is set at 1 % of GDP for the structural deficit of public budgets.

168 BIS Papers No 67

Chart 2
Components of cyclical balance
% of GDP
Aggregated approach (EC method) Disaggregated approach (ESCB method)
-1.0
-0.5
0.0
0.5
1.0
1.5
2.0
1996 1998 2000 2002 2004 2006 2008 2010 2012
personal income tax social security contributions corporate income tax
indirect taxes unemployment expenditure cyclical component of govt. balance
-1.0
-0.5
0.0
0.5
1.0
1.5
2.0
1996 1998 2000 2002 2004 2006 2008 2010 2012

Source: CNB: the CNB forecast is shown in the grey zone.
Two important observations can be made from the cyclically adjusted developments of fiscal
deficits. First, that the Czech public finances are characterised by persistent government
deficits, which have a predominantly structural character. The second distinctive feature is
that Czech fiscal policy has usually been procyclical, especially in the years of economic
boom when extra revenues were typically spent, and unfortunately also in the current period
when there is a need for fiscal consolidation during the time of economic slowdown.
4. Fiscal policy in the CNB forecast
The CNB’s fiscal forecast, which is independent from that of the Ministry of Finance, is an
integral part of the CNB’s quarterly macroeconomic forecast. The medium-term
macroeconomic forecast serves as a main input into the CNB’s board monetary policy
decision-making.8 It includes a forecast of interest rate and exchange rate trajectories. A core
model plays a key role in the preparation of the forecast. Since summer 2008, the “g3” model
(a structural DSGE type of model) has taken over as the CNB’s core model, replacing the
previously used QPM model (which was a relatively small-size gap model).9
There are several interfaces between the fiscal outlook and the core model within the
projection exercise.10 The process starts with the quantification of the demand-side fiscal
impulse, which measures the impact of fiscal policy on GDP dynamics. The fiscal impulse is
derived by fiscal experts based on government plans for both the revenue and expenditure

8 See Czech National Bank (2003) or Král (2005) for more information about the organisation and properties of
the CNB’s forecasting and policy analysis system.
9 Andrle et al (2009) describe the implementation of the “g3” model into the CNB’s forecasting process.
10 The actual core model has no fully fledged fiscal block within its structure as yet and it is therefore dependent
on expert inputs concerning fiscal policy.

BIS Papers No 67 169

sides of the public budget using two alternative methods. The bottom-up approach derives
the impulse by summarising individual revenue and expenditure budgetary measures
expressed as a share of nominal GDP, which is then multiplied by the estimated value of the
fiscal multiplier (at 0.6) to derive the impact on GDP growth dynamics. The top-down
approach (serving as an ex post consistency check) infers the fiscal impulse from the fiscal
position which is defined as a year-on-year change in the structural deficit-to-GDP ratio
expressed in percentage points. The structural deficit is estimated as the average of the two
methods described in the previous chapter. The fiscal impulse in this approach is then
computed by multiplying the fiscal position by the fiscal multiplier.
A final estimate of the fiscal impulse is incorporated into the forecast by influencing core
model mechanisms that describe the behaviour of private consumption (via the savings rate),
investments (via the cost of funds), the exchange rate (reflecting the country risk premium
related inter alia to public indebtedness) and trends in productivity and technology (related to
preferences, institutions and rigidities with respect to the size, features and efficiency of the
public sector). Government and household consumption are the most significant expenditure
items that are most influenced by governmental decision-making. The core model directly
incorporates the expert outlook for government consumption as well as the anticipated
primary effects of indirect taxation changes on inflation (see Chart 3). The CNB distinguishes
between the primary and secondary effects of taxation changes because it does not react to
the primary effects when setting interest rates. Besides the government consumption
forecast, the fiscal unit provides the forecasting team with its outlook for some specific items
of public expenditures such as social benefits, the public sector wage bill and government
investment expenditures. These figures are used in the next stages of the forecasting
process when preparing the final disaggregated macroeconomic story, which contains details
beyond what the core model structure makes available (eg the disposable income of
households, average wage and wage bill in the non-profit sphere etc).
Chart 3
Fiscal outlook in the CNB’s macroeconomic forecast

The completion of the macroeconomic forecast and the fiscal outlook is an iterative process.
After the (draft) macroeconomic forecast is completed (having incorporated all fiscal inputs
mentioned above), the forecast of direct and indirect tax revenues and social security/public
healthcare insurance contributions are computed. To do so, the fiscal experts make use of
the labour market outlook, the prospect for GDP and its structure, estimated profits of firms
etc. After that, the public budget deficit (and government debt) and its structural component
Fiscal Impulse Quantification
Government Consumption
Primary Effects of Tax
Changes
Core Model Forecast
Bottom-up Top-down

170 BIS Papers No 67

are derived with the latter providing an important ex post consistency check of the underlying
fiscal stance arising from the mutual interaction between the fiscal side and the real
economy. If necessary, a few rounds of iterations take place between fiscal and macro
experts during the projection exercise to deliver a consistent economic story.
5. Summary and conclusions
The recent financial crisis highlighted the need for a better understanding of interactions
between fiscal and monetary policy. These interactions are complex and their in-depth
description and analysis is out of this paper’s scope. Instead, we provide some partial
insights into these interactions from the perspective of the Czech National Bank.
The first insight is on the impact of fiscal policy on the interest rate channel of monetary
policy transmission. We show that government bond yields are an important determinant of
client long-term interest rates in the Czech Republic. However, during the financial crisis, the
relationship between the client interest rates and bond yields has weakened with the
exception of interest rates on mortgage loans.
The second insight concerns alternative methods for the cyclical adjustment of the fiscal
balance, which might lead to different assessments of the fiscal policy stance. We present
two distinct methods, one used by the European Commission and another one used by the
European System of Central Banks. The key difference between these two methods is that,
while the first method is based on output gap calculations, the second takes into account the
different cyclical behaviour of the main revenue and expenditure categories. Both methods
provide almost identical estimations of the cyclical component when applied to the Czech
data, with the exceptions being 2003 and 2009. Both methods also point to the existence of
persistent government deficits that have a predominantly structural nature, and to the
procyclicality of Czech fiscal policy in most years.
The final insight relates to how fiscal policy is incorporated in the CNB’s forecast. The CNB
fiscal forecast is an integral part of the CNB’s medium-term macroeconomic forecasts. The
key fiscal variable, ie the fiscal impulse, is derived using two approaches – bottom-up and
top-down. The fiscal impulse influences private consumption, investments, the exchange rate
and trends in productivity and technology. In addition to the fiscal impulse, the inputs to the
CNB’s macroeconomic forecasts include government consumption and the primary effects of
indirect taxation changes on inflation.
References
Alexopoulou, I., I. Bunda and A. Ferrando (2009): Determinants of Government Bond
Spreads in New EU Countries, ECB Working Paper No. 1093.
Andrle, M., T. Hlédik, O. Kameník aQG� -�� 9OþHN� �������� ,PSOHPHQWLQJ� WKH� 1HZ� 6WUXFWXUDO�
Model of the Czech National Bank, CNB Working Paper 2/2009, Prague, October 2009.
Baldacci, E., S. Gupta and A. Mati (2008): Is it (Still) Mostly Fiscal? Determinants of
Sovereign Spreads in Emerging Markets, IMF Working Paper No. 08/259.
Barro, R. (1979): On the Determination of Public Debt, Journal of Political Economy,
Vol. 87 (5), pp. 940–971.
Blinder, A. (1982): Issues in the Coordination of Monetary and Fiscal Policy, NBER Working
Paper No. 982.

BIS Papers No 67 171

Bouthevillain, C., P. Cour-Thimann, G. van den Dool, P. de Cos, G. Langenus, M. Mohr,
S. Momigliano and M. Tujula (2001): Cyclically adjusted budget balances: an alternative
approach, European Central Bank Working Paper No. 77.
Chari, V., L. Christiano and P. Kehoe (1991): Optimal Fiscal and Monetary Policy: Some
Recent Results, Journal of Money, Credit and Banking , Vol. 23, No. 3, Part 2: Price Stability
(August 1991), pp. 519–539.
Czech National Bank (2003): The Czech National Bank’s Forecasting and Policy Analysis
System. Edited byW. Coats, D. Laxton, and D. Rose. Prague, February 2003.
Davig T. and E. Leeper (2011): Monetary-Fiscal Policy Interactions and Fiscal Stimulus,
European Economic Review, Vol. 55(2), pp. 211–227.
‘XPLþLþ�� 0�� DQG� 7�5LG]DN� �������� ‘eterminants of sovereign risk premia for European
emerging markets, Financial Theory and Practice, Vol. 35 (3), pp. 277–299.
Ferrucci, G. (2003): Empirical determinants of emerging market economies’ sovereign bond
spreads, Bank of England Working Paper No. 205.
Kirsanova T., C. Leith and S. Wren-Lewis (2009): Monetary and Fiscal Policy Interaction:
The Current Consensus Assignment in the Light of Recent Developments, Economic
Journal, Vol. 119 (541), pp 482–496.
Král, P. (2005): Forecasting and Policy Analysis System in the Czech National Bank. In:
Practical Experience with Inflation Targeting, Volume of International Conference held at the
Czech National Bank, May 13 – 14, Prague 2004.
Larch, M. and A. Turrini (2009): The Cyclically Adjusted Budget Balance in EU Fiscal
Policymaking. Love at First Sight Turned into a Mature Relationship, Intereconomics,
Vol. 45 (1), pp. 48–60.
López, E., E. Muñoz. and Riquelme V. (2011): Long Term Interest Rate and Fiscal Policy,
Working Papers Central Bank of Chile 633, Central Bank of Chile.
Lucas, R. and N. Stokey (1983): Optimal fiscal and monetary policy in an economy without
capital, Journal of Monetary Economics, Vol. 12(1), pp 55–93.
Sargent, T. and N. Wallace (1981): Some unpleasant monetarist arithmetic, Quarterly
Review, Federal Reserve Bank of Minneapolis, issue Fall.
Schmitt-Grohe, S. and M. Uribe (2004): Optimal Fiscal and Monetary Policy under Sticky
Prices, Journal of Economic Theory, Vol. 114, pp.198–230.
Tabellini, G. (1986): Money, Debt and Deficits in a Dynamic Game, Journal of Economic
Dynamics and Control, Vol. 10, pp. 427–42.
Woodford, M. (2003): Interest and Prices: Foundations of a Theory of Monetary Policy,
Princeton University Press.
Zoli, E. (2005): How does Fiscal Policy Affect Monetary Policy in Emerging Market
Countries?, BIS Working Paper No. 174.

BIS Papers No 67 173

The importance of fiscal prudence under the Linked
Exchange Rate System in Hong Kong SAR
Hong Kong Monetary Authority
Abstract
The Hong Kong SAR has consistently pursued a prudent fiscal policy. Substantial fiscal
reserves have insulated government funding from the volatility of financial market conditions,
and have buffered the economy against shocks, particularly in the absence of a discretionary
monetary policy. Sound fiscal management has also reinforced the credibility of the Linked
Exchange Rate system.

Keywords: Foreign exchange reserves, peg, deficit, fiscal policy
JEL classification: E52, F31, H62

174 BIS Papers No 67

Fiscal position of the Hong Kong government
Hong Kong has an impressive track record of fiscal prudence. The government has
made significant efforts over the years to observe the provisions in the Basic Law1 that call
for a balanced budget over the medium term and require the pace of spending increases to
be commensurate with the GDP growth rate. A case in point is the experience during the
early 2000s, when the government undertook a series of major fiscal reforms (mainly in the
form of spending cuts) in response to a decline in revenues associated with a protracted
economic downturn. With the bursting of US dotcom bubble and the subsequent SARS
epidemic exacerbating the painful adjustment in the aftermath of the 1997–98 Asian financial
crisis, Hong Kong suffered a long period of deflation and anaemic real growth, with nominal
GDP not returning to the 1997 peak level until 2005. However, the government’s reform
efforts helped to keep the fiscal deficits at a relatively low level, and eventually to restore the
budget balance to surplus in 2005 (Chart 1). Even during the 2008–09 global financial crisis,
the government continued to register a small surplus, thanks to a tight control on
expenditure. For the current 2011/12 fiscal year, the budget projects a small deficit, driven
mainly by one-off relief measures.2
Chart 1
Hong Kong’s fiscal performance
and nominal GDP growth
-8
-6
-4
-2
0
2
4
6
8
10
12
19
97
19
98
19
99
20
00
20
01
20
02
20
03
20
04
20
05
20
06
20
07
20
08
20
09
20
10
10.0
12.0
14.0
16.0
18.0
20.0
22.0
24.0Nominal GDP growth (lhs)
Government revenue (rhs)
Government expenditure (rhs)
% of GDP% change

The sustained fiscal discipline has yielded a sizeable accumulation of fiscal reserves.
Unlike most fiscal authorities internationally, the Hong Kong government is a net creditor,
with more financial resources at its disposal than the liabilities it owes to the public.
Outstanding general government gross debt to finance fiscal operations totalled only about
HK$11 billion (or 0.6% of GDP),3 while the pool of fiscal reserves stood at HK$595 billion (or
one third of GDP) as of March 2011. The reserves are large enough to cover about
24 months of government expenditure. Given the considerable fiscal headroom, along with

1 The Basic Law serves as the constitutional document of post-colonial Hong Kong SAR.
2 It now seems likely, however, that the government will record a surplus in 2011/12, partly due to higher-than-
expected revenue collections.
3 This debt was issued in 2004 and 2005, complementing the drawdown from the fiscal reserves. The
government also issues bonds under the Government Bond Programme to promote development of the local
bond market. Proceeds from this programme are credited to a separately managed bond fund and are not
used for fiscal operations. As at the end of October 2011, outstanding bonds under the programme amounted
to HK$43 billion (or 2.4% of GDP).

BIS Papers No 67 175

Hong Kong’s other favourable economic and institutional fundamentals, the government’s
creditworthiness has been strongly endorsed by the rating agencies, with long-term credit
ratings of AAA from Standard and Poor’s and Aa1 from Moody’s.
The need for strong fiscal reserves to cope with shocks
Having adequate fiscal scope for cushioning economic downturns is particularly
important for economies with a fixed exchange rate regime and perfect cross-border
capital mobility, such as Hong Kong. Under a fixed exchange rate regime and with an
open financial account, the monetary authorities cannot pursue an independent monetary
policy, thus leaving fiscal policy, possibly supplemented by other policy tools including
macroprudential measures, as the key levers for managing economic cycles and mitigating
the impact of shocks on the economy. The availability of a sizeable pool of fiscal reserves
provides the government with sufficient financial resources to implement a countercyclical
expansionary fiscal policy during economic downturns, without having to worry about
potentially volatile funding conditions in the government debt market. The unfolding
European sovereign debt crisis has clearly demonstrated how heavy reliance on debt can
expose governments to swings in market sentiment.4
The experience in the years following the Asian financial crisis underscores the role of
strong fiscal reserves in supporting Hong Kong’s economic stability. Due to falling
income in the private sector and sluggish asset markets amid a series of adverse shocks
including the Asian financial crisis, the bursting of the US dotcom bubble and the SARS
epidemic, government revenue dropped by more than 35% in nominal terms between
FY1997/98 and FY2002/03. The marked revenue loss contributed to consecutive years of
budget deficits, from FY1998/99 to FY2003/04 (with the brief exception of FY1999/2000).
While reforms were undertaken to shore up the fiscal position, the needed fiscal adjustment
would have been much larger – and hence the fiscal support for the economy much weaker
– if not for the option to deploy the existing fiscal reserves to finance public spending. A total
of HK$182 billion, or 40% of the fiscal reserves, was drawn down during the period, which
represented about 15% of Hong Kong’s annual GDP at the time.
In addition, the fiscal reserves can contribute to maintaining financial stability in Hong
Kong at times of heightened market stress, as the 2008–09 global financial crisis
illustrates. As in the economic downturn during the early 2000s, the available fiscal
headroom enabled the government to effect significant relief measures – which totalled about
6.3% of annual GDP from 2008/9 to 2010/11 – to bolster the economy against the fallout
from the global financial crisis. As importantly, at the depth of the crisis when concerns about
bank soundness arose, to maintain confidence the government adopted full deposit
guarantees and standby measures to backstop banks’ recapitalisation needs. The credibility
of these policy actions was underpinned by the strong financial backing of the Exchange
Fund, of which the fiscal reserves constitute a major part.5 In other words, the fiscal reserves
substantially augment the government’s resources in supporting the financial sector as
needed during times of extraordinary distress.

4 In addition, for a small economy whose currency is not a major reserve currency, for example Hong Kong, the
local debt market tends to be less liquid and could experience particularly acute sell-off pressures in times of
global stress when investors’ demand for liquid assets surges. In Hong Kong, yields on Exchange Fund bills
rose sharply vis-à-vis their US counterparts during the 2008–09 crisis.
5 The fiscal reserves account for about one quarter of the Exchange Fund assets.

176 BIS Papers No 67

The importance of building up strong fiscal reserves in Hong Kong also reflects
volatile government revenue and future ageing-related fiscal pressures.
• The fluctuations in Hong Kong’s government revenue are among the most extreme
in the world (Porter (2007)).6 This is partly due to the openness of the economy and
its susceptibility to external shocks, but it also reflects a revenue regime that has a
narrow tax base and relies heavily on non-tax income. Taxes are based either on
earnings (not on more stable consumption spending) or dependent on asset market
activities (where stamp duties apply). Also, the volatile non-tax asset income
(comprising the proceeds from land sales and investment income in Hong Kong)
accounts for around 20% of government revenue, one of the highest levels in Asia.
The large fluctuation in fiscal receipts amplifies the importance of maintaining strong
fiscal reserves to support steady government spending.
• The government has large potential contingent liabilities associated with the future
ageing of the population. In the case of Hong Kong, the expected rise in health care
spending due to ageing will put pressure on public finances as health care services
are largely government-funded.7 The availability of fiscal reserves could ensure a
smooth transition for any necessary structural reforms in the revenue regime.
The role of fiscal reserves in fortifying the Linked Exchange Rate system
The government’s fiscal discipline has been a cornerstone of long-term monetary
stability in Hong Kong. As the government holds a substantial amount of net financial
assets, there are few concerns that the Hong Kong Monetary Authority would ever be forced
into monetising government debt, thus undermining the currency board arrangement.
Separately, under the fixed exchange rate, any overheating pressures in the economy would
have to be absorbed solely by domestic price adjustment, thus likely resulting in high
inflation. However, the government’s prudent approach to spending reduces the risk that the
economy will face overly strong demand pressures, thereby helping to support a low level of
inflation in Hong Kong. The Argentina debt crisis in the late 1990s that led to the collapse of
the country’s exchange rate peg is a notable example of the importance of fiscal discipline in
ensuring the feasibility of a currency board system (see, eg, IMF (2003)).8
The absence of government borrowing would also help alleviate the likely interest rate
adjustment in crisis scenarios. In the event of a crisis, money markets typically come
under pressure, causing interbank interest rates to surge. Heavy government borrowing
compounds the liquidity pressures and pushes interest rates even higher, thus exacerbating
the economic downturn and associated asset market correction. The ensuing downturn in
confidence intensifies adverse sentiments towards the local currency. If, on the other hand,
the government has no borrowing requirement, the Hong Kong dollar will experience that
much less downward pressure during a crisis.
More importantly, the fiscal reserves boost the financial resources available to defend
the Linked Exchange Rate system. The fiscal reserves placed with the HKMA, which
constitute about one quarter of the total Exchange Fund, expand the resources that can be
called upon to defend the Hong Kong dollar against speculative attack or a sudden reversal

6 N Porter, “Guarding against fiscal risks in Hong Kong SAR”, IMF Working Paper, no 07/150, 2007.
7 There are estimates that fiscal reserves totalling up to 30% of GDP may be needed by 2030 to anticipate the
ageing-related budget pressures in Hong Kong (Porter (2007)).
8 “Lessons from the crisis in Argentina”, International Monetary Fund (IMF), October 2003, pp 8–11.

BIS Papers No 67 177

in capital flows. In August 1998, to counteract market manipulations and defend the Linked
Exchange Rate system, in addition to dispensing the foreign reserves to meet the Hong
Kong dollar sell-off pressures, the government and the HKMA drew on the Exchange Fund in
an unprecedented operation in the stock and futures markets that involved stock purchases
totalling HK$118 billion. The support of the fiscal reserves alleviated the constraints on the
possible scale of the operation, thus improving the odds that the Hong Kong dollar could be
effectively defended. While the technical refinements undertaken over the years have greatly
improved the robustness of the currency board arrangement,9 the additional financial
resources conferred by the fiscal reserves would certainly still be useful in extreme
circumstances to support our exchange rate regime.

9 These refinements include the formal establishment of two-sided convertibility undertakings and convertibility
zone, and the introduction of a discount window.

BIS Papers No 67 179

The impact of public debt on foreign exchange reserves
and central bank profitability: the case of Hungary
Gergely Baksay, Ferenc Karvalits and Zsolt Kuti1
Abstract
This paper focuses on the interactions between public debt policy and foreign exchange
reserve management. We found that, although foreign currency debt issuance can contribute
significantly to the growth of foreign exchange reserves, it can cause serious difficulties in the
assessment of reserve adequacy, especially during crisis periods. Furthermore, it affects the
profit-loss of the central bank. On the other hand, the accumulation of foreign exchange
reserves may affect the public deficit and debt as well.
Based on these observations, we draw several lessons. We conclude that debt management
policy may result in a suboptimal solution on a consolidated basis if the needs of reserve
adequacy are not taken into account within the decision-making process for foreign currency
debt issuance. In addition, we argue that, if the central bank wants to enhance its capacity to
intervene during a crisis, it should seek to identify and utilise other sources of foreign
exchange liquidity. But the options here are limited: we believe that the most appropriate tool
that would enable a central bank such as the MNB to rapidly obtain an ample amount of
foreign exchange reserves is a foreign exchange swap line provided by a developed country
central bank.

Keywords: Monetary policy, fiscal policy, public debt management, national budget,
sovereign debt, foreign exchange reserve
JEL classification: E52, E58, E62, E63, H63

1 Gergely Baksay is Senior Analyst, Ferenc Karvalits is Deputy Governor and Zsolt Kuti is Senior Economist.

180 BIS Papers No 67

Introduction
Through various channels, the amount and structure of public debt can have a significant
influence on a central bank’s foreign exchange reserve management. On the one hand, the
issuance of foreign currency-denominated debt can boost international reserves. On the
other hand, repayment of public foreign currency debt not only reduces the level of foreign
exchange reserves but can cause transient problems in liquidity management. Furthermore,
the dynamics of public debt influences not only reserve accumulation but may affect the
central bank’s reserve adequacy targets, as an increased level of foreign debt can push up
the reserves, requirement, depending mainly on the maturity structure of public assets held
by non-residents.
The actual level of reserves may also set in motion forces that interact with public debt.
Inadequate foreign exchange reserves would call for an increase in foreign currency debt or
would lead to changes in market perception about the sustainability of the debt. Such
changes would affect the fiscal deficit not only directly through interest costs, but indirectly
through the central bank’s profit and loss. The cost to the central bank of sterilising excess
liquidity in the domestic money market is likely to increase, given that there is a significant
spread between the cost of sterilisation and the yield on foreign exchange reserves.
This paper focuses on how public debt policy and foreign exchange reserve management
have interacted in Hungary. The massive increase in the country’s foreign currency debt and
the changes in reserves in the past decade offer several important lessons. We describe how
the central bank’s room for manoeuvre in reserve accumulation can be constrained by debt
and exchange rate considerations. We evaluate the most important components of such
constraints. We also demonstrate how various state agencies may have diverging goals
related to public debt, and how the potential conflict of interest between these goals can
influence preferences that are reflected in the assessment of reserve adequacy.
The paper proceeds as follows: Section 2 outlines the development of foreign exchange-
denominated public debt in Hungary. Section 3 investigates the effect of this increase on the
international reserves. Section 4 examines the impact on reserves requirements while
Section 5 describes the effects on the central bank’s profit and loss. Section 6 summarises
the most important policy lessons and Section 7 concludes.
The role of foreign currency-denominated debt in Hungary’s public
finances
There is an extensive literature on the benefits or desirability of foreign currency-
denominated public debt. The most important potential benefits of foreign currency debt
include access to a larger investor base, less crowding-out in domestic markets, lower yields
on foreign exchange issuance, access to longer maturities, and the possibility of building up
official foreign exchange reserves and improving short-term stability in hard times.
But foreign currency financing has risks. These include more stringent constraints on
redeeming foreign currency debt relative to debt denominated in the domestic currency,
which can increase the rollover risk. Also, large-scale foreign exchange issuance can
increase the country’s external vulnerability as perceived by investors and credit rating
agencies. Finally, a significant depreciation of the domestic currency may significantly
increase the interest burden as calculated in that currency.2

2 Wolswijk and de Haan (2005) summarise the related literature, where empirical studies suggest that smaller
economies tend to take on more foreign currency debt (Claessens et al (2003)), and the decision seems to be

BIS Papers No 67 181

In Hungary, the Debt Management Agency (ÁKK) incorporated these considerations into a
quantitative model for cost/benefit and risk analysis. They used the model to construct a
reference band for the foreign exchange share of public debt, at 25–32% of total government
debt in 2004. This seemed to be a reasonable choice at Hungary’s pre-crisis public debt level
of 60–67% of GDP. Like many other emerging market economies, Hungary was short of
domestic private savings to cover the government’s large financing requirement. Thus, it was
vital to attract foreign investors. Large-scale issues of foreign currency-denominated bonds
were inevitable because investors were reluctant to purchase domestic currency government
paper in the necessary amounts. According to ÁKK, the benefits of foreign debt would offset
the additional costs and risks in the target range and the annual issuance of foreign currency
debt was broadly in line with the central bank’s foreign exchange reserve target.
Prior to the crisis, the actual share of foreign currency debt remained at the planned levels.
However, the impact of the global turmoil in late 2008 enforced a radical change in debt
management, and the actual share of foreign currency debt jumped to more than 40% of
total debt. After the collapse of Lehman Brothers, the demand for HUF government bonds
plummeted and AKK was forced to suspend primary issuances. Medium- and long-term HUF
bond issuance was suspended between 22 October 2008 and 12 February 2009, when it
was restarted on a smaller scale, returning to normal levels only around July 2009.
To cover its financing needs, the government, together with the central bank, requested an
EU/IMF standby agreement in November 2008. The financing provided by this programme
practically replaced the government’s local currency financing with foreign exchange-
denominated debt in 2009. The EUR 20 billion arrangement served three goals beyond
supporting the financing of the balance of payments. It (i) helped to increase the foreign
exchange reserves of the central bank; (ii) covered the government’s financing needs; and
(iii) it also played a substantial part in stabilising Hungary’s financial system. The government
drew down EUR 12.9 billion over the course of the programme, while the central bank
withdrew an additional EUR 1.4 billion to replenish foreign exchange reserves without
increasing the public debt.
The most direct consequence of the EU/IMF agreement was that gross government debt
surged in 2008, in spite of the historically favourable deficit of 3.7% of GDP. As the
drawdowns were front-loaded, and part of the tranches were deposited at the central bank,
or were onlent to support domestic financial institutions, gross debt increased by more than
the government’s financing needs.3
At the same time the foreign exchange share of public debt exceeded its ceiling, reaching
44.7% by the end of 2009. This radical change was clearly due to practical considerations,
and was not preceded by the overhaul of the debt management strategy. However, the
sudden increase in foreign exchange debt had a substantial impact on interest expenditure,
short-term external debt, foreign exchange reserves, the central bank’s profit and loss, and
the structural liquidity surplus on the interbank money market, as presented in the following
sections.

motivated mainly by practical considerations especially the expected cost of foreign currency debt (Pecchi and
Di Meana (1998)).
3 The front-loaded pattern of the loans only slightly affected the medium-term level of debt, because the excess
withdrawals were gradually used for government financing at a later stage. Thus the debt converged to a level
derivable from a smoother theoretical debt issuance schedule.

182 BIS Papers No 67

Graph 1
Central government debt and foreign exchange reserves
In billions of euros

The effect of foreign currency-denominated public debt on foreign
reserve dynamics
The financial crisis caused significant changes in the structure of the Hungarian foreign
exchange reserves. Before the crisis, reserve levels increased only slowly in line with the
country’s short-term debt dynamics. The authorities preferred not to hold any buffer above
this precautionary level. However, after the crisis deepened, the level of international
reserves doubled in three years. This section describes how the increase of foreign currency
debt contributed to this process and what kind of side effects and constraints arose in the
management of foreign exchange reserves.
Foreign currency debt issuance can contribute significantly to the growth of foreign
exchange reserves
If the central bank manages the government’s transaction account, new foreign exchange
issuances boost the foreign exchange reserve level immediately. If the conversion of the
government’s foreign exchange balances to local currency are also effected by the central
bank, then the additional reserves stay with the central bank even after the government
starts spending the local currency equivalent of the foreign exchange issuance.
In Hungary, the increase in foreign currency debt issuance and the EU/IMF loan were the
most important factors behind the growth of foreign exchange reserves (see Figure 2). In the
last 10 years net debt issues contributed almost EUR 22 billion to the level of reserves. Such
a large external influence has a serious impact on the central bank’s ability to autonomously
determine the desired level of foreign currency reserves.

BIS Papers No 67 183

Graph 2
Growth of foreign exchange reserves and
foreign currency debt issuance
Cumulated changes
-15
-5
5
15
25
35
45
-15
-5
5
15
25
35
45
2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013
EUR BnEUR Bn
Net FX bond and loan issuance IMF – EU loan Other
EU transfers Net interest income Net increase of FX reserves
Forecast

Foreign currency issuances can fall short in times of market stress, resulting in a
lower-than-expected level of foreign exchange reserves
Evidently, foreign currency debt issuance can be a continuous net contributor to reserve
growth only if new issues exceed the amounts maturing in a given year. At times of crisis this
might be difficult to achieve, especially for countries with weak economic fundamentals. Even
to renew maturing debt can be hard and, even if market access is possible, the increased
funding cost can be punitive. In the worst case of a sudden stop, the country is unable to
obtain market funding at any rate.
As a consequence, market debt issuance usually cannot work as an automatic stabiliser for
foreign exchange reserves. Since the expected level of foreign reserves generally increases
in times of stress, any shortfall in foreign currency funding can exacerbate the problem of
inadequate foreign currency reserves.
This problem can be detected in the reserve dynamics of Hungary (see Graph 3). Until 2007
the impact of the debt on reserves was quite limited: the relatively small amount of foreign
currency debt issuance went hand in hand with the slow growth of international reserves. In
2008, however, when an increasing level of short-term debt and the harsher investment
climate called for higher level of foreign reserves, net foreign exchange bond issuances
could not be increased due to the worsening of market conditions.
Finally, the IMF/EU loan resulted in a rapid increase of foreign currency reserves,
contributing some EUR 6–7 billion to reserve growth in both 2008 and 2009. Repayment of
the IMF/EU loan started in the last quarter of 2011 and will peak in 2012–14. This schedule
will compel Hungary to issue foreign currency debt in larger amounts than it did in the pre-
crisis period. This elevated level of foreign currency funding is required if the IMF/EU funds
used to increase central bank reserves are to be replaced.

184 BIS Papers No 67

Graph 3
Growth of foreign exchange reserves and
foreign currency debt issuance
Yearly changes

Hedging foreign exchange risk can cause relatively large swings in foreign currency
reserves
In addition to the net foreign currency debt issues there is another important factor that can
have a large direct impact on the level of foreign exchange reserves, namely the flows
related to the margin accounts of derivative positions held by the public sector.
As a policy matter, the Hungarian debt management agency only accepts EUR/HUF risk on
its foreign currency debt. As a significant part of the foreign currency issuance occurs in
currencies other than the euro, such as the US dollar or Japanese yen, the state runs a large
FX swap book that is used to hedge the cash flows of non-euro denominated foreign
currency bonds into euros.
As the market rate of the euro changes, the swap counterparties evaluate their positions and
adjust collateral as necessary. The Hungarian debt management agency must pay out cash
collateral to its counterparties when the euro appreciates against the other foreign currency
while, if the euro weakens, its counterparties will pay out. Since these flows are directly
debited or credited to the debt agency’s foreign currency account kept at the central bank,
the level of foreign currency reserves changes accordingly.
Before 2008 these flows did not significantly influence the level of Hungarian reserves. From
that year, however, these flows have contributed significantly to the volatility of the foreign
exchange reserves. The crisis has resulted in a much more volatile exchange rate
environment, increasing the need for collateralisation. Furthermore, the currency structure of
Hungary’s issues has changed following the issue of a USD 3.75 billion dollar-denominated
bond in early 2011. These three factors have contributed to the increasing volatility of margin
call flows: before 2010 even the largest quarterly changes in the level of net cash collateral
did not exceed EUR 200 million whereas in the last two years there have been quarters
when the net flows reached EUR 1 billion.

BIS Papers No 67 185

The effect of foreign currency-denominated public debt on the required
level of foreign reserves
The size and structure of foreign currency-denominated public debt also influences the
necessary or optimal level of foreign exchange reserves. We demonstrate this using the well
known Guidotti-Greenspan rule, which states that reserves should cover a country’s short-
term external debt. Several channels exist through which public debt can have an impact on
reserve requirements.
At the end of the 1990s, Hungary’s external short-term debt was limited, and the public
sector had no reliance on short-term external funding (see Graph 4). From 2000 onwards,
the Guidotti-Greenspan rule started to indicate an increasing need for reserves. Up until
2003, the increasing public sector debt was the sole contributor to this process, pushing up
the public sector’s share within short-term external debt to 50%.
After 2003, private sector external indebtedness began to grow apace whereas the public
sector’s short-term debt fluctuated at around EUR 4–5 billion. These patterns have changed
again after the outbreak of the financial crisis: both public and private sector short-term debt
have started to increase significantly. However, the former has outpaced the latter, again
resulting in an increase in public sector’s contribution to short-term debt dynamics.
Graph 4
Public and private sector short-term debt

Based on these tendencies we can conclude as follows:
The total and relative amounts of public debt matter for the reserve requirement
Short-term debt tends to increase in line with the total amount of public debt. Additional
public debt tends to crowd out private companies from local financial markets, pushing them
to seek financing from external sources. Both of these factors lift the Guidotti-Greenspan
indicator, raising the optimal level of reserves.
Furthermore, although international comparisons can provide a wide range of examples, in
general, investors respond to a heavy public debt burden by shifting their activity towards
shorter-term debt with the aim of reducing their interest rate and default exposures. This
shortening process results in a larger reserve requirement, which usually leads to trouble

186 BIS Papers No 67

during a crisis. Such a situation is exacerbated by the liquidity shortages that typically occur
during a crisis episode.
Heavy public indebtedness can cause significant interest rate differentials which can
push up carry trader’s demand for short-term assets
Again, tendencies can vary from country to country, but emerging economies with heavier
public debt burdens and weaker fundamentals usually need to offer higher interest rates on
their public debt. However, high interest rates can attract carry traders especially when the
interest rates offered by the “safe haven” currencies are low and there is abundant liquidity in
the global financial system. Such flows directly push up reserve requirements but, since they
do not contribute to the foreign exchange reserves themselves, reserve adequacy is eroded.
Flows from margin calls can also contribute to the volatility of the reserve requirement
We have already mentioned that margin calls on the state debt management agency’s FX
swap contracts can cause relatively large swings in foreign exchange reserves. But it is
worth emphasising that this can also affect the required level of reserves. All inflows from
margin calls are regarded as short-term funding. If the debt agency becomes a net debtor at
the individual counterparty level, any additional funding obtained from margin calls boosts
both the foreign exchange reserves and the reserve requirement.
Long-term debt, usually considered to be stable, can also be a source of instability
when liquidity needs suddenly increase during sudden sell-offs
Hitherto, we have used the Guidotti-Greenspan rule as a leading indicator to assess the
reserve requirement. However, the experience of the recent crisis suggests that using short-
term debt as an indicator for the foreign exchange reserve requirement can be misleading.
During the crisis many countries, including Hungary, found that the volume of capital outflows
depended much more on the type of capital rather than on its maturity. For example, short-
term parent bank lending to subsidiaries proved to be highly stable whereas foreign portfolio
investors sold large amounts of Hungarian long-term financial assets. Non-resident holdings
of HUF-denominated government bonds declined by approximately 30% in the second half of
2008, which put heavy pressure on the local spot and FX swap currency markets.
The effect of foreign currency-denominated public debt on the central
bank’s profit and loss
A heavy public sector foreign currency debt burden affects the consolidated expenditures
and revenues of the public sector via the net interest expenditures of (i) the government and
(ii) the central bank. We analyse the net effect of these relatively separate channels on a
consolidated level, merging the costs and benefits for both the central bank and the
government.
One of the key arguments for foreign currency-denominated government issuance is the
lower interest rate relative to local currency funding. In the case of Hungary, this was
particularly true for the EU/IMF loan package. While the average actual interest rate on the
EU/IMF loans was 2.9% between 2008 and 2011, the average five-year HUF bond yield was
8.1%. If we consider only that part of the international financial package that was used for
actual government financing, the comparison between the actual interest payments to the
estimated interest expenditure on the HUF bonds that might have been issued instead

BIS Papers No 67 187

suggests that around EUR 0.8 billion was saved thanks to the IMF-EU loans between 2008
and 2011.4
However, the resulting growth in the foreign reserves of the Magyar Nemzeti Bank (MNB)
increased the central bank’s net interest expenditures through the higher cost of sterilisation.
As the government converted the majority of the loans at the central bank to cover its debt
repayment and regular expenditures, it not only raised the MNB’s foreign exchange reserves,
but also increased its HUF liquidity. The MNB sterilises the structural liquidity surplus via the
two-week MNB bill, its main policy instrument. The resulting surge in liquidity considerably
increased the MNB’s interest expenditure, while the higher foreign exchange reserves were
not able to offset this effect on the interest revenue side due to lower foreign exchange
yields. Although the foreign exchange reserves increased somewhat more than the stock of
MNB bills, the higher HUF interest rates imposed a higher overall net interest expenditure on
the central bank. This effect has been augmented by the increasing spread between the
MNB’s policy rate and the yield on its foreign exchange reserves.5
Graph 5
Change of outstanding MNB bills and
its effect on profit/loss of the central bank

On a consolidated level, the balance of government savings and the central bank’s net extra
interest expenditure is likely to be positive because of the high opportunity cost of HUF bond
issuance. When the government covers its financing requirement by foreign currency debt
issuance, and converts the funds raised at the central bank for HUF liquidity, then the
government gains the difference between the interest rate on foreign exchange loans and the
rate on HUF bonds. At the same time, the central bank’s loss amounts to the difference
between the policy rate and the yield on foreign exchange reserves. The average difference
of the aforementioned spreads for the period between November 2008 and November 2011
is around 90 basis points, which means that the government’s gain has so far exceeded the

4 Of course the counterfactual assumption on HUF bond issuance is somewhat unrealistic, as the primary
market dried up totally at beginning of the period.
5 The excess withdrawals were held at the central bank, increasing the interest revenue on FX reserves, but
without any effect on the structural liquidity.

188 BIS Papers No 67

actual losses of the central bank. This gain may decrease, or turn into a loss, once the
EU/IMF loan is repaid and replaced by market issuance.
Graph 6
Actual cost of IMF-EU loans on a consolidated level

Lessons from the viewpoint of foreign reserve management
As we have seen, the level and dynamics of foreign currency public debt can heavily affect
the central bank’s ability to formulate its own strategy on foreign reserves. This section
summarises the most relevant constraints and policy conclusions.
Growing public debt tends to increase the level of uncertainty in the dynamics of
foreign exchange reserves
Foreign currency debt issuance can contribute significantly to the growth of foreign exchange
reserves. Yet, reserve accumulation via this channel is strongly countercyclical: during
periods of abundant market liquidity and low risk awareness, there are practically no
constraints. In bad times, however, when the need to use these reserves arises, an emerging
economy has very limited scope to issue new debt in the necessary amounts. Furthermore,
hot money investors tend to become more active as public debt grows, causing an ever-
increasing volume of short-term funding to flow in and out of the country.
The central bank needs to have a buffer above necessary level of reserves
All the above factors tend to increase the uncertainties in the dynamics of foreign exchange
reserves: that is, reserves are likely to fall below the expected level when sudden and large
shifts occur in reserve requirements. In such a situation, simply targeting the reserve level at
the precautionary level (ie at the level indicated by the Guidotti-Greenspan rule or similar)
can create difficulties, given that the replenishment of reserves will take time and that,
especially during a period of market disorder, the required funds will be difficult to obtain from
the market. This suggests that the central bank will be better off if it maintains an additional
buffer, over and above the precautionary level of reserves.

BIS Papers No 67 189

However, maintaining excess reserves implies extra costs too. The size of these costs is
determined by the differential between the financing cost and the yield on reserve assets.
This gap is usually positive and during an episode of market disorder it tends to widen.
The central bank’s autonomous instruments may not be sufficient to counteract
volatility in reserve levels and reserve requirement
Although foreign currency debt issuance plays crucial role in foreign exchange reserve
growth in Hungary, the central bank has only a limited degree of influence over foreign
currency debt management. The debt management agency may take reserve adequacy
considerations into account, but the major drivers behind its decisions are its own
preferences on the optimal level of foreign currency debt. In theory, the central bank can also
issue foreign currency debt, booking it on its own balance sheet. Statistically speaking, this
would not increase the government’s debt, but there are several disadvantages to this
approach. Thus, this is not common practice internationally.
In Hungary, the central bank has only limited scope for building foreign exchange reserves
through the use of its own instruments. The direct spot purchase of foreign exchange is not
practicable, because of the negative revaluation effect on the private foreign currency debt
stock. Nor is it possible to impose reserve requirements on local banks based on their foreign
exchange liabilities: as the banking system also suffers from a foreign exchange liquidity
shortage, such a measure would increase their participation on the central bank foreign
exchange tenders, thus wiping out any increase in the foreign exchange reserves. The
central bank could, in principle, conduct repo or FX swap deals with large foreign banks, but
these agents would be willing to provide short-term funding only. Thus reserve adequacy
would not be bolstered in this way. FX swap lines extended by developed country central
banks might provide an appropriate tool, but the options for Hungary are rather limited in this
respect.
These considerations suggest that the central bank would face difficulties if it were forced to
quickly replenish foreign exchange reserves. An important lesson of the recent crisis was
that investors expected countries to hold the optimal (ie Guidotti-Greenspan rule-based) level
of foreign exchange reserves even at the height of the crisis. All this implies that central
banks that have not previously built up large reserve buffers may find their capacity for
intervention severely circumscribed in a crisis.
Efficient coordination between government agencies is vital
Serious conflicts of interest can arise between different government institutions (such as the
debt management agency, finance ministry or central bank) which can strongly influence the
evolution of the foreign exchange reserves. For example, changes in the preferences of
policymakers regarding the structure of the public debt can greatly affect the reserves
accumulation process. A particularly crucial decision is the level at which the target ratio of
the foreign component in total debt is determined. Any move to reduce this ratio can easily
conflict with policy targets related to foreign exchange reserves. Furthermore, any early
repayment of foreign currency public debt can have a negative side effect on foreign
exchange reserves.
Clearly, swings in debt management policy can cause major difficulties for the management
of foreign exchange reserves. In addition, if the issue of foreign exchange reserve adequacy
has only a limited weight in the decision-making process, foreign currency debt management
may lead to suboptimal results at the consolidated level. In order to avoid such an outcome
and to optimally coordinate the different interests, we believe that a long-term debt issuance
strategy should be defined in which both the central government and the central bank have a
say in determining the size and the timing of foreign exchange issuance – a strategy which
would also be binding on the debt management agency.

190 BIS Papers No 67

Additional sources of foreign currency liquidity could play an important role
Acknowledging the central bank’s dependence on the government’s foreign currency debt
issuance, the Hungarian central bank seeks to identify and utilise other sources of foreign
currency liquidity from both market and official sources. However, as we have briefly outlined
in Section 6.4, the options are limited.
The only tool that would allow central bank to quickly obtain an ample amount of foreign
exchange reserves is a FX swap line extended by developed country central banks. Based
on the MNB’s previous experience, the existence of a such a line can significantly improve
market sentiment6 even if actual utilisation is limited. Market participants “reward” such
agreements not only because they represent a potential source of foreign exchange liquidity
but also because they are a token of support from a developed country central bank.
The costs and benefits of foreign currency debt should be considered on a
consolidated basis
When the effect of foreign currency debt on the central bank’s net interest expenditure is
taken into account, the overall cost of public debt financing in foreign currency may be
significantly altered. Usually governments pay lower interest rates on their foreign currency
debt than on domestic issuance. This is, of course, one of the most appealing features of
foreign currency financing. On the other hand, the conversion of foreign exchange loans
increases the domestic liquidity which the central bank will likely be obliged to sterilise. When
it does so, the sterilisation cost may partially or totally offset the government’s saving on its
interest payments. If the increased interest expenditure for the central bank causes losses,
the government is ultimately obliged to reimburse that loss, and thus sterilisation costs
should be taken into account when the cost of foreign exchange financing is assessed. The
consolidated outcome depends on the difference between the interest rate on international
loans and the rate on HUF bonds compared to the difference between the policy rate and the
yield on foreign exchange reserves.
If the net interest expenditures of the government and the central bank are consolidated, we
estimate that the cost of the foreign exchange loans is likely to be slightly lower than the
equivalent domestic issuance would be. However, this gain may turn negative in future if
foreign exchange loans are rolled over into new foreign exchange bond issues with higher
yields.
Conclusion
This paper focuses on the interactions between public debt policy and foreign exchange
reserve management. We found that, although foreign currency debt issuance can contribute
significantly to the growth of foreign exchange reserves, it can cause serious difficulties in
assessing reserve adequacy. This is especially the case during a crisis when it becomes
almost impossible to refinance maturing debt at a time when, for various reasons, the
reserve requirement may be rising still further. On the other hand, the accumulation of
foreign exchange reserves may affect the public deficit and debt, both directly and indirectly,
especially if it is implemented through foreign currency debt issuance by the government.
Based on these observations, several lessons could be drawn. Rising public debt tends to
increase the uncertainty in foreign exchange reserve dynamics. The reserve requirement can

6 Markets responded favourably in the case of other countries obtaining an FX swap line from a developed
country central bank: see Aizenman and Pasricha (2009).

BIS Papers No 67 191

fluctuate within a wider range but the central bank has only limited influence on the reserve
accumulation process. As a consequence, the debt management policy may result in a
suboptimal solution on a consolidated basis if the needs of reserve adequacy are not taken
into account within the decision-making process on foreign currency debt issuance. To avoid
such negative side effects, we believe that a long-term debt issuance strategy should be
defined where both the central government and the central bank have a say in determining
the size and the timing of foreign exchange issuance. Further, if the central bank wants to
enhance its capacity to intervene during a crisis, it should seek to identify and utilise other
sources of foreign exchange liquidity. But the options here are limited: we believe that the
most appropriate tool that would enable a central bank such as the MNB to rapidly obtain an
ample amount of foreign exchange reserves is an FX swap line provided by a developed
country central bank.
References
Aizenman, J., and Gurnain Kaur P. (2009): “Selective swap arrangements and the global
financial crisis: analysis and interpretation”, NBER Working Papers, 14821.
Wolswijk, G and de Haan, J (2005): “Government debt management in the euro area. Recent
theoretical developments and changes in practices”, ECB Occasional Paper Series, No. 25,
March 2005.
Eichengreen, B., Hausmann, R., and Panizza, U., (2003): “Currency mismatches, debt
intolerance, and original sin: why they are not the same and why it matters”, NBER Working
Paper, 10036.
Eichengreen, B., Hausmann, R. (1999): “Exchange rates and financial fragility”, NBER
Working Paper, 7418.
de Fontenay, P., Milesi-Ferretti, G. M., Pill, H. (1995): “The role of foreign currency debt in
public debt management”, IMF Working Paper, 1995/21.
Claessens, S., D. Klingebiel and S. Schmukler (2003): “Government bonds in domestic and
foreign currency: the role of macroeconomic and institutional factors”, CEPR Discussion
Paper, No. 3789.
Pecchi, L., and A. di Meana (1998): “Public foreign currency debt: a cross-country evaluation
of competing theories”, Giornale degli Economisti e Annali di Economia, No. 2, September.

BIS Papers No 67 193

Sovereign debt management in India:
interaction with monetary policy
R Gandhi1
Abstract
India’s expansionary fiscal policy during the recent crisis resulted in higher government
borrowing through 2008–09 and 2009–10. This borrowing requirement came in about 83%
above the budget estimate in 2008–09, and 65% above the previous year in 2009–10. The
debt-to-GDP ratio rose from 69% before the recent global financial crisis to 73% in 2010,
creating a severe challenge for the Reserve Bank of India (RBI) in meeting the public
borrowing requirement without causing market disruption. To hold borrowing costs down
while scheduling issue maturities so that rollover risk was kept to a minimum, the RBI
followed a multi-pronged strategy.
The potential for interaction between public debt management and monetary policy has
undoubtedly increased during the recent global crisis. This is due to the increase in short-
term debt, which can jeopardise both the signalling of monetary policy and its transmission.
India’s particular dilemma, however, was related to systemic liquidity, ie the system would
preferably be in deficit for monetary policy transmission whereas a system in surplus would
be more favourable for debt management. The RBI has resolved this dilemma by putting in
place a monetary policy operating framework whereby the system is ideally allowed to be in
deficit (or surplus) to the extent of the frictional component ie 1% (+/-) of the banking
system’s net demand and time liabilities (NDTL). In this setup, the structural liquidity deficit
(or surplus) is met through OMOs and adjustments in the cash reserves.
Against the background of the increased interaction between sovereign debt management
(SDM) and monetary policy, two important issues urgently need to be addressed. These are:
(i) to ensure seamless coordination between SDM and monetary policy, especially during
turbulent periods; and (ii) to revisit the role of central banks in public debt management.

Keywords: National debt, debt management
JEL classification: E610, E630, H630, H740

1 Executive Director, Reserve Bank of India.

194 BIS Papers No 67

Sovereign debt management (SDM) is important for other macroeconomic policies,
especially monetary policy setting and transmission. The recent global crisis has brought
sovereign debt to the forefront as debt surged to unsustainable levels in many advanced
countries, triggering sovereign debt crises. Reinhart and Rogoff (2011) find that, for countries
with systemic financial crises and/or sovereign debt problems (Greece, Iceland, Ireland,
Portugal, Spain, the United Kingdom and the United States), average debt levels are up by
about 134% since 2007, surpassing by a significant margin the three-year 86% benchmark
that the same authors (2009) find for earlier deep post-war financial crises.2 These debt
levels have posed severe challenges for other macroeconomic policies and objectives. Some
central banks, especially in advanced countries, have applied unconventional monetary
policy measures, ie outright purchase of long-term government bonds to influence long-term
interest rates. However, the success of such measures has yet to be proven. Nevertheless, a
broad consensus has emerged in academia and among policy practitioners on the
importance of coordination between debt management and other macroeconomic policies,
especially monetary policy.
The sovereign debt composition in terms of maturity, instruments and currency could also
have grave implications for other macroeconomic policies. For instance, heavy government
borrowing combined with the outright purchase of government securities by central banks
has heightened the interaction between monetary policy and SDM. Rising sovereign default
risks and increased volatility in markets for government securities have serious implications
for financial markets and financial stability, given that government securities constitute a
large part of banks’ and financial institutions’ portfolios.
India’s debt-to-GDP ratio rose from 69% before the recent global financial crisis to 73% in
2010. This increase was mainly due to India’s fiscal stimulus measures, which were similar to
those implemented by sovereigns worldwide after the financial crisis. This note covers the
various issues arising from the Indian experience with public debt management and the
challenges it poses to monetary policy.
Public debt management framework
The Reserve Bank of India (RBI) is responsible for managing India’s public debt, especially
debt denominated in the domestic currency. The management of the central government’s
debt is conducted by RBI under statutory provisions that oblige the central government to
delegate its debt management to the RBI. The debt of the sub-national governments, on the
other hand, is managed by the RBI under bilateral agreements. The RBI seeks to hold the
government’s borrowing costs to a minimum over the medium to long term, while keeping the
associated risks to a prudent level. The cost objective is largely met by deepening and
widening the government securities market, while rollover risk is contained by fixing upper
limits for yearly maturity buckets as well as individual securities. These limits are set
according to the government’s repayment capacity and the probable demand for government
securities. Further, the maturity of each new issue of government debt is influenced by the
interest rate cycle; shorter maturities are considered when the yield curve is steep and
vice versa.
Two landmark developments have shaped India’s public debt management framework,
namely (i) the March 1997 supplemental agreement between the RBI and the government

2 The combination of high and climbing public debts (a rising share of which is held by major central banks) and
the protracted process of private deleveraging makes it likely that the 2008–17 period will be aptly described
as a decade of debt

BIS Papers No 67 195

and (ii) the 2003 Fiscal Responsibility and Budget Management (FRBM) Act. The
supplemental agreement discontinued the issuance of ad-hoc treasury bills by the
government to the RBI to finance the fiscal deficit, while the FRBM Act prohibits the RBI from
participating in the primary auctions for government loans. Together, these measures
prevent the fiscal deficit from being monetised.
Apart from its role as debt manager, the RBI also acts as a banker to both central and sub-
national governments. Thus, the RBI provides Ways and Means Advances (WMA) and
limited overdrafts to both the central and sub-national governments allowing them to meet
any temporary mismatch between receipts and payments. Further, the RBI acts as a fiscal
adviser to both the central and the sub-national governments. For example, most of the sub-
national governments have adopted fiscal responsibility legislation that was originally
proposed by an RBI working group.
Debt management experience during the crisis
The expansionary fiscal policy adopted during the recent crisis resulted in higher government
borrowing during 2008–09 and 2009–10. The government’s gross market borrowing was
estimated at INR 1,497.80 billion in the 2008–09 budget. However, actual government
borrowing during 2008–09 amounted to INR 2,730 billion, about 83% higher than the budget
estimate. Gross borrowing increased further to INR 4,510 billion during 2009–10 reflecting
continued fiscal expansion. The challenge for the RBI was to manage a government
borrowing programme on the required scale without disrupting markets, especially in an
environment of uncertainty and heightened risk aversion among investors. The borrowings of
sub-national governments also increased by about two thirds in 2008–09 over the previous
year, as they also undertook countercyclical measures. The sub-national governments raised
from the market a gross amount of INR 1,181 billion and INR 1,311 billion during 2008–09 and
2009–10, respectively.
The associated challenges need to be viewed in the context of the fiscal stimulus packages
implemented worldwide after the crisis to offset falling consumption and investment. In India,
the most significant challenge for the RBI was to manage the sudden large increase in the
borrowing requirement during the crisis period. Second, liquidity in the system had dried up
due to large capital flow reversals as foreign investors withdrew funds from EMEs. Third,
uncertainty and general risk aversion in financial markets further complicated the task of the
debt manager in completing the borrowing programme without disrupting markets. To meet
these challenges, while also seeking to keep borrowing costs low over time and to mitigate
rollover risk, the RBI followed a multi-pronged strategy that included the following elements:
• front-loading of borrowing to make use of more favourable market conditions in the
first half;
• the Market Stabilisation Scheme (MSS), which was primarily used by RBI for
managing liquidity infused by capital flows, was de-sequestered to partly fund the
GFD alleviating pressure on fresh government borrowings;
• use of Treasury bills to partially fund the increased gross fiscal deficit;
• shortening of average maturity to lower effective borrowing costs. The average
maturity of India’s public debt was sufficiently long (ie 10.59 years as at end-March
2008) to allow scope for some shortening without a significant increase in rollover
risk;
• continued use of the RBI’s uniform price auction format to allow aggressive bidding
by investors in an uncertain market environment; and

196 BIS Papers No 67

• increased communication between the RBI and market participants through press
releases, meetings, and information on evolving issues and policy decisions.
The weighted average cost of borrowing through dated securities fell from 8.50% in 2007–08
to 8.23% during 2008-09 and further to 7.89% in 2009-10. The issuance of government
dated securities with maturities of five years or less increased during the crisis period, with
the weighted average maturity of dated securities issued during the year shortening from
14.9 years in 2007–08 to 13.81 years in 2008–09 and further to 11.16 years during 2009–10.
Interaction with monetary policy
The interaction between SDM and monetary policy operations is a topic that has attracted an
increasing amount of attention from both scholars and policymakers in recent years. When
the financial crisis forced a sharp rise in sovereign borrowing, debt managers in many
countries (eg the euro area) shifted the maturity structure of fresh borrowing towards the
short term. Issuance of short-term debt increased significantly in almost all OECD markets
during the crisis period (Blommestein (2010))3. Hoogduin et al (2010) note that the potential
for interaction between public debt management and monetary policy has risen due to the
increase in short-term debt during the recent global crisis period. Sovereign debt managers
generally operate over the medium to long term; but their increased short-term fund-raising
could potentially come into conflict with the monetary policy transmission mechanism.
Further, the greater reliance on short-term borrowing (for example, Treasury bills and cash
management bills in India) could distort the yield curve in a thin market, jeopardising
monetary policy signalling and its transmission mechanisms, besides having serious
implications for public welfare as the yield curve is a public good.4
Another possible interaction between SDM and monetary policy could be through the central
bank’s open market operations and the new issuance of securities by the debt manager.
Since the onset of the international crisis, central banks in many advanced economies and
emerging markets (EMs) have purchased government bonds in the secondary market as part
of unconventional monetary policies. However, the intended effect of purchasing long-term
securities (open market operations) by the central bank could be offset by a concurrent
decision by the sovereign debt manager to issue long-term securities. In this regard,
Mohanty and Turner (2011) note that the recent central bank operations in government debt
markets to influence the long-term interest rate are usually defended on the grounds that
monetary easing is constrained once the policy rate approaches zero. Furthermore, the
liquidity and monetary management operations of the monetary policy also interact with SDM
operations as government bonds are used as collateral in open market operations and other
liquidity facilities.
These potential interactions between monetary policy and SDM could be smoothened
without any adverse impact through seamless coordination between the monetary
policymaker and the debt manager. Such coordination, however, is more difficult when these
activities are conducted by different agencies. It has been argued that independent sovereign
debt managers, seeking solely to keep costs low, are tempted to prioritise their short-term

3 The explosion in the supply of public debt happened at a time when even sovereign issuers were experiencing
liquidity problem in their secondary markets.
4 When conventional monetary policy uses policy interest rate adjustments and signalling as the instrument,
central banks typically operate such that their transactions in government debt markets have only a minimal
impact on yields, so as not to undermine the usefulness of the yield curve as an indicator of macroeconomic
expectations (BIS (2011)).

BIS Papers No 67 197

goals. For example, the share of short-term issuances has recently increased significantly in
the sovereign debt of countries such as Greece, Portugal, Ireland and Spain, where SDM
has been segregated from monetary policy into a separate debt management agency.
Hoogduin, et al (2010) have analysed debt managers’ behaviour in the euro area, where
sovereign debt is managed by independent debt management agencies, finding that debt
managers are apt, in the interests of cost mitigation, to shift excessively towards short-term
borrowing in response to a steepening of the yield curve or other interest rate movement. If,
however, the central bank is also empowered to manage the country’s sovereign debt, it is in
a position to ensure seamless coordination between both activities. This kind of coordination
was evident in India during the recent global financial crisis, when it became vital to efficiently
manage the steep increase in government borrowing.
In India, debt management is currently carried out by the RBI’s Internal Debt Management
Department (IDMD), which is functionally separate from monetary policymaking. The debt
management strategy is formulated by the Monitoring Group on Cash and Debt
Management, which is the apex coordinating body between the RBI and the Ministry of
Finance. Contrary to the popular perception of a conflict between monetary policy and debt
management, there exists a strong confluence of interest in these two activities that are
undertaken by the RBI. In fact, any perceived conflict of interest was resolved by two
measures, namely (i) the March 1997 agreement between Government of India and RBI that
discontinued the issuance of ad-hoc treasury bills by the government to RBI, which
effectively put an end to the automatic monetisation of the fiscal deficit; and (ii) the 2003
Fiscal Responsibility and Budget Management (FRBM) Act, which debars the RBI from
participating in the primary market auction for government borrowing. Further, the open
market operations (OMO), in which the RBI purchases and sells government securities, are
coordinated with the government’s borrowing programme, ruling out any potential for conflict
between these activities. If there is a dilemma for RBI with regard to monetary policy and
debt management, it is related to systemic liquidity. That is, the system may need to be in
deficit for monetary policy transmission, whereas a system in surplus would be more
favourable for debt management. But RBI has resolved this dilemma by putting in place a
monetary policy operating framework whereby the system is allowed to be in deficit (or
surplus) to the extent of its frictional component, ie 1% (+/-) of the banking system’s net
demand and time liabilities. At the same time, any structural liquidity deficit (or surplus) is met
through OMOs.
The perception that a conflict exists between monetary policymaking and debt management
misses the point that monetary policy lies at the core of debt management. Without inflation
at a low and stable level, it would be very difficult to sell fixed coupon government securities,
particularly of longer maturities. Low and stable inflation since the mid-1990s has made it
possible to extend India’s sovereign yield curve. The RBI has also been actively engaged in
developing the government securities market, inter alia, in terms of instruments and investor
base, and the Bank has put in place an efficient infrastructure for trading, payment and
settlement. These efforts have helped to contain the cost of government borrowing over the
medium term. Therefore, a central bank that is also responsible for debt management can be
equally committed to price stability, particularly when debt management is its statutory
responsibility.
In this regard, Goodhart (2010) argues that debt management is again becoming a critical
element in the overall conduct of macroeconomic policy. Hence, he suggests, central banks
should be encouraged to revert to their role of managing the national debt. Subbarao (2011)
also concludes that, on balance, and as long as there are institutionalised mechanisms to
negotiate the various trade-offs within the overarching objective of achieving monetary and
financial stability, the separation of debt management from central bank would seem to be a
sub-optimal choice.

198 BIS Papers No 67

Further issues
The increased interaction between SDM and monetary policy raises two important issues
that urgently need to be addressed: (i) to ensure seamless coordination between SDM and
monetary policy, especially during turbulent times; and (ii) to revisit the role of central banks
in public debt management. In countries where debt management has been separated from
the central bank and entrusted to an independent debt management office (DMO), an
institutional mechanism may exist for coordination between debt management and monetary
policy. But the larger question is whether the desired coordination is taking place in practice,
as the central bank and the DMO may at times find their objectives in conflict. Thus, the
coordination mechanism needs to be reviewed, especially against the backdrop of auction
failures in the United Kingdom and Germany in the recent past, and the sub-optimal debt
structures implemented by some DMOs. In India, the 2007–08 budget announced that an
independent DMO would be set up and a middle office has already been set up in the
Ministry of Finance. If an independent DMO is established in countries where the
responsibility for SDM currently lies with the central bank, then challenges might arise when
seeking to ensure seamless coordination between monetary policy and debt management.
This is particularly the case where the level of sovereign debt is high, as in India where
government borrowing has increased in parallel with the fiscal deficit. A further challenge
would be to ensure that the borrowing programmes of the central government and the sub-
national governments are fully coordinated.
References
BIS (2011): “Interaction of sovereign debt management with monetary conditions and
financial stability”, CGFS Papers, no 42.
Goodhart, C (2010): “The changing role of central banks”, paper for the ninth BIS Annual
Conference.
Hoogduim, L, B Ozturk and P Wierts (2010): “Public debt managers’ behaviour: interactions
with macro policies”, DNW Working Paper, no 273.
Mohanty, M S and P Turner (2011): “Monetary policy in overindebted economies”, paper
presented at Scottish Institute for Research in Economics and the Money, Macro and
Finance Research Group, Heriot-Watt University, 29-30 September.
Reinhart, C and K Rogoff (2010): “Debt and growth revisited”, MPRA Paper, no 24376
(http://www.voxeu.org/index.php?q=node/5395).
——— (2010): “A decade of debt”, Centre for Economic Policy Research, DP 8310.
Subbarao, D (2011): “Central bank governance issues: some RBI perspectives”, comments
at the meeting of the Central Bank Governance Group in Basel on 9 May 2011.
Wolswijk, G and J d Haan (2005): “Government debt management in the euro area: recent
theoretical developments and changes in practices”, ECB Occasional Paper Series, no 25.

BIS Papers No 67 199

Fiscal policy, public debt management and
government bond markets in Indonesia
Mr Hendar1
Abstract
Over the past several years, the Indonesian government has pursued a prudent fiscal policy
while still promoting economic growth. Since 2005, the government has shifted the source of
deficit financing from external to domestic debt via the issuance of government securities. In
doing so, it has sought to lengthen the maturity of local currency government bonds and to
construct a yield curve. Meanwhile, the global excess of liquidity has driven foreign investors
to seek for higher yields. With its strong fundamentals and attractive yields, Indonesia has
therefore been the recipient of massive capital inflows, most of which have been invested in
stock and government bonds. As the central bank, Bank Indonesia has adopted a mixture of
monetary and macroprudential policy measures to manage these capital inflows and excess
liquidity. From early 2008, the Bank has conducted daily operations with government
securities to manage liquidity in the market.

Keywords: Central Banks, Monetary Policy, Fiscal Policy, Indonesia
JEL classification: E58, E63

1 The author would like to thank all colleagues who contributed to this paper, as follows Clarita Ligaya, Erwin
Hutapea, Rosita Dewi, Elpiwin Adela, Dythia Sendrata and Aldy Mochamad.

200 BIS Papers No 67

Fiscal policy overview
Fiscal policy indicators show that the Indonesian government has managed to pursue
a prudent fiscal policy while still promoting economic growth. The fiscal policy stance is
measured by estimating the overall balance, the primary balance, and fiscal impulse. Over
the past several years, a relatively low deficit has been reflected in the ratio of the overall
balance, which indicates the difference between revenues and grants, and expenditures. The
exception was in 2009, when the overall fiscal deficit reached 1.6% of GDP as the
government sought to cushion the impact of the global crisis on the Indonesian economy. At
the same time, the fiscal impulse – which indicates the role of government in increasing (or
dampening) aggregate demand – also showed an expansion, in line with the government’s
prioritisation of economic growth.
Indonesia overall and primary balance Fiscal impulse
����
����
����
����
����
����
����
���
���
���
���
���
���
���
����
����
����
����
���
���
���
���
���
���
���� ���� ���� ���� ���� ���� ����
� ����������������� ������� ���� �������������� �������
�

����
����
����
����
���
���
���
���
���
���� ���� ���� ���� ���� ���� ���� ���� ����
���������
�����������

Medium-term pressures could potentially arise from official subsidies, especially those for
fuel and electricity. Rising as it does in line with the oil price, the cost of fuel subsidies has
the potential to undermine the state budget. In 2012, the government has sought to place
limits on the fuel subsidy, with a view to reallocating the subsidy budget to more productive
expenditure such as infrastructure and capital development. However, the medium-term
demographic threat to the government budget from pensions or health care spending is
limited, we believe, given that around 65% of Indonesia’s population lies within the
economically active 15–64 age bracket.
Energy subsidies as a share of GDP Population demography
���
���
���
���
���
���
���
���
��� ���
���
��� ���
���
����
����
����
����
����
����
�����
�
��
��
��
��
���
���
���
���
���
���
���
���
���
���
���
���� ���� ���� ���� ���� ���� ����
��������������� ������� ���������������������� ������� ���������
����������

��
���
���
���
���
���
���
���
���
���
����
���� ���� ���� ���� ���� ���� ���� ���� ���� ���� ���� ���� ���� ����
������������ ������������� ������������� �������������
������������� ������������� ����� ��������
The government has shifted the financing of the government deficit from external to
domestic financing. Since 2005, government securities have become the primary financing
instruments for the government deficit. The ratio of domestic debt relative to total debt

BIS Papers No 67 201

increased from 45% in 2005 to 54% in 2010. However, this trend was interrupted in 2005 and
2008, when new public debt was issued in US dollars. The government’s aim is to refinance
maturing foreign debt by issuing securities in the domestic market, to achieve a sound
balance of foreign and domestic debt, and to strengthen the domestic financial market. Thus,
financing from external borrowing has trended downwards since 2004. From that point,
planned loan repayments have been set to exceed disbursements.
Central government debt Composition of central government debt

Source: Debt Management Office, Ministry of Finance.
Bonds market, money market and monetary policy implementation
The government is seeking to lengthen the maturity of its domestic bonds and to
construct a yield curve. More than 90% of the outstanding central government debt is in
medium- to long-term maturities. Since 2002, the government has conducted a reprofiling
strategy to improve the maturity profile of government debt securities, hence reducing
refinancing risk. Capital flows from abroad, combined with a relatively flat yield curve, have
pushed down government bond yields. With its strong fundamentals and attractive yields,
Indonesia has received massive capital inflows from yield-seeking foreign investors. Most of
these inflows are invested in stocks and government bonds. Coupled with the limited supply
of paper, this strong demand has further reduced the yield on government bonds.
Maturity profile of government debt Reprofiling strategy for government debt

����������
����
�����
�����
�����
�����
�����
�����
������
�����
�������������� ����
������
������
������������
������������
�����
���������
���
�
��
��
��
��
���
���
���
������� �������������������� �����������������������
�������

Source: Debt Management Office, Ministry of Finance.

202 BIS Papers No 67

Yield curve for government debt
���
���
��� ���
���
���
���
��� ��� ���
���
���
���
���
���
��� ��� ���
���
��� ��� ���
���
���
���
���
���
��� ��� ���
���
���
��� ���
��� ���
���
���
���
���
���
���
���
���
���
� � � � � � � � � �� �� �� ��
������ � ����� ������
�

The Indonesian money market is dominated by interbank call money. In terms of the
underlying transaction assets, the Indonesian money market can be categorised into
collateralised and uncollateralised segments. Based on transaction volume, more than 70%
of money market transactions, in daily volume terms, are conducted in interbank call money.
Owing to the excess liquidity in the economy, the money market rate tends to hover around
the lower border of the central bank’s corridor, ie the deposit facility rate. As the repo market
is undeveloped on account of this excess liquidity, demand for short-term liquidity is largely
met by interbank call money. Repo rates are also higher than those for interbank call money,
despite the lower credit risk.
Bank Indonesia has taken various monetary and macroprudential policy measures to
manage capital inflows and curb excess liquidity. To maintain financial stability amid
massive portfolio capital inflows, Bank Indonesia has imposed a six-month holding period for
Bank Indonesia bills (formally known as Certificates of Bank Indonesia or SBI). At the same
time, this policy measure supports the central bank’s aim of managing excess liquidity by
lengthening the maturity profile of monetary instruments. In this regard, the Bank is currently
issuing only nine-month SBIs (in monthly auctions) with a view to absorbing liquidity over a
more sustained period.
Daily average volume Interbank money market rate
���� ���� ���� ���� ���� ����
����������� � ����� ����� ����� ����� ����� ������
�������� ���� ���� ���� ���� ���� ����
������������� ���� ���� ���� ���� ���� ����
������������������� ���� ��� ���� ��� ���� ����
�
�
�
�
�
��
��
��
����
��������

0
5
10
15
20
2006 2007 2008 2009 2010 2011
%
������ ������� ������� ������� ����

Since early 2008, Bank Indonesia has conducted daily operations to manage liquidity
in the market. Among others, one aim of this policy enhancement is to reduce fluctuations in
the overnight interbank call money rate, which is the operational monetary policy target rate.
Daily liquidity management operations have relied mainly on term deposits, reverse repo
transactions (with government bonds as underlying securities) and deposit facilities (standing

BIS Papers No 67 203

facilities). As the market believes that Bank Indonesia will act to maintain the stability of the
overnight interbank call money rate, it is expected that the effect of these measures will be
transmitted to the longer-term interest rates too.
The coordination of monetary policy and fiscal policy
The monetary authority coordinates with the fiscal authority in conducting macro
policy. According to the Central Bank Act, Bank Indonesia has operational independence in
determining monetary policy. However, to improve the effectiveness of monetary policy and
fiscal policy, there is close cooperation between the central bank and the fiscal authority.
Moreover, Bank Indonesia is prohibited from buying government securities in the primary
market for its own account, except when purchasing short-term government securities solely
for the purpose of monetary operations. In the primary market for short-term government
securities, Bank Indonesia acts as a non-competitive bidder. At the same time, according to
the Government Securities Act, Bank Indonesia is the auction and administration agent for
government securities. In this connection, the government is required to consult with Bank
Indonesia before issuing government securities.
Since 2002, Indonesia’s public debt management has been conducted by the Debt
Management Office within the Ministry of Finance. For its part, Bank Indonesia is
responsible for managing the country’s foreign exchange reserves, conducting
various foreign exchange transactions, and receiving the proceeds of foreign
borrowing. The central bank and the Ministry of Finance coordinate closely with the aim of
improving the quality of macroeconomic policy. In addition, the Central Bank Act requires the
Ministry of Finance to consult with Bank Indonesia before any issuance of debt. The Act
further states that Bank Indonesia must evaluate the monetary implications of such issuance
and advise the Ministry of Finance on its terms. As the auction agent for domestic
government securities, Bank Indonesia announces the auction plan for government
securities, conducts the auctions, and announces the auction results. The Bank also has a
role in administering government securities, including their registration, clearing and
settlement, and repayment in both primary and secondary markets. For global issuance, the
Ministry of Finance appoints external agents to conduct the auctions as well as clearing and
settlement.
Since 2011, Bank Indonesia has increased its use of government bonds in its reverse
repo operations with the aim of absorbing excess liquidity. As part of its enhanced
monetary policy implementation, Bank Indonesia has gradually shifted from the issuance of
central bank bills (SBI) to reverse repo transactions as a means of absorbing excess liquidity.
This strategy has a twofold benefit, in that it stimulates the development of the government
bond market and reduces price discrepancies between different government securities.

BIS Papers No 67 205

The interaction between monetary and fiscal policy:
insights from two business cycles in Israel
Kobi Braude and Karnit Flug1
Abstract
Comparing the two significant recessions Israel experienced over the last decade, we
highlight the importance of sustained fiscal discipline and credible monetary policy during
normal times for expanding the set of policy options available at a time of need. In the first
recession Israel was forced to conduct a contractionary fiscal and monetary policy, whereas
in the second one it was able to pursue an expansionary policy. The difference in the effect
of the policy response between the two recessions is sizable: it exacerbated the first
recession while it helped to moderate the second one.

Keywords: Fiscal policy, fiscal discipline, public debt, monetary policy, counter-cyclical policy,
business cycles
JEL classification: E52, E62, H6

1 Bank of Israel. We thank Stanley Fischer and Alon Binyamini for helpful comments and Noa Heymann for her
research assistance.

206 BIS Papers No 67

Introduction
Over the last decade Israel experienced two significant business cycles. The monetary and
fiscal policy response to the recession at the end of the decade was very different from the
response to recession of the early 2000s. In the earlier episode, following a steep rise in the
budget deficit and a single 2 percentage point interest rate reduction, policy makers were
forced to make a sharp reversal and conduct a contractionary policy in the midst of the
recession. In the second episode, monetary and fiscal expansion was pursued until recovery
was well under way. This note examines the factors behind the difference in the policy
response to the two recession episodes.
Comparing the two episodes, we highlight the importance of fiscal discipline and the
reduction of public debt over time for allowing counter-cyclical fiscal and monetary policy
during recessions. In particular, we show that the improved fiscal situation on the eve of the
last recession, along with other factors, played an important role in allowing the Bank of
Israel to pursue a highly expansionary monetary policy during the recent recession, which
helped to moderate and shorten it. By contrast, the poor fiscal situation that preceded the
previous crisis and too-sharp an instant interest rate cut that proved unsustainable not only
prevented any monetary expansion during that crisis; it actually forced the central bank to
raise its interest rate in the midst of the crisis. This actually exacerbated the recession. A
rough estimate shows that the difference in the effect on GDP of the policy response
between the two episodes was sizable.
Two recessions – different circumstances
During the years 2001-2003 Israel experienced its worst recession in decades, which
included four consecutive quarters of negative GDP growth (Figure 1). The dramatic change
in the state of the economy, which came after an exceptional 9 percent GDP growth rate in
2000, was due to the unpleasant combination of the burst of the global hi-tech bubble and a
sharp deterioration in Israel’s security situation (the Intifada). Unemployment rose sharply,
peaking at about 11 percent (Figure 2), and began to decline only after about 3 years. The
budget deficit peaked at almost 6 percent of GDP in 2003 (Figure 3), and the public debt,
which was high to begin with, reached almost 100 percent of GDP that year (Figure 4).
SOURCE: Based on Central Bureau of Statistics data.
Figure 1: Growth Rate of GDP
2000-2011
(Quarterly, seasonally adjusted, annual rates of change)
-7
-4
-1
2
5
8
11
14
17
03
/0
0
09
/0
0
03
/0
1
09
/0
1
03
/0
2
09
/0
2
03
/0
3
09
/0
3
03
/0
4
09
/0
4
03
/0
5
09
/0
5
03
/0
6
09
/0
6
03
/0
7
09
/0
7
03
/0
8
09
/0
8
03
/0
9
09
/0
9
03
/1
0
09
/1
0
03
/1
1
%
SOURCE: Based on Central Bureau of Statistics data.
Figure 2: Unemployment
2000-2011
5
6
7
8
9
10
11
12
03
/0
0
09
/0
0
03
/0
1
09
/0
1
03
/0
2
09
/0
2
03
/0
3
09
/0
3
03
/0
4
09
/0
4
03
/0
5
09
/0
5
03
/0
6
09
/0
6
03
/0
7
09
/0
7
03
/0
8
09
/0
8
03
/0
9
09
/0
9
03
/1
0
09
/1
0
03
/1
1
%

BIS Papers No 67 207

* The data refer to the deficit excluding the Bank of Israel’s profits, and excluding credit extended.
SOURCE: Bank of Israel.
Figure 3: Budget Deficit*
1995-2010
(Percent of GDP )
4.8
5.8
4.4
5
3.6
1.4
4
5
5.8
4
2.4
1
0.2
2
5
3.4
0
1
2
3
4
5
6
1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010
%
SOURCE: Bank of Israel.
Figure 4: Gross Public Debt
1995-2010
(Percent of GDP )
102.3 100.3 99.3 101
94.8
84.3
89
96.7 99.3 97.7 93.7
84.7
78.1 77 79.4 76.1
0
20
40
60
80
100
120
1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010
%

The 2008-2009 recession in Israel was somewhat different: it was due entirely to external
factors – the global crisis – and was milder and shorter than the 2001-2003 recession. In
particular, growth was negative for only 2 quarters, and the rise in unemployment, while quite
sharp, was short-lived: it peaked within just 3 quarters and started declining thereafter. It
should be emphasized, however, that it is only in retrospect that we can characterize the
outcomes of the recent crisis as milder. The shock to real activity in Israel may have been not
much smaller than in the previous recession, and the shock to its financial markets was
certainly larger. The fact that, ex post, the recent crisis in Israel turned out to be milder than
the previous one, and milder than feared in real time, is in part due to the more aggressive
policy pursued in Israel, and in part to the success of policy measures abroad in containing
the crisis.
Several important differences in the circumstances under which the two recessions evolved
should be noted:
The relative nature of the shock: As noted, in 2001 Israel faced a unique combination of
shocks such that it was hit more severely than the rest of the world and its relative risk
increased. In contrast, the core of the recent crisis did not include Israel, and in many
respects it fared better than most advanced countries during the recession. For example, the
fall in exports and in GDP in Israel in 2001 was much larger than in the advanced
economies, whereas in the 2008-2009 crisis Israeli exports decreased by roughly the same
as in the advanced economies and GDP fell considerably less (Figures 5 and 6). Notably,
while sharp declines in housing prices and housing investment played a major role in the
development of the recent crisis in the US and other advanced economies, demand and
prices in the Israeli housing market rose during 2009. This rise was partly driven by the Bank
of Israel’s interest rate cuts and reflected the desired transmission of its expansionary
monetary policy to the construction sector and, through it, to the economy at large.2 The
considerable pressures for an appreciation of the shekel during much of the recent crisis,
which the Bank was trying to moderate by purchasing foreign currency, also reflected Israel’s
relatively favorable position at that time.

2 Activity in the Israeli housing market slowed down at the peak of the crisis (late 2008 and the beginning of
2009), but accelerated thereafter. The acceleration during 2009 stood in sharp contrast to the falling real
prices and low level of activity in this market in the preceding decade.

208 BIS Papers No 67

Figure 5: Change in Exports of Goods and Services
2000-2010
(Volume change, annual)
-2
0
2
4
6
8
10
2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010
%
Advanced economies Euro area Emerging and developing economies Israel USA
SOURCE: IMF World Economic Outlook, September 2011.

Figure 6: Rate of GDP Growth
2000-2010
(Annual)
-6
-4
-2
0
2
4
6
8
10
2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010
%
Advanced economies Euro area Emerging and developing economies Israel USA
SOURCE: IMF World Economic Outlook, September 2011.

World policies: Low interest rates and substantial monetary and fiscal expansions around
the world during the recent crisis made such policies in Israel more feasible and acceptable
to financial markets. This is particularly true regarding the very low level which the interest
rate in Israel reached. However, the low levels of interest rates abroad do not tell the entire
story, since it is not just the level that was much lower in Israel during the recent recession
compared with the 2001-2003 one. Figure 11 shows that the interest rate differentials were
also much lower. This reflects a decline in Israel’s risk premium and underlines the
importance of the improvement in its particular situation compared with the 2001-2003
episode, over and above the global circumstances.
The state of the economy in the years preceding the crisis: The recent crisis hit Israel
after about 5 years of exceptionally high growth (about 5 percent a year). This was robust
and sustainable growth3 in the sense that it was driven by strong fundamentals: strong export
growth driven by world demand and sound macroeconomic policy, which included both fiscal
discipline and monetary credibility, as well as structural reforms. Growth rates in the years
preceding the previous recession were slower4 and were not sufficiently based on strong
fundamentals and fiscal discipline.
The current account and IIP: As in other respects, Israel’s current account on the eve of
the last recession was in much better shape than in the run-up to the 2001-2003 recession
(Figure 7). Israel had been running a deficit on that account prior and during most of that
recession, while it has had a substantial surplus since 2003. These accumulated surpluses
also resulted in an improved IIP at the onset of the recent crisis, which increased Israel’s
resilience to the crisis. In fact, Israel’s net foreign liabilities have neared zero since 2008. In
particular, it has held a positive and growing net asset position in debt instruments since
2003 (Figure 8).5 However, it is noteworthy that neither of the two crises were essentially
balance-of-payments crises. Moreover, the current account deficit at the onset of the
previous recession was not large (1.6 percent of GDP) and followed a trend of decline in
those years. Hence, the importance of the balance of payments notwithstanding, this is

3 Absent the global crisis the robust growth could have to continued, though likely at somewhat lower rates as
the economy was gradually shifting from a cyclical expansion to long-term growth.
4 As noted the spectacular growth in 2000 was exceptional and unsustainable. It collapsed at once at the end of
that year.
5 The position in debt instruments has been shown to be particularly associated with debt crises around the
world. See the IMF World Economic Outlook Sep. 2011. Box 1.5.

BIS Papers No 67 209

apparently not the major difference between the two crises in regard to the economy’s
situation.
SOURCE: Balance of Payments, Central Bureau of Statistics.
Figure 7: Current Account of the Balance of Payments
2000-2010
(Percent of GDP )
4.8
-1.7
3.6
2.9
0.9
2.73.1
1.7
-1.1
0.5
-1.6
-6
-4
-2
0
2
4
6
2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010
%
SOURCE: Bank of Israel.
Net external debt assets
Net foreign assets – total
Net external equity assets
Figure 13: Israel’s International Invesment Position
1996-2010
(Percent of GDP )
-50
-40
-30
-20
-10
0
10
20
30
1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010
%
Figure 8:

Real time perception of the severity of the crisis: Beyond the objective circumstances
described above, differences in the real time assessment of the severity of the situation also
contributed to the different policy responses. The dramatic events in global markets as the
recent crisis evolved, combined with the tremendous uncertainty and concerns over the
potential deterioration, pointed to a possibility that this crisis would be substantially worse for
Israel than the 2001-2003 one. As policy makers had to act under such uncertainty and react
in a timely and preemptive manner, this real time perception played an important role in
motivating the aggressive policy response in Israel. The fact that, ex post, the recent crisis in
Israel turned out to be milder than feared is in part due to the more aggressive policy
pursued in Israel, and in part to the success of policy measures abroad in containing the
crisis.
A different policy response
The different circumstances noted above notwithstanding, both recessions were deep
enough to require a significant counter-cyclical fiscal and monetary policy response.
However, such policy was pursued only in the 2008-2009 recession.
In the first recession, the increase in the budget deficit (Figure 3) brought about by the fall in
economic activity caused yields on government bonds to soar in 2002-2003 (Figure 9) as
financial markets were reluctant to finance the growing debt. Thus Israel was forced to
pursue a procyclical fiscal policy, cutting public spending drastically. This is well illustrated by
the decrease in the cyclically adjusted budget deficit6 during 2001-2002 (Figure 10). The
excessively sharp cut in the Bank of Israel interest rate – 2 percentage points at once –
turned out to be unsustainable (Figure 11). Thus, the response of financial markets, for
example the rise in yields and the depreciation of the shekel, forced the Bank to raise its
interest rate sharply in the midst of the recession by about 5 percentage points within a few
months and maintain it at a high level for a considerable period of time.

6 The adjustment is based on the tax revenues that could be expected if GDP were currently at its potential
level.

210 BIS Papers No 67

SOURCE: Bloomberg.
Figure 5: 10-Year Government Bond Rate
2001-2011
3
6
9
12
15
05
/0
1
11
/0
1
05
/0
2
11
/0
2
05
/0
3
11
/0
3
05
/0
4
11
/0
4
05
/0
5
11
/0
5
05
/0
6
11
/0
6
05
/0
7
11
/0
7
05
/0
8
11
/0
8
05
/0
9
11
/0
9
05
/1
0
11
/1
0
05
/1
1
%
Figure 9:
SOURCE: Bank of Israel.
Figure 10: Cyclically-Adjusted Budget Deficit
1999-2010
(Percent of GDP )
0.4
3.2
1.8
2.2
3.2
0.3
0.1
1.2
2.2
3.4
2.9
3.1
0
1
2
3
4
5
6
1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010
%

The recent recession looks very different in terms of fiscal and monetary policy. This time the
government did not have to cut total spending at all, and in fact let the tax-revenue automatic
stabilizers act in full, allowing the budget deficit to rise to 5 percent of GDP in 2009, in line
with the decline in tax receipts.7 This was reflected in a rise in the cyclically adjusted budget
deficit during 2008-2009 (Figure 10), which was exceptional in view of previous recessions in
Israel in which policy was typically procyclical.8 The policy was well received by the financial
markets, as reflected in the relative stability of government bond yields in 2008 (which even
declined somewhat at the beginning of 2009), in sharp contrast to their surge in 2002.
The difference in monetary policy between the two episodes is perhaps even more striking.
The Bank of Israel responded to the recent crisis with an unprecedented monetary
expansion. Starting in October 2008 the Bank cut its interest rate by 3.75 percentage points
bringing it to 0.5 percent in April 2009, its lowest level ever (Figure 11). In contrast to the
sharp rate cut at the end of 2001 which proved unsustainable, this time the Bank was able to
maintain the rate at its near-zero level, and started raising it in September 2009 in view of the
rapid improvement of the economic situation and the resumption of growth, not because of
pressures from the financial markets. As the monetary rate approached its near-zero level,
the Bank also started implementing quantitative easing by purchasing government bonds in
the secondary market. At the same time, the Bank continued its purchase of foreign
currency. The Bank had begun these purchases about a year earlier as a preemptive
measure to moderate the appreciation of the shekel, which could prove particularly harmful
to Israeli exports when combined with the contraction in world demand associated with the
global recession. These measures, which were quite unthinkable during the previous
recession and would have probably had a major destabilizing effect on financial markets at
that time, did not invoke any irregular reactions in the markets during the last recession. As
noted, yields remained relatively steady in 2008-2009.
The evolution of the exchange rate provides a further illustration of the difference between
the two episodes (Figure 12). Following the interest rate cut at the end of 2001, the shekel
depreciated sharply, and remained at its highly depreciated level for more than a year,
despite the sharp increase in the interest rate which followed immediately after its attempted

7 Governments in Israel tend to raise the rates of indirect taxes (mainly the VAT) during recessions. In that
respect even the 2008-2009 recession was no exception, as described in Strawczynski and Weinberger
(2011). However, unlike in previous recessions, this increase was explicitly designed to allow a corresponding
increase in expenditure, and not to offset the effect of the automatic stabilizers.
8 The cyclicality of fiscal policy in Israel is discussed in Strawczynski and Zeira (2007).

BIS Papers No 67 211

cut. This underlined the inability to pursue a monetary expansion at the time, due in part to
the fiscal situation and the undermined credibility of monetary policy. In contrast, during the
recent recession not only did the Bank cut the interest rate and narrow the rate differential
with world rates, it also purchased substantial amounts of foreign currency in order to induce
a depreciation of the shekel and support exports. The sharp depreciation of the shekel during
the second half of 2008 and the beginning of 2009 reflected these purchases, as well as a
change in the perception of markets regarding the resilience of the domestic economy to the
global recession.9
SOURCE: Bank of Israel.
BoI interest rate
FED interest rate
Figure 11: BoI and FED Interest Rate
2000-2011
0
2
4
6
8
10
01
/0
0
07
/0
0
01
/0
1
07
/0
1
01
/0
2
07
/0
2
01
/0
3
07
/0
3
01
/0
4
07
/0
4
01
/0
5
07
/0
5
01
/0
6
07
/0
6
01
/0
7
07
/0
7
01
/0
8
07
/0
8
01
/0
9
07
/0
9
01
/1
0
07
/1
0
01
/1
1
07
/1
1
%

Figure 12: The Nominal Shekel/Dollar Exchange Rate
2000-2011
3
3.2
3.4
3.6
3.8
4
4.2
4.4
4.6
4.8
5
01
/0
0
07
/0
0
01
/0
1
07
/0
1
01
/0
2
07
/0
2
01
/0
3
07
/0
3
01
/0
4
07
/0
4
01
/0
5
07
/0
5
01
/0
6
07
/0
6
01
/0
7
07
/0
7
01
/0
8
07
/0
8
01
/0
9
07
/0
9
01
/1
0
07
/1
0
01
/1
1
07
/1
1
%
SOURCE: Bank of Israel.
Why was the policy response so different?
The difference in the policy responses between the two crises, which were both severe
enough to call for counter-cyclical fiscal and monetary measures, raises the question as to
why a policy like the one undertaken in the 2008-2009 recession was not feasible in
2001-2003. In other words, why were financial markets willing in 2008-2009 to accept fiscal
and monetary policy which they would by no means tolerate in the 2001-2003 recession?
We argue that initial conditions at the onset of the crisis are crucial for understanding this
difference. As noted above, Israel entered the recent crisis in a much better position in terms
of the state of the economy. The different nature of the crises and the global environment are
also important in this respect. However, the most important factor is probably the difference
in the state of fiscal and monetary policy on the eve of the two crises. In the recent crisis
Israel reaped the benefits of several years of sound macroeconomic policy, particularly in
terms of persistent fiscal discipline, which was crucially lacking at the onset of the previous
crisis, and in terms of the credibility of its price stability target, which had been established
over time.

9 It should be noted that the pass-through of exchange rate movements to the consumer price index had
declined over time in Israel (due in large part to reduced indexation of housing prices to the shekel/dollar
exchange rate). Thus the depreciation during the recent crisis exerted much less inflationary pressure than it
did in the 2001-2003 recession. This mitigated the need to strike a balance between the benefits of
depreciation for exports and its inflationary costs.

212 BIS Papers No 67

Fiscal policy: In the years preceding the recent crisis, Israel pursued a very disciplined fiscal
policy.10 It avoided substantial increases in public spending despite its rapid growth, and
used the large cyclical tax revenues and receipts from privatization to reduce public debt.
The budget deficit declined steadily between 2003 and 2007 (the budget was almost
balanced in 2007), as did even the cyclically adjusted one until 2006 (Figures 3 and 10).
Thus the public debt decreased from almost 100 percent of GDP in 2003 to 77 percent in
2008 (Figure 4).
Such fiscal discipline was lacking in the years preceding the 2001-2003 recession, and the
government was not making much progress in terms of fiscal consolidation at the time.
During 1995-1999 the budget deficit was around 4-5 percent of GDP, and public debt was
around 100 percent of GDP, showing no real signs of embarking on a downward path.11 Thus
Israel entered the crisis in 2001 with alarmingly high public debt, a poor fiscal reputation and
a troubling outlook for its fiscal standing. This state of things made it particularly vulnerable to
shocks. As soon as the economy was hit by (a combination of) shocks and the deficit
increased due to the fall in tax revenues, yields surged, as did the exchange rate. In fact,
policy makers lost all degrees of freedom: they were forced to tighten fiscal policy and more
than offset the effect of the automatic stabilizers. The problematic fiscal circumstances
reflected on monetary policy as well: in view of these circumstances, financial markets were
also reluctant to tolerate a monetary expansion.
Monetary policy: Despite considerable fluctuations in actual inflation during 2003-2007,
inflation expectations remained relatively stable and almost entirely within the inflation target
range during that time (Figure 13). This reflects the degree of credibility that monetary policy
had established in those years. This credibility played an important role in allowing the highly
expansionary monetary policy during the recent crisis without jeopardizing the stability of
prices and financial markets: in spite of the sharp interest rate cut, the quantitative easing,
and actual inflation exceeding the upper bound of the target range during the crisis, inflation
expectations remained within the target most of time and their fluctuations were smaller than
those of actual inflation.12
Such credibility of monetary policy had not been sufficiently established by the time the
2001-2003 recession hit Israel. During the second half of the 1990s Israel was still
proceeding with its disinflation process. While inflation had been lowered substantially in
those years, in fact falling below the inflation target, inflation expectations remained as
volatile as actual inflation and credibility had yet to be consolidated (Figure 13). Under these
circumstances, the unduly sharp single rate cut at the end of 2001 undermined credibility.
Building on the credibility established in recent years and spreading the rate cut over several
months, the Bank was able to sustain a much larger cumulative rate cut (3.75 percentage
points) during the 2008-2009 recession.13

10 Brender (2009) studies Israel’s fiscal policy during 1985-2007 and concludes that during those years only two
periods, 1985-1992 and 2002-2006, can be characterized as episodes of sustainable consolidation.
11 The sharp decrease in public deficit and debt in 2000 is not a reflection of fiscal consolidation but rather the
(short-lived) result of the exceptional (and equally short-lived) GDP growth rate in that year. Moreover, as
Brender (2009) notes, the government actually raised the deficit target for that year and introduced several
expansionary policy initiatives.
12 In fact, inflation expectations at the end of 2008 and the beginning of 2009 fell below the lower bound of the
range, reflecting in large part fears of the crisis and its potential deterioration. A major concern of monetary
policy at that time was indeed to prevent a deflationary spiral.
13 The intolerance of financial markets to the 2001 rate cut was due to additional factors. This cut was supposed
to be a part of a package deal in which the government was to take immediate measures to reduce its deficit.
However, it did not. It also appears that the interest rate had been kept too high for too long and that a more
gradual reduction over time might have turned out to be more sustainable. The surprising and dramatic
2 percentage point cut took markets by surprise and seemed like a breach in policy.

BIS Papers No 67 213

* Expectations derived from the capital market.
SOURCE: Bank of Israel.
Consumer price
index
Inflation
expectations
Figure 13: Inflation over Past 12 Months, Inflation Target and
Inflation Expectations*
1997-2011
-3
-1
1
3
5
7
9
11
01
/9
7
09
/9
7
05
/9
8
01
/9
9
09
/9
9
05
/0
0
01
/0
1
09
/0
1
05
/0
2
01
/0
3
09
/0
3
05
/0
4
01
/0
5
09
/0
5
05
/0
6
01
/0
7
09
/0
7
05
/0
8
01
/0
9
09
/0
9
05
/1
0
01
/1
1
09
/1
1
%

An illustrative estimate of the effect of the different policy responses
In this section we provide a rough estimate of the effect of the policy response in each crisis.
We estimate the (counterfactual) cumulative loss of GDP that would have been caused by
the exogenous factors absent any policy response, and compare it to the actual cumulative
loss of GDP.14 We attribute the difference between the two losses in each episode to the
effect of the policy response (fiscal and monetary) in that episode.
The main exogenous factors affecting GDP growth in the 2001-2003 recession were the
Intifada and, to a lesser extent, the slowdown in world trade. In the 2008-2009 recession, the
decline in world trade was the major exogenous factor, and an additional important factor
was a wealth effect driven by the decline in the value of financial assets, which affected the
purchase of durable goods.15 We calculate the effect of world trade on GDP in each episode
drawing on the unit elasticity of Israeli exports to world trade, which has been found in many
studies, and applying the share of exports in GDP as well as the value added of exports that
were relevant in each period. The loss of GDP due to the Intifada is calculated using
estimates published by the Bank of Israel (2001-2003) in its annual reports. The effect on
GDP of the decline in the purchase of durables arises mainly through import taxes. This is
because a substantial part of these goods in Israel are imported and the significant taxes on
these imports are part of GDP in accordance with national accounting conventions. We thus
estimate the loss of GDP due to the loss of these tax revenues. The actual loss of GDP in
each recession is calculated as the cumulative difference between potential and actual
growth during the respective period.
The results of our calculations are reported in Table 1. The first two columns show that the
2001-2003 recession was more severe than the 2008-2009 one in terms of both the
magnitude of the exogenous shocks (the first column) and the actual loss of GDP (the
second column). However, for our purposes, the main point is given by the last column: in
the 2001-2003 recession the actual loss of GDP was 1.5 percentage points larger than the
loss attributable to the exogenous shocks. That is, the contractionary monetary and fiscal

14 We consider the years 2001 through the first half of 2003 for the first recession, and the third quarter of 2008
through the second quarter of 2009 for the second recession.
15 Credit constraints affecting consumers and the housing market were relatively mild in Israel during the
2008-2009 recession.

214 BIS Papers No 67

policy response at that time exacerbated the crisis. The opposite was the case in the
2008-2009 recession: the expansionary policy response to this crisis helped moderate its
effect on the economy, so that the actual loss of GDP is estimated to have been
0.9 percentage points smaller than the loss that would have been caused by the shocks
absent a policy response.
Our calculation probably underestimates the loss of GDP caused by the exogenous shocks
in the recent recession since we do not account for all of their financial effects, such as the
increase in the cost of credit for firms. This implies that the contribution of the policy
response to mitigating the crisis in Israel in 2008-2009 was probably larger than reported in
Table 1.

Table 1
The effect of shocks and policy on GDP
in the 2001-2003 and 2008-2009 recessions
Cumulative effect, percent of GDP

Loss of GDP due to
exogenous shocks
Actual loss of GDP
Policy effect
on GDP
2001-2003 7.6 9.1 -1.5
2008-2009 5.1 4.2 0.9

A separate calculation using the Bank of Israel DSGE model for the Israeli economy yields
similar results.16 It shows that implementing the contractionary policy of the previous
recession during the 2008-2009 recession would have resulted in a loss of 2.6 percent of
GDP. The DSGE calculation further implies that about three quarters of the loss are due to
monetary policy, and the remaining loss to fiscal policy. Our calculation, which derives the
effect of policy as a residual, does not allow for such decomposition between monetary and
fiscal policy.
Looking ahead – confronting the looming crisis
The possibility of a second global crisis triggered by the current events in Europe raises the
question whether Israel can repeat its monetary and fiscal policies that seem to have worked
well in the recent crisis. The answer is not straightforward.
As noted, Israel was affected relatively mildly by the 2008-2009 crisis and has recovered
rapidly, enjoying growth rates that were higher than in most advanced countries. It avoided
the large increases in public debt which many advanced countries experienced during the
crisis, and has also maintained fiscal discipline since emerging from the crisis. Hence, in
terms of debt- and deficit-to-GDP its situation compared with other advanced economies has
improved in recent years. It has also accumulated substantial amounts of foreign currency
reserves in recent years. Additionally, it has raised its interest rate several times in the last
two years, while most advanced counties have left it at a very low level. All this would seem
to suggest that Israel has ample room to pursue fiscal and monetary expansion – allowing

16 We thank Alon Binyamini for providing this calculation. The MOISE DSGE model is described in detail in
Argov et al. (2012).

BIS Papers No 67 215

automatic stabilizers to increase the deficit and bring its interest rate back to a near-zero
level.
However, several circumstances have changed since the last crisis. It seems that in view of
lessons learned from fiscal policies in the recent crisis, the debt crisis in Europe and the state
of public finance in the US, global financial markets are less tolerant to budget deficits than
they were in 2008-2009. Combined with the risk of contagion among markets, Israel’s fiscal
performance in recent years may not suffice to allow it to increase the deficit by as much as it
did in the recent crisis.
Concluding remarks
The experience Israel has had with two recessions over the last decade provides an
interesting example regarding the interaction between monetary and fiscal policy, and the
conditions under which policy makers can pursue counter-cyclical policies. The main lessons
are that favorable initial conditions and sound macroeconomic policy during normal times
expand the set of policy options available to policy makers at a time of need.
In this note we have focused on fiscal and monetary policy but the lesson applies more
generally: good policy in good times pays off handsomely in bad times. Good policy in that
respect means sustained fiscal discipline during the upside of the business cycle, which
credibly aims at an acceptable level of the public debt-to-GDP ratio and pursues a steadily
declining path of this ratio over time, along with monetary policy that promotes price stability.
Such policy is awarded by the tolerance of financial markets to fiscal and monetary
expansion during a recession: yields, risk premia and the exchange rate remain reasonably
stable as the central bank cuts the interest rate and the automatic stabilizers are allowed to
act, temporarily raising the budget deficit and the public debt. Our calculation shows that the
effect on GDP of such a policy response can be sizable.
Looking ahead in view of current developments abroad, Israel is relatively well positioned to
confront another crisis. It has some room to increase the deficit and cut the interest rate.
References
Argov, Eyal, Emanuel Barnea, Alon Binyamini, Eliezer Borenstein, David Elkayam and Irit
Rozenshtrom (2012), MOISE: A DSGE Model for the Israeli Economy, Discussion Paper
No. 2012.6. Research Department, Bank of Israel.
Bank of Israel (2001-2003), Annual Report.
Brender, Adi (2009), “Targets or Measures? The Role of the Deficit and Expenditure Targets
in Israel’s Fiscal Consolidation Efforts, 1985-2007,” The Israeli Tax and Economics Quarterly
Vol. 33 (129), pp. 7-33.
International Monetary Fund (2011), World Economic Outlook, September.
Strawczynski, Michel and Gila Weinberger (2011), The Cyclicality of Statutory Taxes in
Israel, Bank of Israel, mimeo.
Strawczynski, Michel and Joseph Zeira (2007), “Cyclicality of Fiscal Policy in Israel,” Israel
Economic Review, Vol. 5 (1), pp. 47-66.

BIS Papers No 67 217

Public debt and monetary policy in Korea
Dr Geum Wha Oh1
Abstract
This note reviews Korea’s fiscal policy and public debt management, and discusses some of
the constraints that bind the Bank of Korea in its conduct of monetary policy. Fiscal prudence
and low public debt in Korea have allowed monetary policymakers to focus on inflation
control without worrying about public debt dynamics. Such fiscal prudence is mainly
attributable to the strong and long-standing commitment to a balanced budget. However,
recently, fiscal policy has been managed in a more countercyclical manner within the
framework of medium-term fiscal planning. During the recent global financial crisis, Korea
implemented large-scale countercyclical fiscal policies to counteract the contractionary effect
of the crisis.
Meanwhile, the Korean government bond (TB) market has grown rapidly. Such a
development can potentially be helpful for implementing countercyclical fiscal policy against
crises, by acting as a low-cost funding source during crises. The Korean government has
made various efforts to develop an efficient bond market, such as introducing a system of
fungible issuance and opening the market to foreign investors. A recent phenomenon is the
increase of official investment in TB by Asian countries, including China and Thailand. The
opening-up of the financial market, however, has also complicated the implementation of
monetary policy because capital flow also affects market liquidity and the exchange rate. A
recent study on the transmission channel shows that the bank lending channel is the most
effective one, while the scope for other channels to operate (eg through the yield curve) is
limited. This result indicates that monetary policy may have been constrained in reacting to
inflationary pressure after the global crisis.
While it is true that public debt sustainability is currently not an issue in Korea, it is also true
that sovereign debt management could face significant challenges arising from population
ageing and ballooning social welfare expenditures. Other risk factors for public debt
dynamics are unfunded government liabilities, public agency or state-owned enterprise debt
that is not counted as sovereign debt, and the cost of unification.

Keywords: Fiscal policy, public debt management, capital inflow, monetary policy
JEL classification: E61, E62, H60

1 Bank of Korea

218 BIS Papers No 67

I. Introduction
Fiscal policy, public debt management, and monetary policy are closely interlinked. Both
fiscal and monetary policy enter the government’s inter-temporal budget constraint, and
directly affect the dynamics of public debt. As proven by the recent euro zone debt crisis,
public debt is a highly significant factor in financial stability particularly if it is unsustainably
high and constitutes a significant part of bank assets. But the converse is also true, in that
the euro zone debt crisis is in large part a legacy of the global crisis.
Public debt management and monetary policy have been fairly independent in Korea, and
are expected to remain so in the foreseeable future. Fiscal prudence and low public debt has
allowed monetary policy to focus on inflation control without concerns for public debt
dynamics. Korea’s fiscal policy and public debt management will pose no immediate threat to
monetary policy and financial stability. But, in the long run, they do face significant
challenges from population ageing and growing social demands for social welfare spending,
among other factors.
This note reviews Korea’s fiscal policy and public debt management, and discusses the
constraints faced by the Bank of Korea in conducting monetary policy.
II. Fiscal policy and public debt management: an overview
Institutional setup for fiscal policy
Korea’s fiscal policy and public debt management have been prudent since the 1980s by any
international standards. Indeed, Korea is one of the least indebted among the OECD
countries with public debt standing at about 33.4% of GDP at end-2010 and with no signs of
investor concerns over fiscal sustainability (Figure 1). Fiscal prudence has been aided by the
sustained solid growth of the economy and by stable financial conditions.

The backbone of fiscal prudence and transparency is the five-year medium-term fiscal plan
first introduced in 2004. The plan serves not only as the basic framework for each year’s
budget formulation and fund management but also as a basis for establishing targets for
fiscal balance and public debt over the medium term. Experts from the private sector

BIS Papers No 67 219

participate in the planning process, and the final plan is reported to the National Assembly
not later than 90 days prior to the start of a new fiscal year (which coincides with the calendar
year). Strictly speaking, the plan is not a fiscal rule as it is not legally binding. Rather, it is an
apparatus for encouraging fiscal discipline.
For transparency and administrative efficiency, fiscal activities are categorised into three
accounts: (1) the general account, (2) special accounts and (3) public funds accounts. The
general account covers general fiscal activities while special and public funds accounts are
for public projects funded by taxes and by levies earmarked for specific purposes. As of
2011, there were 18 special accounts and 64 public funds accounts. Public funds accounts
had been allowed a relatively high level of flexibility and autonomy for efficiency reasons but
only at the expense of reduced transparency and prudence in their management. For this
reason, they have required the review and approval of the National Assembly since 2002.
They are also subject to a need test every three years conducted by an evaluation committee
comprising a group of civil experts.
The consolidated fiscal balance, which includes all three fiscal accounts, recorded a surplus
in the mid-1980s after a series of deficits. Since then, it has fluctuated over the business
cycle, but remains broadly balanced once adjusted for cyclical factors. The consolidated
balance net of social security funds, which is known as the “managed fiscal balance”, has
remained close to a balance except for the periods of severe recession or financial crisis.
According to the budget, the consolidated and managed fiscal balances are expected to
record a surplus of 0.4% of GDP and a deficit of 2.0% of GDP, respectively, in 2011
(Figure 2).

Local governments enjoy only a limited degree of fiscal autonomy in their fiscal management,
given their role in executing various policies on the central government’s behalf in the area of
education, social welfare and industrial policy. Nevertheless, their autonomy has recently
been on the rise in line with growing fiscal decentralisation. Local governments’ debt net of
borrowing from the central government is only modest, standing at 1.6% of GDP or
18.4 trillion won at end-2010.
Prudent fiscal management, aided by solid economic growth and stable inflation, has
resulted in low government indebtedness. Public debt stood at 33.4% of GDP
(392.2 trillion won) in 2010, the majority of which is owed by the central government

220 BIS Papers No 67

(Figure 3). The debt ratio is low by international standards and also relative to the 60%
threshold of danger zone for EMEs as estimated by Reinhart and Rogoff (2010). Foreign
currency-denominated debt accounted for less than 3% of total public debt at end-2010,
some 84% of which is denominated in US dollars.

The general government’s financial assets have increased rapidly since the mid-1990s,
thanks to the surpluses of the national pension fund, and reached 73% of GDP by 2009,
which is more than twice its debt. The general government remains a net creditor even if the
assets of the national pension fund (which stood at 39.7% of GDP) are excluded. However, it
should be noted that government assets in the form of loans, stocks and proprietary equities
are illiquid.
Fiscal policy during crisis periods
Fiscal policy was procyclical until the 1997 crisis due mainly to a strong and long-standing
commitment to a balanced budget. In view of the critical role it played in the recovery from
the 1997 crisis, fiscal policy has since been managed in a more countercyclical manner and
also within the framework of medium-term fiscal planning. During the more recent global
crisis, Korea deployed a large-scale countercyclical fiscal policy to counteract the
contractionary effects of the crisis. To be specific, discretionary government spending
increased by 6% of GDP during 2008–10. This policy response proved successful in
mitigating the fallout from the crisis – indeed, Korea recovered quickly from the global crisis
with growth of more than 6% in 2010.
Success in crisis management was not free of cost, however. Public debt increased to 33.8%
of GDP by 2009, up from 30.7% of GDP in 2007, largely driven by swollen deficits, before
falling slightly to 33.4% in 2010 as the economy recovered. In order to strengthen fiscal
discipline and prevent rapid increases in the public debt ratio, the medium-term fiscal plan for
2011–15 targeted a balanced budget by 2013 and restricted expenditure growth to
4 percentage points below revenue growth until 2013 (Figure 4).

BIS Papers No 67 221

III. Sovereign debt market
Sovereign debt market structure
The government issues several types of bonds for its financing, including treasury bonds
(TBs), foreign exchange equalisation bonds (FEEBs), and national housing bonds (NHBs).

Table 1
Treasury bonds: maturities and issue/tender dates
Maturities Three years Five years 10 years1 20 years
Tender date Monday of the
first week
Monday of the
second week
Monday of the
third week
Monday of the
fourth week
Issue dates for
bonds
June 10 and
December 10
March 10 and
September 10 September 10 March 10
Note: (1) Inflation-linked bonds of 10-year maturity are issued on Wednesday in the third week of every month.

TBs are issued on a regular basis in four fixed maturities – for three, five, 10 and 20 years –
at the market interest rate through public tender (Table 1). The terms of TB issuance are also
standardised under the system of fungible issuance which was introduced in 2004 as part of
a package of development measures for the domestic bond market. Fungible issuance has
allowed the government to improve TB liquidity and cut its debt financing costs. FEEBs are
issued at irregular intervals when the need arises to curb excessive volatility in the exchange
rate. While they are typically issued overseas in foreign currency, the government issued
them in the domestic market in 1998 after the crisis. The outstanding volume of FEEBs stood

222 BIS Papers No 67

at about $7 billion at end-2011 with debt service being well dispersed between $0.5 and
$2.5 billion each year over the 2013–25 period (Table 2).

Table 2
Foreign exchange equalisation bonds: issuance and maturity
2003 2004 2005 2006 2009
Amount of issuance $1
billion
$1
billion
$0.4
billion
€0.5
billion
$0.5
billion
€375
million
$1.5
billion
$1.5
billion
Maturity (year) 2013 2014 2025 2015 2016 2021 2014 2019
Source: Ministry of Strategy and Finance.

Since March 2007, the 10-year inflation-linked bond has been issued using the three-month-
ahead consumer price index as the underlying index for inflation. This bond is also regularly
issued through public tender. As the volume issued has not been large, however, inflation-
linked bonds have only limited market liquidity and thus give little indication of inflation
expectations. Inflation-linked issues accounted for less than 2% of total issuance of TBs in
2010, and none were issued in 2009 (Table 3).

Table 3
Inflation-linked bonds
2007 2008 2009 2010 2011.1–11
Amount of issuance
(trillion won) 1.9 0.8 0.0 1.3 1.1
Ratio to total issuance of TBs
(%) 3.8 1.6 0.0 1.6 1.5
Source: Ministry of Strategy and Finance.

The volume of TBs issued increased somewhat after 2008 to finance the large-scale fiscal
stimulus aimed at mitigating the recessionary impact of the global financial crisis. The
amount of net issuance remained at 10–16 trillion won during 2007–08, but almost tripled to
30–45 trillion won in 2009–10. These large increases in TB issuance were accompanied by
changes in the maturity composition. Following the collapse of Lehman Brothers, the share
of TBs with maturity of five years or less jumped to from the previous 60% to 92% in
November 2008 before falling back to the pre-Lehman level by 2010 when financial instability
was largely resolved (Figure 5).

BIS Papers No 67 223

Despite the large increase in shorter-term bond issuance immediately after the global crisis,
the average remaining maturity of TBs increased rather than decreased. This was because
the financial market had been stabilised in a relatively short period and because the 20-year
bonds – first issued in 2006 – had increased as a share of total issuance. Nevertheless, the
effective (average) interest rate on TBs has declined significantly, aided by the
accommodative monetary policy stance after the global crisis (Table 4).

Table 4
Effective interest rate and average maturity of TBs
2003 2004 2005 2006 2007 2008 2009 2010
Average interest payment
(%) 4.76 4.38 4.57 5.05 5.18 5.37 4.64 4.48
Average remaining
maturity (years) 3.73 4.04 4.17 4.55 4.68 4.85 4.96 5.33
Source: Ministry of Strategy and Finance.

The global crisis affected the secondary TB market only moderately. In fact, the monthly
transaction volume of TBs in the OTC market increased sharply from about 100 trillion won
during 2007–08 to more than 180 trillion won in 2009, against the backdrop of a tripled
volume of new TB issuance in 2009 relative to previous years. It increased further to
266 trillion won in 2010–11, comprising mainly longer-term bonds with maturity of three years
or longer (Figure 6). Underlying these sharp increases were continued foreign capital inflows
into domestic bond markets after 2009.

224 BIS Papers No 67

The foreigners’ share in the total outstanding volume of government debt has rapidly
increased recently (Figure 7). Foreigners held 14.9% of the total or 58.8 trillion won at the
end of the third quarter of 2011 (as compared to the similar share of 31.2% in the stock
market in 2010). The remainder was split between financial institutions (55.6%) and public
agencies including the national pension fund (23.9%). Foreigners purchased mainly TBs with
maturities of three years or shorter in 2007 when foreign bond investment started to expand,
after which they seem to have gradually increased the share of longer maturity bonds in their
holdings (Figure 8).

BIS Papers No 67 225

Increased investor demand for TBs and the accommodative stance of the monetary policy
after the global crisis has resulted in a downward shift of the yield curve (Figure 9). The yield
curve became quite flat by the end of 2011, reflecting global economic woes and the influx of
global liquidity into domestic markets.

Debt instruments of the central bank
The Bank of Korea (BOK) issues and purchases its own debt instrument – monetary
stabilisation bonds (MSBs) – as part of its open market operations and liquidity control. The
issuance is subject to the ceiling on total outstanding volume set every three months by the

226 BIS Papers No 67

Monetary Policy Committee, the supreme organ of monetary policy. MSBs are issued in
13 standardised maturities ranging from 14 days to two years. As of the end of 2010, the
average remaining maturity is 0.8 years (Table 5). Short-term MSBs, which have maturity of
less than 28 days, are issued at irregular intervals depending on market conditions, while the
bidding date for MSBs with maturity of 28 days or longer is pre-fixed every month. Given that
TBs and MSBs are close substitutes from the perspective of investors, the Ministry of
Strategy and Finance (MOSF) and the BOK coordinate their issues to minimize any overlap
in the maturity structure between the two debt instruments. At present, TBs are issued with
maturities of three years or longer while MSBs are issued at shorter maturities.

Table 5
Remaining maturity distribution of MSBs
(End-2010)
Below one year (%) Between one and three years (%)
Above three years
(%)
Average remaining
maturity (yr)
63.5 36.5 0.0 0.8
Source: Bank of Korea.

The outstanding volume of MSBs fell in 2008 but has been on the rise since 2009, reflecting
the sterilisation operation against capital inflows to the stock market (Figure 10). The
outstanding balance of MSBs amounted to 13.9 % of GDP, or 163.5 trillion won, at end-2010
(Figure 11).

BIS Papers No 67 227

IV. Capital flows and monetary policy
Following the 1997 financial crisis, the capital account was widely liberalised to a degree
comparable to that of advanced economies. Since then, monetary policy has been
increasingly subject to the influence of capital flows on market liquidity and the exchange
rate. For example, monetary policy was tight during 2006–07 but asset price inflation
remained high partly due to a surge in capital inflows driven by abundant global liquidity. The
outbreak of the global crisis triggered an abrupt and sharp reversal in capital flows, which
soon gave way in its turn to a renewed surge of inflows. Specifically, foreign investors
withdrew $62.4 billion from the stock market during 2007–08 before reinvesting $48.5 billion
over the next two years. Bond investment recorded net outflows during the second half of
2008 but these gave way to a $43.5 billion inflow during 2009–10. Such swings in capital
flows were repeated in 2011, albeit to a lesser degree, in line with the heightened market
uncertainty emanating from the euro area fiscal crisis (Figure 12).

228 BIS Papers No 67

More recently, official investment in TBs by Asian countries including China, Thailand,
Malaysia and Singapore has increased, which has offset in part the bond outflows driven by
European investors. This new development seems to reflect the interest rate differentials
between home and abroad, the diversification needs of official investors related to foreign
reserve management, and the positive market perception of the safety of Korean TBs. At the
end of 2011, the outstanding balances of stock and bond investment funds stood at
$41.3 billion and $25.0 billion (or 14% and 34% of total foreign investment), respectively.
Monetary policy was often complicated by these volatile and large capital flows, which
distorted the transmission channel and created unpleasant policy trade-offs. While the policy
rate (and the overnight call rate) have been raised several times since 2010, longer-term
market interest rates have been on the decline since 2009. As a result, the spread between
long and short rates narrowed to less than 1 percentage point by end-2011 – for example,
the spread between the overnight call rate and the three-year TB yield fell from
0.64 percentage points in 2006 to 0.55 percent points in 2011, a level even lower than
observed in the pre-crisis period. In contrast, the risk premium between three-year corporate
and treasury bonds has remained significantly higher during the post-crisis period than in the
pre-crisis period but nevertheless declined from 4.8 percentage points in December 2008 to
less than 1 percentage point by 2011 (Figure 13).

While many factors including negative outlooks for the global economy may have contributed
to the opposite movements in short and long rates, the prime suspect has been large capital
inflows, suggesting that the so-called Greenspan conundrum is no longer a phenomenon
confined to advanced countries. A recent study on the transmission channel shows that the
bank lending channel (through changes in bank lending rates) is the most effective one while
the scope for other channels (eg through the yield curve) to operate is limited. These results
indicate that monetary policy has been constrained in reacting to inflationary pressure after
the global crisis.
Though there have been huge inflows of foreign capital since the global financial crisis, there
is no clear evidence that they have eroded Korea’s international competiveness. The level of
the effective exchange rate is somewhat lower compared with that of the pre-crisis period.
The growth rates of the monetary aggregate, which have shown a downward trend, seem to
indicate that the capital inflows have not expanded domestic liquidity to any great extent.

BIS Papers No 67 229

V. Further considerations
Long-run fiscal challenges
Public debt sustainability is currently not an issue which requires immediate policy attention
or corrective action. And the risk of fiscal dominance is only a remote possibility. But
sovereign debt management will face significant challenges arising from population ageing
and ballooning social welfare expenditures (Figure 14). Social welfare expenditure has risen
rapidly to almost 9% of GDP by 2010, up from less than 5% of GDP in the early 2000s while
the dependency ratio is projected to rise above 50% by the late 2020s (Figure 15).

230 BIS Papers No 67

Other risk factors to public debt dynamics are unfunded government liabilities (such as future
pension payments), public agency or state-owned enterprise debt which is not counted as
sovereign debt, and the cost of unification. It is very difficult, if not impossible, to estimate the
cost of unification with any precision but the impact of unification on the government budget
would in all likelihood be enormous.
The state-owned enterprises (SOEs) are making profits (eg 2.9 trillion won in 2010) and
maintain a positive net asset position (assets 444.6 trillion won, debt 271.8 trillion won) and
there is no sign of imminent financial problems. However, the SOEs have increased debt
rapidly in recent years in order to finance large-scale multi-year investment in infrastructure
and the energy sector. This may potentially give rise to a public debt problem and needs to
be monitored carefully.
Financial risks to the Bank of Korea
The Bank of Korea is subject to financial risks as it issues its own debt in domestic currency
and purchases foreign assets. Financial risks involve interest rate, credit and currency risks.
Interest payments on MSBs have typically accounted for more than half of the total operation
cost of the Bank (Table 6). Moreover, the Bank incurred a large loss in 2005–07 because of
the Korean won’s steep appreciation. Acknowledging such financial risks, the recent revision
of the Bank of Korea Act allows the Bank to maintain a higher reserve balance (to be used to
compensate for losses) by withholding up to 30% of its operational profit (previously 10%).

Table 6
Operation balance of BOK
(Billion won, %)
2001 2002 2003 2004 2005 2006 2007 2008 2009 2010
Operation cost (A) 6,243 6,223 6,405 7,899 9,344 11,945 14,081 16,916 12,915 11,876
Interest payment
for MSB (B) 4,873 4,802 4,963 5,584 6,144 6,806 7,478 7,200 6,228 6,010
(B/A, %) (78) (77) (77) (71) (66) (57) (53) (43) (48) (51)
Operation profit 6,124 4,149 3,198 35 –1,919 –1,762 –491 3,363 3,823 4,560
Appreciation rate
of won to dollar –14.2 3.1 4.7 4.0 10.5 6.7 2.8 –18.7 –15.8 9.4
Source: Bank of Korea.

Household debt and monetary policy
While sovereign debt is low by international standards, household debt has recently
increased to an alarming level. The ratio of household debt to disposable income stood at
132% at the end of 2010, a level broadly comparable to the corresponding figures of many
advanced economies, including the United States (Figure 16). In addition, the characteristics
of this debt are less than benign. The majority of household debt comprises bullet loans from
banks and non-banks at variable interest rates and a relatively short maturity (typically of
three years). This suggests that households are exposed to liquidity risk and remain
vulnerable to income shocks.

BIS Papers No 67 231

The recent increases in the household debt ratio seem to reflect not only consumption
smoothing at a time of sluggish income growth and low interest rates but also financing of
home purchases in anticipation of rising housing prices (particularly in non-metropolitan
areas). The accommodative monetary policy stance, maintained since the global crisis, may
also have influenced the increase in household debt.
At present, the underlying macrofinancial risk related to high household indebtedness seems
to be less than one would infer from the debt-to-disposable-income ratio itself. To be specific,
according to a household financial survey data, most household debt is owed by middle- to
high-income groups (ie the top three quintiles), whose repayment capacity is assessed as
relatively strong (Figure 17). Moreover, these groups have financial assets that can be drawn
down if necessary for debt repayment. Indeed, the household debt ratio is estimated to fall to
below 80% if measured against the broader financial resource base comprising disposable
income and long-term time deposits. Last but not least, housing loans have been subject to
tight prudential regulations, including LTV and DTI restrictions since 2002.

232 BIS Papers No 67

While there seems to be no immediate threat to financial stability at present, high and rising
household debt, if persisting over a long period, would in all likelihood lead to increased
systemic risk and undermine the scope for monetary policy to preserve price and financial
stability. In view of the high uncertainty in global financial markets and real activities, Korea’s
monetary policy has so far remained accommodative despite a five-step increase in the
policy interest rate since mid-2010. And the likelihood of monetary tightening in the near
future seems low in view of the deteriorating global outlook. In this light, the supervisory
authority has taken macroprudential action to slow the speed of household debt increases, if
not prevent it from rising further.

BIS Papers No 67 233

Banco de México and recent developments
in domestic public debt markets
José Sidaoui, Julio Santaella and Javier Pérez1
Abstract
This paper describes major policy actions that have recently contributed to the development
of the Mexican domestic-currency debt market, and concomitant benefits. Among the most
important are a significant reduction in exchange-rate exposure and a decline in refinancing
risk for the government and private sectors alike. Another positive outcome of the
development of government securities markets has been investor base diversification. This
paper explains how capital inflows have translated into larger and more stable foreign
investor participation in local debt markets. Empirical evidence presented suggests that
these capital inflows have had positive funding implications, lowering both interest-rate levels
and volatility.

Keywords: Public debt management, financial markets, Latin America
JEL classification: H63, E63, N26

1 The authors thank Dorothy Walton and Armando Gonzalez Torres for their assistance. The opinions in this
paper are solely those of the authors and do not necessarily reflect those of Banco de México.

234 BIS Papers No 67

1. Introduction
The development of financial markets has yielded several benefits for the Mexican economy.
Among them are a significant increase in the share of domestic-currency liabilities in the
total, which has reduced the exchange-rate exposure of government and private-sector debt
servicing. Other developments are a lengthening of the yield curve, which has limited
refinancing risk for the government and provided reference rates for private issuers, and an
ample peso interest rate swap (IRS) market, allowing investors a better risk distribution. More
liquid and deeper markets, in turn, have contributed to a more efficient allocation of
resources by reducing transaction costs. They have also been essential in the economy’s
ability to take advantage of capital inflows by channeling them towards more productive
uses.
The law governing the central bank, Banco de México, requires that in addition to the
mandate to safeguard price stability, the bank must promote the healthy development of
financial markets and serve as the fiscal agent of the federal government. Hence, in
coordination with the government and other financial supervisors, it has played a key role in
putting into place the building blocks of the institutional framework that has enabled the
development of the domestic debt markets. Undoubtedly, the current liquidity and stability
that the domestic public securities market enjoys, even amid severe external financial
turmoil, result from this progress, as well as from sound macroeconomic and financial
stewardship.
This paper describes major policy developments in recent years and identifies some of their
benefits. We build on previous research and document the latest advances in government
securities markets, notably, the lengthening of the sovereign yield curve and the increase in
the share of local-currency government securities in the total. We discuss Banco de México’s
role in this process. One outcome of the development of government securities markets has
been the diversification of the investor base. This has been accompanied by capital inflows
that have translated into higher foreign participation in local debt markets, apparently
investors with more stable profiles. Empirical evidence suggests that these capital inflows
have had positive funding implications, decreasing interest-rate levels and volatility.
Adequate fiscal and monetary policies and a respect for freely functioning markets, together
with international recognition (e.g., the inclusion of Mexican government securities in
Citigroup’s WGBI), have all been pivotal to these results.
This note is structured as follows. Section 2. discusses various policy actions taken by the
authorities to develop the government securities market. Section 3. briefly describes the
recent evolution of interest-rate derivatives, which have served as a complement to the
government securities market. Section 4. summarizes how these actions brought about
further development of local debt markets. Section 5. advances the hypothesis of a change
in investor profile towards more stability. Finally, section 6. offers conclusions.
2. Policy aimed at developing local debt markets
For several years now, Mexican policymakers have been committed to developing domestic
debt markets and have taken steps in this direction as conditions have made it possible.
Previous research (see, for instance, Sidaoui, 2002; Pérez-Verdía and Jeanneau, 2005)
refers to many important institutional achievements over the last decade. This section
describes how sound macroeconomic policies, together with financial reforms, have
contributed to the development of local debt markets.

BIS Papers No 67 235

2.1 Sound macroeconomic policy
Sound fiscal, monetary and debt management policies pursued since the Mexican crisis of
1995 have been conducive to higher macroeconomic stability and to the development of
financial markets. Pérez-Verdía and Jeanneau (2005) show how most public finance
indicators improved since 1995: both fiscal-deficit-to-output and public-debt-to-output ratios
followed decreasing paths between 1995 and 2005. As a consequence of the 2008 financial
crisis, the Mexican government provided fiscal stimulus to moderate the dampening effect of
external conditions on the domestic economy. Although this stimulus temporarily weakened
public finance numbers, the financial position of the government remained sound. By the end
of 2010, total public debt as a proportion of GDP was 32.2% (versus 21.8% in 2005), while
the narrowly defined fiscal deficit was 2.8% of GDP (which compares to 0.1% in 2005).2
Pérez-Verdía and Jeanneau (2005) show as well inflation converging towards Banco de
México’s target of 3%.
2.2 Minimal market intervention
The Mexican authorities are convinced that free markets are best suited to allocating
resources and determining prices. Therefore, they have facilitated the development of the
debt markets by liberalizing almost every segment of the financial sector. Furthermore,
consistent efforts have been made to provide markets with sound legal, operational, and
institutional infrastructure. Government interference with the market price-discovery
mechanism has been avoided. On the very few occasions there has been a market
intervention, it has been under extreme market stress and limited to providing liquidity.3 No
capital controls have been imposed, even in view of massive capital inflows in the recent
past. Instead, the strategy to cope with capital inflows has been a combination of various
policies believed to increase the chances of attracting potentially longer-term investors, thus
decreasing the likelihood of sudden stops without provoking severe market distortions. The
hallmarks of the strategy are a stable macroeconomic outlook together with predictable and
reliable policymaking, and sterilization by the central bank of the impact of capital surges on
domestic liquidity (see section 2.6 for additional details). As a result of these and other
events where the propensity to intervene has been tested, the Mexican government has
earned a reputation for promoting the independent and orderly functioning of markets.
2.3 Pension system reforms
Debt market development cannot proceed unless there is a dependable supply of long-term
loanable funds from institutional investors. Reforms to the Mexican pension system have
strengthened the demand for government securities. The transformation in 1997 of a pay-as-
you-go system into an individual contributory pension system for private workers resulted in a
surge of large pension funds. Later on in 2007, the pension system of public employees went
through a similar reform which further increased assets managed by pension funds, hence
stimulating additional demand for securities. By the end of 2010, the net assets of these
funds amounted to 1.4 trillion pesos, equivalent roughly to 10% of GDP. The resources under

2 Source: Finance Ministry (SHCP) public finance and debt statistics.
3 During periods when liquidity tends to dry up, causing high market volatility, authorities have introduced
auctions to sell foreign currency. The auction mechanism is pre-announced to the public, it is for a limited
amount of foreign currency, and it starts at a floor of 2% above the previous day’s reference exchange rate, or
the Fix. This scheme was used in the late 90s, during the Lehman crisis in 2008, and, more recently, since
November 30, 2011. Additionally, a few extraordinary auctions of foreign currency were carried out on days of
particularly scarce liquidity in 2008 and 2009.

236 BIS Papers No 67

their custody have grown very fast; just a few years back, in 2004, they were 6.5% of GDP.
Almost half of their assets are invested in government securities, which account for 12% of
the total outstanding. Aside from being major investors in local government debt, pension
funds have contributed to the demand for long-term securities. This is a natural result of their
investment horizon. As of October 2011, the average duration of government securities they
held was 7.3 years (the average maturity of outstanding government debt is 4.5 years).
2.4 Improved securities clearing and settlement systems
Reliable clearing and settlement systems are key components for the sustainable operation
of financial markets. Banco de México, the National Banking and Securities Commission
(CNBV), and the National Securities Depository Institute (Indeval) have worked closely
together to develop centralized and automatized clearing and settlement systems. The legal
framework has been enhanced to provide certainty to market participants carrying out market
transactions. Indeval, the centralized securities custodian, offers services such as the
settlement and transfer of securities, collateral management, securities lending, and the
infrastructure for repo operations. Currently, operations are settled on a near real-time
Delivery versus Payment (DvP) protocol. Furthermore, Indeval is linked to foreign clearing
and settlement systems such as Euroclear and Clearstream in order to facilitate trading with
securities issued abroad by Mexican firms and institutions (Jiménez Vázquez, 2011).
2.5 Completing the market’s information set
Efficient resource allocation depends on reliable prices. In this regard, Banco de México has
been providing reference interest rates and securities prices to the market for a long time. On
a daily basis, the central bank publishes the peso/dollar Fix rate (an auction-determined
exchange rate for U.S. dollar liabilities payable in Mexico), as well as the 28- and 91-day
reference interest rate, or TIIE (the 182-day TIIE is published once a week). As documented
by Sidaoui (2002), the TIIE has become a widely used benchmark for loans, yields, and as
an underlying rate for futures and swaps. Moreover, daily, Banco de México publishes the
price vector it uses to mark to market its holdings of government securities.4 In addition,
private price vendors have been authorized in order to preclude conflicts of interest among
market participants.
Furthermore, in a permanent effort to improve upon transparency and information quality,
Banco de México publishes an array of financial and economic indicators which have been
progressively standardized to meet international criteria (e.g., IMF, CUSIP). Also, the central
bank publishes its policy stance and various reports on inflation, monetary policy, and the
financial system on a timely basis and in accordance with a pre-determined calendar. These
announcements provide news agencies with updated information.
To improve the predictability of the issuing patterns for government securities, the
government, Banco de México, and the Institute for the Protection of Bank Savings (IPAB,
the nation’s deposit insurance agency),5 preannounce their issuance program on a quarterly
basis. The program includes details on the securities as well as on the amounts to be
auctioned each week. In addition, the government has published debt guidelines on a yearly
basis since 2004. This has the advantage of allowing investors to estimate the supply of
securities ahead of time and to adjust their investment strategies accordingly.

4 Banco de México averages information from private price vendors to create its own price vector.
5 The deposit insurance agency regularly issues debt, which is viewed by the markets as quasi-government
debt.

BIS Papers No 67 237

2.6 The use of government securities as monetary policy instruments
Although the central bank can issue its own paper to carry out open market operations (see
below), it has chosen to use government paper instead. This decision was made in order to
foster the development of the government debt markets. Banco de México is prohibited by
law from financing the government. Thus, a mechanism was designed that allows it to use
government paper without acting as financier (see Box 1). In the last couple of years, there
has been a significant increase in reserve accumulation, and the use of government
securities to sterilize the resulting liquidity has proved effective. In what follows, we describe
how the central bank has alternated the use of its own and government securities for liquidity
management.
Banco de México only used government securities to manage liquidity until 2000.
Nevertheless, in order to test the market acceptance of central bank paper that year, the
bank issued its own securities and used them to implement monetary policy. At that time, a
market niche was perceived due to the fact that investors were looking for an instrument to
reduce price sensitivity to interest rates because they feared episodes of higher interest-rate
volatility. Thus, Banco de México began issuing Monetary Regulation Bonds (BREMs), which
were bonds indexed to the daily overnight interbank lending rate.6
Six years later, the central bank and the government reached the conclusion that it was to
their mutual benefit to use one security. They substituted the BREMs with government paper
with identical characteristics, Bondes-D. An objective of the switch was to facilitate the
government’s interest-rate and currency-exposure reduction strategy (to be described in
section 4.1). Another aim was to enhance the liquidity of these securities given the fact that
both entities were to use the same paper for their financing needs.
Currently, the federal government auctions Bondes-D every second Tuesday, while Banco
de México does so every Thursday. These floating-rate bonds are completely fungible from
the market’s perspective.7 Since both risk-free issuers use the same instrument, predation is
avoided. At the same time, the use of Bondes-D has two appealing advantages: it preserves
the government’s floating-rate niche, and the markets of other government securities are not
distorted as a consequence of Banco de México’s liquidity management operations.
2.7 The issuance of warrants
A part of the active debt management strategy has been to identify opportunities to cater to
particular investor needs. In November 2005, the government realized there was demand
from investors exposed to Mexican sovereign debt in dollars for an instrument that had an
embedded option for the same credit risk in pesos. Since then warrants have been sold on
different occasions to fulfill this demand. These instruments entitle their holders to exchange
securities denominated in foreign currency for securities denominated in local currency. At
the time of issue, the warrants establish the ratio at which, during their validity, specific
securities may be exchanged. The warrants give investors the option to hold sovereign risk
constant, but at the same time to manage their currency exposure. The option granted by the
warrants becomes profitable as the spread between foreign-currency and local-currency
yields narrows. As section 4.1 will explain, to date, the warrants have contributed to reducing
the government’s foreign-currency liabilities.

6 It should be noted that these changes have been coordinated with the government in order to protect its debt
segments and to avoid distortions in the securities market.
7 The reason for making this distinction is to provide the market with information on how much the government
and the central bank are issuing each time.

238 BIS Papers No 67

Box 1
The use of government securities for liquidity management
Direct sales of government securities are one of the instruments Banco de México (henceforth “the
Bank”) uses to manage liquidity (mainly to sterilize international reserve accumulation). Legal
support for these operations stems from articles 7 and 9 of the law that governs Banco de México.
Article 7 entitles the Bank to deal with this class of securities. Article 9 contains accounting
guidelines for the operations:
“Banco de México shall not lend securities to the Federal Government nor purchase
securities from it, except when purchases of securities that are payable by the
Government comply with one of the two following conditions:
I. When said purchases are covered by cash deposits, made by the Government in the
Bank with the proceeds of the placement of said securities, and which may not be
withdrawn before their maturity date; the amounts, terms and yield on these deposits
must be equal to the amounts, terms and yield of the securities being traded; [ ]”
Therefore, when the Bank purchases securities from the government for liquidity management
purposes, its balance sheet is affected as follows. On the liability side, the Bank constitutes a cash
deposit in favor of the government; the deposit cannot be withdrawn before the securities mature.
These deposits are labeled in the Bank accounts as a “government-securities monetary-regulation
deposit” (henceforth “deposit”). On the asset side, the Bank records the holdings of purchased
government securities.
Once the securities are on the books of the Bank, they are marked to market daily. This procedure
ensures that before the securities are sold to the market, all the holdings on the asset side are
perfectly matched by the deposit on the liability side. Further, if the securities are coupon bonds, the
accrued interest of current coupons is computed and provisioned both on the asset side
(government securities holdings) and liabilities (deposit) side of the balance sheet. Whenever
coupons are due, the Bank simultaneously debits the provisioned interest and credits the
government account.
When the Bank sells securities to the market, it creates an imbalance between its assets and
liabilities. This happens because the deposit is held as a liability until the securities mature. In the
case of coupon bonds, an additional imbalance arises from the fact that the Bank continues to
credit the deposit with the accrued interest on current coupons from securities sold, but ceases to
provision this interest on its asset side. When a coupon matures, the Bank transfers resources to
the government account to pay the bond holders.
Similarly, when securities mature, the Bank debits unsold securities at face value from the
government’s account, and credits the government with the face value of securities sold. Finally, the
Bank pays bond holders by debiting the account of the government. The re-pricing of securities as
well as interest is then reflected in Banco de México’s P & L.

2.8 Additional measures to enhance the liquidity of government securities
In order to foster liquidity, several other measures have been adopted. They include market
makers,8 a strips market,9 and the reopening of previously issued securities. Additionally,

8 As Pérez-Verdía and Jeanneau (2005) explain, market makers are financial institutions that commit themselves
to bid for a minimum amount of securities at primary auctions of government securities, to always make two-
way quotes for a minimum amount of fixed-income securities, and to maintain a cap on the bid-offer spread. In
exchange, market makers are entitled to take part in green-shoe auctions, to hold regular meetings with
government debt-management authorities, and to have access to Banco de México’s securities lending facility.
9 The strips program was launched in 2005 to foster liquidity in the secondary market. Nevertheless, very few
bonds have so far been stripped, and turnover of these securities is minimal (García Padilla, 2011).

BIS Papers No 67 239

several exchanges and repurchases of securities have been carried out to smooth the
maturity profile or to increase the liquidity of particular issuances. More recently, to support
the potential inclusion of Mexican government securities in the WGBI10 (which took place in
October 2010), a syndication program was introduced in February 2010. One goal of this
program was to furnish new issues with an acceptable initial total outstanding amount, thus
enhancing their liquidity from the outset and enabling them to be included in global fixed-
income indices.
The creation of a securities-lending scheme provides another way to add liquidity to the
market for government paper. To illustrate this point, consider a long-term bond owned by a
pension fund. Typically, given its investment horizon, the pension fund would very likely hold
this bond until maturity. However, by lending this security to another investor, the pension
fund would earn a fee and get the bond back upon expiration of the lending agreement. As a
consequence of similar lending operations, the liquidity of the market for the securities
increases.
The central bank proactively carried out regulatory modifications to facilitate repo
transactions and securities lending. In particular, a master contract for both operations was
designed in 2007 in accordance with international guidelines from the Public Securities
Association, the Bond Market Association, and the Securities Industry Association. Further,
to foster the development of the private securities lending market, Banco de México
increased the cost of its securities lending facility. Currently, there are two privately owned
firms that provide securities lending intermediation: Accipresval, owned by Citibank, and
Valpre, owned by Indeval.
3. The development of derivatives markets
In principle, derivatives add liquidity and depth to government securities, since they offer
hedging possibilities for different portfolios, therefore expanding the range of investors
demanding the underlying assets. Hence, Banco de México, in coordination with other
regulators, made institutional arrangements to provide the legal and operational framework
for a derivatives exchange. MexDer, the Mexican derivatives exchange, was created in 1998
in order to provide a standardized environment for trading commonly used derivatives.
Asigna, MexDer’s clearinghouse, guarantees that obligations arising from transactions in
MexDer will be honored. More recently, in 2006, Banco de México revisited the regulation
that establishes which underlying assets are eligible to become derivatives and the type of
market participants that may trade these securities. Furthermore, any intermediary that
wishes to participate in this market has to comply with minimum requirements the central
bank imposes for management, operations, and internal governance.11

10 The WGBI (World Government Bond Index) is an index of fixed-income sovereign securities from 24 countries
that is constructed by Citigroup. Eligibility criteria include a minimum total outstanding amount of each bond (at
least USD 20 billion a year), a minimum credit rating (BBB- by Standard and Poor’s or Baa3 by Moody’s), and
low barriers to entry. Mexican government bonds already met the last two requirements: long-term local
sovereign Mexican debt is rated A by Standard and Poor’s and Baa1 by Moody’s, and it can be settled on
Euroclear. As of March 2011, the market value of assets linked to the WGBI was approximately
USD 18.1 trillion (Tapia Rangel, 2011). Mexico was the first Latin American country to be included in the
WGBI.
11 There is a guideline, known as “Banxico’s 31 points,” which states minimum requirements by which institutions
trading derivatives must abide, regulated by Banco de México (Circular 4/2006).

240 BIS Papers No 67

Derivatives markets have complemented the government securities market. There are no
comprehensive measures of the peso OTC interest-rate derivatives market.12 However,
interest-rate swaps (IRS) within Mexican financial institutions and between these institutions
and other investors, which are systematically reported to Banco de México, could serve as a
lower bound for the volume operated in the peso OTC interest-rate derivatives market, since
a significant fraction is traded off-shore. Using this information, Figure 1 illustrates the size of
the peso IRS market, as measured by outstanding IRS contracts reported to the central
bank, and the traded volume of these derivatives. Although the market shrank during the
recent financial crisis in 2009, it should be noted that traded volume has gradually recovered
over the past three years.
Figure 1
Peso-denominated IRS reported by financial institutions
to Banco de México: outstanding contracts and traded volume
In billions of pesos: 2005-2011

Source: Banco de México
Filter: 20-day moving average
NB: Total outstanding IRS are calculated by adding the absolute
value of both legs, since the goal is to gauge the value of the total
number of outstanding contracts, regardless of which party is on
the other side of the contract.
Perhaps the most interesting development in the derivatives market has been the
lengthening of the horizon of interest-rate derivatives investors have access to. This has
allowed investors to pair derivatives with investment strategies on the full span of the yield
curve. In MexDer, futures on the 28-day TIIE are by far the most widely traded13 interest-rate
derivative and are available up to a five-year horizon. However, MexDer offers other
instruments that may well meet investors’ needs to hedge virtually every portfolio of
government securities: 2- and 10-year interest rate swaps, as well as futures on 91-day
Treasury certificates (Cetes), and on 3-, 5-, 10-, 20-, and 30-year bonds. Additionally, futures

12 Today there is an ample IRS market both in nominal and real (inflation-linked) interest-rate instruments.
13 According to MexDer’s website, open interest in TIIE 28 futures accounted for more than 96% of open interest
in interest-rate derivatives by the end of October 2011.

BIS Papers No 67 241

on UDIs, the inflation-indexed units on which real-rate government securities are based, are
also available. Data on the OTC IRS market collected by Banco de México reflect as well the
usage of longer-term derivatives. In fact, as depicted by Figure 2, the proportion of longer-
term swaps, mainly in the “� 10 year” bucket, has steadily increased, while the proportion of
short-term swaps (less than 1 year) has become smaller.
Figure 2
Outstanding peso-denominated IRS reported by financial institutions
to Banco de México by maturity
% of total: 2005-2011

Source: Banco de México
Filter: 20-day moving average
NB: Total outstanding IRS are calculated by adding the absolute
value of both legs, since the goal is to gauge the value of the total
number of outstanding contracts, regardless of which party is on
the other side of the contract.
4. The advantages of a well developed government securities market
Sound policymaking and the growth of complementary markets have enabled the Mexican
government to pursue a strategy to mitigate vulnerabilities by reducing interest rates and
currency risks. They have also helped to finance the government’s deficit and to develop
domestic securities markets. We next describe how the policy actions explained in the
previous sections have contributed to this strategy.
4.1 Lower currency risk exposure
The growth of domestic financial markets has allowed the Mexican public sector to stop
relying on foreign markets for its financing requirements. The debt denominated in foreign
currency has declined from 35% of local-currency debt at the beginning of 2002 to 16% in

242 BIS Papers No 67

late 2011 (Figure 3).14 This has reduced public-sector exposure to exchange-rate risk, a key
financing vulnerability. In fact, since 2001, the government has had the capability to
completely finance its deficit in the domestic markets at its choosing.
Figure 3
Ratio of foreign-currency to domestic-currency government securities
%: 2002-2011
��
��
��
��
��
��
��
��
��
��
��
��
��
��
��
��
��
��
��
��
��
��
��
��
��
��
��

Source: Banco de México and the Finance Ministry (SHCP)
The reduction in external debt was achieved through several actions. First, the government
was able to prepay an important portion of its external debt between 2005 and early 2008
thanks to two factors: its ability to get funding in the local markets at reasonable terms, and
its ability to buy a large amount of dollars (USD 25.5 billion) without distorting the foreign
exchange market. The foreign currency was purchased from the central bank’s international
reserves at market prices (at the Fix). By using international reserves, the government and
the central bank avoided sending unintended signals that could have been read mistakenly
by the market as changes in the stance of monetary or exchange policy.
Second is the issuance of warrants described in Section 2.7. These warrants granted
investors wishing to incur some sovereign risk but limit their currency exposure the option of
exchanging foreign-currency-denominated government bonds for nominal and inflation-
indexed local securities. As a result of this program, the government has reduced its external
liabilities by nearly USD 5 billion since 2005.
Third, all of the maturing foreign-currency bonds were replaced with debt issued in the
domestic markets.
4.2 Lower interest-rate risk exposure
The government has continuously sought to alter the mix between foreign-currency-
denominated and floating-rate securities on the one hand, and peso-denominated fixed-rate
or inflation-indexed securities on the other, in favor of the latter. Figure 4 clearly illustrates
this point: the share of fixed-rate peso-denominated securities has more than doubled from
28% of the total in 2002 to 60% of the total in late 2011. Lower and more stable inflation has

14 The pace at which this share has decreased slowed in 2009, when the federal government turned to external
markets to take advantage of better funding conditions.

BIS Papers No 67 243

arguably contributed to the increase (Pérez-Verdía and Jeanneau, 2005). However, more
stability has not only contributed to a larger share of fixed-rate securities, but also to
extended government-debt maturities.
Figure 4
Outstanding federal government securities by currency and rate type
% of total: 2002-2011
��
���
���
���
���
���
���
���
���
���
����
��
��
��
��
��
��
��
��
��
��
��
��
��
��
��
��
��
��
��
��
����������������
�������������
����������������������
����������

Source: Banco de México
4.3 Lower refinancing risk for public and private issuers
The government has gradually increased the length of its nominal yield curve from 6 months
in 1995 to 30 years since 2006. This action has reduced its refinancing risk and provided the
market with long-term reference rates. A major benefit is that private issuers have been able
to rely more on the domestic markets for their financing needs. Debtors have extended their
debt maturities, which in all probability has also decreased their refinancing and foreign-
exchange risks (see Figure 5).
Figure 5
Federal government debt average maturity (in years) and
the distribution of private securities by maturity
% of total: 2002-2011
�
�
�
�
�
�
�
�
��
���
���
���
���
���
���
���
���
���
����
��
��
��
��
��
��
��
��
��
��
��
��
��
��
��
��
��
��
��
��
������
��������
��������
��������
��������
ч�ϭ�LJƌ
��������������� �������
�����

Source: Banco de México

244 BIS Papers No 67

Indeed, private and public corporations’ domestic debt issuance has seen a significant
increase since 2000 (see Figure 6). This has been supported, in part, by a legal reform that
created a new instrument (the “Certificado Bursátil”), making the process to access markets
much easier. The new law eliminates red tape and simplifies the issuance process:
authorities grant an “umbrella permit” to place debt and the choice of instrument and maturity
is left to the issuer’s discretion. In contrast, the previous procedure required case-by-case
authorization.
Figure 6
Outstanding securities issued by the domestic private sector
In billions of pesos: 1996-2011

Source: Banco de México
Excludes securities issued by financial institutions.
Among the debt instruments that have benefited from longer risk-free reference rates are
mortgage-backed securities. The development of this market allowed mortgage
intermediaries to broaden their funding sources (see Figure 7).
Figure 7
Mortgage-backed securities: total outstanding
(in billions of pesos) and remaining maturity
In years: 2002-2011
�
�
��
��
��
��
��
�
��
���
���
���
���
���
��
��
��
��
��
��
��
��
��
��
��
��
��
��
��
��
��
��
��
��
�����������������������
��� �������� �������������

Source: Banco de México

BIS Papers No 67 245

4.4 Higher liquidity
During the recent financial crisis, global liquidity almost seized up, and this had a sizeable
effect on the turnover of several sovereign debt markets (Figure 8). The Mexican market was
not an exception. Figure 9 shows turnover and volume of government bonds from 2003 to
2011. Traded volume increased until 2007 and then sharply decreased in 2008. Turnover
increased during 2007, reversing a downward trend documented by Pérez-Verdía and
Jeanneau (2005). However, turnover also fell during 2008. As global liquidity has begun to
return, both turnover and traded volume have rebounded, although they have not yet
reached their pre-crisis levels.
Figure 8
Turnover ratio (% of total outstanding)
of selected government debt markets:
2002-2011
Figure 9
Turnover ratio (% of total outstanding) and
traded volume (in billions of pesos) of
Mexican government bonds:
2002-2011
���
���
���
���
���
���
��
��
��
��
��
��
��
��
��
��
��
��
��
��
��
��
��
��
��
��
�����
���������
��������

�
��
��
��
��
��
��
��
�
�
��
��
��
��
��
��
��
��
��
��
��
��
��
��
��
��
��
��
��
��
��
��
��
�������������� ������������ �������

Source: AsianBondsOnline
Filter: yearly moving average of quarterly data
Source: Banco de México
Filter: yearly moving average of quarterly data
Section 2.6 argued that a rationale for using government-issued (as opposed to central-bank
issued) securities to sterilize international reserve accumulation was to avoid predation
between similar securities (in this case, floating-rate bonds) from two institutions with
comparable risk profiles (in this case, the federal government and Banco de México). In fact,
there is evidence that predation could have been taking place. Figure 10 displays two
liquidity measures (turnover on the left panel and traded volume on the right) for floating-rate
securities issued by the federal government (Bondes, including Bondes-D) and by Banco de
México (BREMs). IPAB securities are also included as a proxy for secular liquidity trends for
public floating-rate securities. The fact that Bondes and BREMs exhibit a negative correlation
for both liquidity measures could be indicative of how substitutable they were. In other words,
to a certain extent, BREMs could have crowded out liquidity from Bondes. Unsurprisingly,
turnover and the volume of BREMs traded sharply decreased once the central bank
substituted them with Bondes-D. Although the liquidity of Bondes had already been
increasing prior to the introduction of Bondes-D, the increase accelerated once the new
securities were issued (particularly, traded volume, on the right panel). Therefore, it seems
that substituting BREMs with Bondes-D has promoted the liquidity of the market for
government floating-rate securities.

246 BIS Papers No 67

Figure 10
Liquidity measures for floating-rate securities:
Federal Government, Banco de México, and IPAB: 2002-2011
Turnover ratio (% of total outstanding) Traded volume (in billions of pesos)
�
�
�
�
�
�
�
�
��
��
��
��
��
��
��
��
��
��
��
��
��
��
��
��
��
��
��
��
��������
���� �
����
���������������������
���� ��� �������������

���
���
���
���
���
���
���
��
��
��
��
��
��
��
��
��
��
��
��
��
��
��
��
��
��
��
��
��������
���� �
����
���������������������
���� ��� �������������

Source: Banco de México
Filter: yearly moving average

4.5 Higher diversification of the investor base for domestic public debt
As put forward by Pérez-Verdía and Jeanneau (2005) and by Tapia Rangel (2011), an
enhanced institutional framework in tandem with macroeconomic stability has also promoted
the diversification of the investor base. While the government has gained more confidence
from foreign investors, the growth of domestic institutional investors, such as pension and
mutual funds or insurers, has guaranteed a healthy demand for government securities. This
development could be represented by a Herfindahl index of government securities holdings
by investor category. Lower levels of this index would be related to a more diversified
investor base. Figure 11 suggests that the investor base has indeed become more
diversified; it suffices to compare the value of the Herfindahl index in 2002, around 0.5, with
the 0.2 it attained in December 2011.
A major advantage of the increased investor diversification could be lower debt market
sensitivity to shocks that are idiosyncratic to each investor profile. In addition, there is some
evidence that suggests that the profiles of both institutional and foreign investors are of a
more stable nature. Figure 12 shows rolling standard deviations of daily percentage changes
in government securities holdings for some investor categories. The purpose is to capture
how the volatility of holdings of each investor type has evolved over time. According to this
measure of volatility, it appears that the holdings of mutual funds and foreign investors have
become more stable over the last decade.
In what follows, we will discuss the role foreign investors have played lately in government
securities markets.

BIS Papers No 67 247

Figure 11
Herfindahl index of government securities holdings by investor category:
2002-2011
���
���
���
���
���
���
���
���
���
��
��
��
��
��
��
��
��
��
��
��
��
��
��
��
��
��
��
��
��
�����
�����
�����������������������
�������������������
�����������������

Source: Banco de México
Filter: 60-day moving average
The Herfindahl index ranges between 0 and 1. Higher levels of
the index are associated with lower investor base diversification.
Investor categories used to compute the index are: banking
sector, pension funds, mutual funds, insurers, other domestic
investors, foreign investors, repo operations with Banxico, and
collateral received by Banxico.
Figure 12
Rolling standard deviation of daily percentage changes in
government securities holdings of selected investor categories:
2002-2011
�
�
�
�
�
�
�
�
�
�
��
��
��
��
��
��
��
��
��
��
��
��
��
��
��
��
��
��
��
��
��
�������������
� �����������
����������

Source: Banco de México
Filter: 60-day moving average
Rolling standard deviations are computed over 60-day windows.

248 BIS Papers No 67

5. The stability of recent capital inflows
Many emerging economies have recently received significant capital inflows. In Mexico,
these inflows have resulted in unprecedented participation by foreign investors in the local
government securities market. Indeed, as Figure 13 shows, foreign investor holdings of
government securities have tripled since 2009, while their share of the market has more than
doubled over the same time period. Figure 13 also shows that the Mexican government
securities market stands out in emerging economies as one of those foreign investors favor.
Figure 13
Participation by foreign investors in selected emerging public securities markets:
2009-2011
Holdings in local currency
Sep 2009 = 100
Holdings as a share of the total outstanding
Sep 2009 = 100
�
��
���
���
���
���
���
���
��
��
��
��
�
��
��
��
�
��
��
��
��
�
��
��
��
�
�
��
��
��
�
��
���
��
�
��
��
��
��
�
��
��
��
�
��
��
��
��
�
��
��
��
�
�
��
��
��
�
��
���
��
�
��
��
��
��
�
��
��
��
�
������
� �����
� �� ��������
������
������
����

�
��
���
���
���
���
���
���
��
��
��
��
�
��
��
��
�
��
��
��
��
�
��
��
��
�
�
��
��
��
�
��
���
��
�
��
��
��
��
�
��
��
��
�
��
��
��
��
�
��
��
��
�
�
��
��
��
�
��
���
��
�
��
��
��
��
�
��
��
��
�
������
� �����
� �� ��������
������
������
����

Sources: Banco de México, the Reserve Bank of New Zealand, the Central Bank of the Republic of Turkey, the Turkish
Treasury Secretariat, the Peruvian Ministry of Economics and Finance, the Polish Ministry of Finance, the Brazilian
National Treasury.
In practice, it is not easy to measure the extent to which these flows are attracted by
temporary arbitrage conditions, or by fundamentals. However, as previously suggested in
Figure 12, a number of factors point to more stable foreign investment, presumably because
investors have not only been attracted to Mexican financial markets by carry-trade
opportunities but also by fundamentals. As Sidaoui, Ramos-Francia and Cuadra (2010)
explain, Mexican public finances are in good shape, inflation has steadily converged to the
central bank’s target, financial system resilience indicators pass international standards with
flying colors, and international reserves (together with liquidity arrangements with major
foreign institutions) convey confidence in the ability to finance external accounts should
external conditions significantly worsen. Finally, commitment to a floating exchange-rate
regime is well established.
Another factor that lends support to more stable higher participation by foreign investors in
the Mexican government securities market is the inclusion of Mexican bonds in Citigroup’s
WGBI since 2010. In addition to the fact that the market considered Mexico’s inclusion in the
WGBI to be yet another stamp of outside approval, the WGBI also raised awareness among
foreign investors of the availability of opportunities to invest in Mexican local markets. Finally,
the WGBI might have induced participation in government securities by those investors who
replicate this index.
Foreign investor participation in Mexican government securities has been fairly resilient to
global financial turmoil at least since 2009. Figure 14 depicts the exposure of foreign

BIS Papers No 67 249

investors to several Mexican public debt instruments and uses the CDS of European banks
as a proxy for adverse external conditions. The left panel shows information on foreigners’
short-term (Cetes) positions, with their positions in interest-rate derivatives netted out.
Although foreign investors have sharply decreased their short-term exposure since August
2011, their short-term positions have remained quite steady in spite of a worsening external
outlook. The right panel of Figure 14 offers a more supportive argument to the resilience of
foreign investor participation: regardless of a riskier environment in Europe, their holdings of
long-term government bonds have continued to increase, although at a much slower rate.
Figure 14
Participation by foreigners in Mexican government securities
and CDS of European banks (in basis points):
2009-2011
Short-term exposure
in billions of USD
Holdings of nominal peso bonds
in billions of pesos
�
���
���
���
���
���
���
���
���
���
���
��
�
�
��
��
��
��
��
��
��
��
�
��
��
��
�
��
��
��
��
�
��
��
��
�
�
��
��
��
�
��
���
��
�
��
��
��
��
�
��
��
��
�
��
��
��
��
�
��
��
��
�
�
��
��
��
�
��
���
��
�
��
��
��
��
�
��
��
��
�
�������
���� ����
�����
���������� ������������
�������������������������

�
���
���
���
���
���
���
�
���
���
���
���
���
���
���
���
��
��
��
��
�
��
��
��
�
��
��
��
��
�
��
��
��
�
�
��
��
��
�
��
���
��
�
��
��
��
��
�
��
��
��
�
��
��
��
��
�
��
��
��
�
�
��
��
��
�
��
���
��
�
��
��
��
��
�
��
��
��
�
�����
�������������������������

Sources: Banco de México and Bloomberg (European banks’ 5-year CDS).
It is still a matter of debate whether or not capital inflows have added stability to local debt
markets. Peiris (2010)15 offers empirical evidence for a number of emerging economies,
including Mexico, suggesting that the effect on interest-rate volatility of higher foreign
investor participation in domestic government securities markets tends to be either negative
or negligible.16 This result is unsurprising for Mexico due to the aforementioned evidence for
foreign investors being attracted by fundamentals. Figure 15 depicts the negative correlation
that has recently been observed between changes in foreign investor participation in
Mexican government securities and interest-rate volatility (as measured by the historical
volatility of 10-year bond yields).17

15 The countries considered in Peiris (2010) are Brazil, the Czech Republic, Hungary, Indonesia, South Korea,
Malaysia, Mexico, Poland, Thailand, and Turkey. The analysis spans from 2000 to 2009.
16 South Korea is the only country in Peiris (2010) that seems to have experienced higher interest-rate volatility
along with higher foreign investor participation.
17 Causality cannot be inferred from Figure 15 alone: in principle, it is not possible to disentangle whether
increased foreign investor participation lowers interest-rate volatility or whether causation occurs in the
reverse direction. The picture is offered as an illustration of the negative correlation between the two variables,
for which a more causal empirical analysis is in Peiris (2010), where the panel dimension is exploited to
address endogeneity.

250 BIS Papers No 67

Figure 15
The volatility of 10-year bond yields and changes
in foreign investor participation (%):
2009-2011
���
���
���
���
���
���
���
���
���
���
���
�� �� � � � �
�
��
��
��
��
��
��
��
��
��
���
��
��
���
��
��
���
��
��
��
��
��
��
��
��
� ��������������������������������������������������������������������������
�����������������������������
Source: Valmer and Banco de México
Volatility was computed with 10-year constant-maturity bond yields.
Increased foreign investor appetite for government securities, ceteris paribus, would bring
about lower interest rates. Pradhan et al. (2011)18 use a panel of emerging economies to
analyze the effect of foreign participation on interest-rate levels. They conclude that
emerging economies, including Mexico, have recently benefited from better funding
conditions arising from stronger external demand for public debt.
Evidence largely suggests that recent capital flows could be of a more stable nature and
have had a positive effect on the government securities market by diversifying the investor
base and by generating cheaper borrowing opportunities. However, in the current uncertain
environment, new episodes of turbulence in international markets could lead to higher risk
aversion. Under such circumstances foreign investors usually do not discriminate among
countries with better fundamentals. Mexico could be vulnerable in such a case. But,
eventually, sound macroeconomic fundamentals will attract abundant and more stable capital
inflows.
6. Concluding remarks
The development of domestic debt markets has contributed to better government financing
terms and has helped the central bank to carry out its open market operations more
effectively. It has also provided economic agents with a wide range of products for saving
and obtaining financing as well as hedging risks. Several factors have contributed to the
development of the Mexican local-currency securities market and span various fronts. Sound

18 The countries considered in Pradhan et al. (2011) are Brazil, Indonesia, South Korea, Malaysia, Mexico,
Poland, Thailand, and Turkey. The analysis spans from 2000 to 2010.

BIS Papers No 67 251

macroeconomic policy, minimal market intervention, pension system reforms (which led to
the expansion of large institutional investors), improved market transparency, and safer
clearing and settlement of securities have all created a more robust institutional framework.
These advances have fostered the development of secondary and derivatives markets,
which in turn feeds back to the debt market.
The government has pursued an active debt management strategy in order to reduce its
refinancing and interest-rate and currency-exposure risks. As a consequence of a more
stable macroeconomic outlook, it has been able to lengthen its yield curve and to increase
the share of fixed-rate, peso-denominated securities in the total. This has had positive spill-
over effects on the private securities market by providing longer-term reference rates, thus
allowing for an increase in the maturity of private debt instruments.
Banco de México has played an important role in the development of the government
securities market. The federal government has prepaid its outstanding foreign liabilities in
different ways. The currency needed for these operations was taken from Banco de México’s
international reserves, with virtually no effect on the exchange rate. The central bank has
also contributed to the improvement of the liquidity of the government securities market by
using these securities in its monetary operations.
Sound monetary and fiscal policies have led to international recognition of Mexican debt.
Government securities have attracted considerable interest from foreign investors. Mexican
bonds have been included in Citigroup’s WGBI, fueling greater investor awareness of
Mexican debt markets.
Overall, the Mexican government has succeeded in developing its local-currency securities
market. So far, in the current context of high external volatility, this strategy has paid off fairly
well. The investor base for government debt is more diversified, and increased foreign
investor participation seems to have favored more advantageous borrowing opportunities for
the government.
Still, some challenges remain for the attainment of higher levels of liquidity and greater
government securities market depth. Although securities lending has increased, most of the
activity is done by market makers through the central bank’s lending facility, in spite of efforts
to encourage lending among private parties (e.g., an increase in the cost of using the central
bank facility). Another positive development would be an increase in the maturity of repo
operations (currently, the bulk is overnight), which would ultimately lower market participants’
refinancing risk.
7. References
ACOSTA ARELLANO, R. and ÁLVAREZ TOCA, C., 2011, “Tipos de instrumentos y su colocación”,
El mercado de valores gubernamentales en México, Banco de México: Mexico, mimeo.
BANCO DE MÉXICO, 2000, Informe Anual 2000, Banco de México: Mexico
______, 2005, Informe Anual 2005, Banco de México: Mexico
______, 2006a, Informe Anual 2006, Banco de México: Mexico
______, 2006b, Reporte sobre el sistema financiero 2006, Banco de México: Mexico
______, 2007, Informe Anual 2007, Banco de México: Mexico
______, 2008, Informe Anual 2008, Banco de México: Mexico
______, 2009, Informe Anual 2009, Banco de México: Mexico
______, 2010, Informe Anual 2010, Banco de México: Mexico

252 BIS Papers No 67

GARCÍA PADILLA, J., 2011, “Mercado secundario”, El mercado de valores gubernamentales
en México, Banco de México: Mexico, mimeo.
JIMÉNEZ VÁZQUEZ, L., 2011, “Liquidación de valores”, El mercado de valores
gubernamentales en México, Banco de México: Mexico, mimeo.
PRADHAN, M., BALAKRISHNAN, R., BAQIR, R., HEENAN, G., NOWAK, S., ONER, C., and PANTH,
S., 2011, “Policy Responses to Capital Flows in Emerging Markets”, IMF Staff Discussion
Note SDN/11/10, International Monetary Fund: Washington, D. C.
PEIRIS, S. J., 2010, “Foreign Participation in Emerging Markets’ Local Currency Bond
Markets”, IMF WP/10/88, International Monetary Fund: Washington, D. C.
PÉREZ-VERDÍA, C. and JEANNEAU, S., 2005, “Reducing financial vulnerability: the
development of domestic government bond market in Mexico”, BIS Quarterly Review,
December 2005, pp. 95 – 107.
SIDAOUI, J., 2002, “The role of the central bank in developing debt markets in Mexico”, BIS
Papers No. 51, Bank for International Settlements: Basel, pp. 151 – 164.
SIDAOUI, J., RAMOS-FRANCIA, M., and CUADRA, G., 2010, “The global financial crisis and
policy response in Mexico”, BIS Papers No. 54, Bank for International Settlements: Basel,
pp. 279 – 298.
TAPIA RANGEL, C., 2011, “Base de inversionistas”, El mercado de valores gubernamentales
en México, Banco de México: Mexico, mimeo.

BIS Papers No 67 253

Fiscal policy considerations in the design of
monetary policy in Peru
Renzo Rossini, Zenón Quispe and Jorge Loyola1
Abstract
We evaluate the financial and real linkages between fiscal and monetary policy in Peru, and
show that during the recent export commodity price boom, public finances supported the
implementation of monetary policy. In particular, the reduction of the net public debt has
translated into a greater capability by the Central Bank to sterilize its FOREX interventions.
Also, an active policy to enhance the development of the local capital markets, using the
issuance of public bonds denominated in local currency as a benchmark, has created the
incentive to de-dollarize banking credit. On the other hand, difficulty in fine-tuning public
investment around the business cycle in recent years has led to periods of a fiscal stance
that does not counteract the real business cycle. This raises the question of the possibility of
adopting a structural rule for the public sector balance, based on structural fundamentals.

Keywords: Central bank monetary policy, fiscal policy, macroeconomic stabilization
JEL classification: E52, E58, E63

1 Central Reserve Bank of Peru.

254 BIS Papers No 67

Introduction
According to the Peruvian Constitution, the Central Reserve Bank of Peru (BCRP) is an
independent public institution that has the objective of preserving monetary stability through
the regulation of money and bank credit. On this basis, monetary policy in Peru follows a
modified form of inflation targeting, in which the policy interest rate is used to counteract
deviations of inflation with respect to the target of 2% (price stability), but also includes a set
of additional, unconventional, instruments aimed at avoiding an overreaction of bank lending.
Fiscal policy aims to create equal opportunities among citizens, ensure sustained growth and
defend the public credit (ensuring fiscal solvency to avoid a financial crowding out). The last
two objectives are related to the stabilization policies and the public sector asset and liability
management, and therefore overlap with monetary policy. For this reason, the design and
implementation of monetary policy take into account the impact on aggregate demand of a
positive or negative fiscal impulse, and in this way the BCRP seeks not to overreact to or
accommodate a fiscal shock. Additionally, in the short run, the monetary operations take into
account the Treasury cash flow and other financial operations.
The surge of commodity prices in the last 10 years has led to a significant increase in
earnings in Peru, including tax revenue. This environment has been favorable to the surge in
foreign direct investment and other forms of capital inflows, which have become a source of
risk of macroeconomic overheating. To avoid a pro-cyclical stance, the public sector has
achieved an annual financial surplus several times since 2006, and has reduced the size of
the public debt, increased the size of the Fiscal Stabilization Fund, and augmented the
amount of other public sector deposits at the BCRP. Given this favorable financial position,
the fiscal authorities have adopted an expansionary stance to protect the economy from the
adverse shock created by the international financial crisis in 2009.
In this article, we assess the policy coordination between the BCRP and the Ministry of
Finance at two levels: financial and macroeconomic stability. In the first part we show that
Peru’s fiscal policy has been able to save part of the non-structural revenues, complementing
the monetary policy. In the second part, we evaluate Peru’s fiscal stance in terms of its
supporting role for macroeconomic stability.
Financial Policy Coordination
During the last 10 years, average annual GDP growth in Peru has been 6.3%, reflecting,
among other things, the increase of export commodity prices (see Figure 1 and 2), which on
average grew 14% per year. The impact on revenues from the export boom has been
significant, with tax revenues growing from 14% of GDP in 2001 to 18% in 2011. This has
become more noticeable since 2005, given the increase in mineral prices in international
markets. The average price of Peruvian exports was 126% higher in 2005-2011 than in
2001-2004. As a result, revenues from mining exports as a percentage of total fiscal
revenues rose from 5.9% in 2005 to nearly 10% in 2011.

BIS Papers No 67 255

Figure 1
�
�������
������
����
����
������
������
������
������ ������ ��� ���� ���� ���� ����
� ����������������������������������������������������
� �����������������������������
������� ���������
�������������������
�� ���������
����
�����
����
�����
����
����
����
����
����
�����
����
����
����
�
�
Figure 2
0
50
100
150
200
250
300
350
400
12
14
16
18
20
2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011
CENTRAL GOVERNMENT REVENUES AND EXPORT PRICE
INDEX: 2001-2011
Other incom e (left) Mining revenues (left) Export Price Index (right)
% GDP Index (1994=100)
�
The size of the expenditures in the public sector budget has not increased at the same pace
as the tax revenue, resulting in a considerable reduction in the net debt of the public sector,
from 38% of GDP in 2001 to 8% in 2011. This figure includes public sector liquid assets of
about 14% of GDP. Regarding the latter, two major issues in the management of public debt
in the last decade have to do with the increase in the share of domestic debt relative to the
total public debt from 23% in 2001 to 43% in 2011; and a significant extension of the average
debt maturity from 7 years in 2002 to 13 years in 2011 (see Table 1 and Figure 3). These
fiscal results show that part of the non-structural revenues was saved.

256 BIS Papers No 67

Table 1

Figure 3
�
��
��
��
��
��
���� ���� ���� ���� ���� ���� ���� ���� ���� ���� ��������
�����������
������ �
��� ����������
�������������
��������
������������
��������������������
�
The main components on the asset side of the public sector balance sheet are BCRP
obligations (9.2% of GDP), including Fiscal Stabilization Fund deposits (3.3% of GDP), other
Treasury deposits (3.5% of GDP), and sub-national government deposits (2.1% of GDP).
The Fiscal Stabilization Fund was created by the Fiscal Responsibility Act as a buffer during
recessions (Table 2).
���� ���� ����
���������������������������
�������������� ���� ���� ����
������������ ���� ���� ���
������� ���������� ���� ��� ����
����������������� ���� ���� ����
��������� ������������������
������� ���������� ��� ���� ����
����������������� ��� ��� ����
��������� ��� ��� ����
��������������������
�������� ���������� ��� ���� ���� ����
����������������������������
�������������������� ���� ���� ����
� ���������������������������������
��������������������
�������������������� � ���� ����
������������������

BIS Papers No 67 257

Table 2
�
Since 2002, the fiscal authorities’ liability management has contributed to creating a
benchmark for issuances of long-term obligations in domestic currency. The actual size of
public bonds in domestic currency placed in the local market is equivalent to 6.1% of GDP,
with an average maturity of 15.7 years and a yield of 5.85%, representing 29% of the total
public debt. The short end of the yield curve is made up of BCRP Certificate issuances with
maturities of up to one year. As shown in Figure 4, the short- and long-run segments of the
benchmark yield curve are well connected. BCRP Certificates were created in 1991 to
sterilize FOREX intervention. The alternative of issuing public debt to sterilize the liquidity
created by FOREX interventions has so far been discarded because of the lack of flexibility
of debt operations compared with monetary operations.
Figure 4

Assets
Central Reserve Bank 9.2 Bonds (foreign currency)* 5.5
-Treasury 3.5
-Sub-national governments and others 2.1 Other external debt 6.3
-Consolidated Pension Reserve Fund 0.3
-Fiscal Stabilization Fund 3.3 Bonds (domestic currency)** 6.5
Banco de la Nación 2.1 Lima Municipal Bonds 0.0
-Treasury 0.2
-Sub-national governments and others 1.9 Pension Recognition Bonds 1.6
Commercial Banks 2.3 Credits from Banco de la Nación 0.3
Rest of Financial System 0.2 Short-term 0.7
Total 13.8 Total 20.9
Net Debt (Liabilities – Assets) 7.1
*Include global bonds (US$ 9 312 million) and bonds of financial system (US$ 122 million of domestic debt)
** Include sovereign bonds (US$ 14 048 million and debt exchange bonds (US$ 682 million of domestic debt)
Public Sector Balance Sheet
(As percentages of GDP, September 2011 figures)
Liabilities
4.17 4.17
4.20
4.24
4.15
4.20
4.25
4.30
Y
ie
ld
s
(%
)
Term to maturity (in months)
Peru: Central Bank CDs yield curve1
(December 2011, in percentages)
������������������������������������������������������������������������
4.5
4.8
5.7
6.2
6.5
6.6
4.0
4.5
5.0
5.5
6.0
6.5
7.0
0 5 10 15 20 25 30 35 40
Y
ie
ld
s
(%
)
Term to maturity (years)
Peru: Sovereign yield curve1
(December 2011, in percentages)
������������������������������������� ������������������������������������ ������������������������������������
1 day

258 BIS Papers No 67

Public sector deposits at the BCRP are the main source of sterilization of FOREX operations
(Table 3). The BCRP accumulates international reserves as a preventive measure,
considering the risks associated with a partially dollarized financial system. FOREX
intervention has reduced exchange rate volatility, thus avoiding deterioration in the quality of
banks’ loan portfolios. The size of net international reserves increased from 18.7% of GDP in
2006 to 27.8 % of GDP in 2011. Reserve requirements (with higher rates on banks’ short-
term foreign exchange liabilities) provide another source of international reserves.
Table 3
�
Inflation targeting in Peru, as described in Rossini et al. (2011), gives special consideration to
financial stability, given the weakness associated with financial dollarization (see Figure 5).
Therefore, additional instruments, like reserve requirements, are used to avoid significant
swings in bank credit like those than can emerge from sharp exchange rate fluctuations,
bank runs on dollar deposits, and capital inflows. In this regard, close attention is paid to
deviations of bank credit as a percentage of GDP with respect to the trend (Figure 6). In
terms of policies, there are preventive measures to ensure an adequate level of bank
liquidity, using reserve requirements kept at the Central Bank in the form of international
reserves.
Figure 5

��
��
��
��
��
��
��
��
��
��
������ ������ ������ ������ ������ ������ ������ ������ ������� ������ ������� ������ ������� ������
��������������������������������������������������������

�� �����������������
Assets Liabilities
International reserves 27.8 Public sector deposits 10.9
In domestic currency 7.1
In foreign currency 3.8
Reserve requirements 7.8
In domestic currency 2.2
In foreign currency 5.7
Central Bank instruments 3.3
Cash holdings 5.2
Other liabilities 0.5
Peru: Central Reserve Bank Balance Sheet
(As percentages of GDP, November 30, 2011 figures)

BIS Papers No 67 259

One consideration regarding sterilized interventions is the net cost of the additional monetary
liabilities, compared with the returns on international reserves. Table 4 shows that in 2011
the returns from international asset management (1.52%) were above the average cost of
the BCRP’s liabilities (1.45%). This result is influenced by the zero cost of the currency and
the interest paid on public sector deposits. It is worth mentioning that the valuation effect of
exchange rate fluctuations on the BCRP’s net foreign currency assets is not part of its profit
and loss statement, since it is registered in a separate line in the capital account. Despite the
lack of general accounting rules for central banks, the rationale for this method to register
valuation changes is that a depreciation of the currency should not generate profits, nor
should an appreciation create accounting losses.
Figure 6
�
�
Figure 7
�
����
��
��
��
��
��
��
��
��
���
�
��
� ��
� ��
���
�
��
���
� ��
� ��
�
��
���
�
��
���
� ��
� ��
� ��
���
� ��
���
� ��
� ��
� ��
���
�
��
���
�
��
� ��
�
��
���
�
��
���
�
��
���������������� ����������� ��������������
����������������
� �������
����
��
��
��
��
��
��
��
��
���
�
��
� ��
�
��
���
�
��
���
�
��
� ��
�
��
���
� ��
���
� ��
� ��
� ��
���
� ��
���
�
��
� ��
� ��
���
�
��
���
�
��
� ��
�
��
���
�
��
���
�
��
���������������� ������������������������
����������������
� �������
�������
����
����
����
����
����
����
����
����
����
����
���
� ��
���
� ��
���
� ��
���
�
��
���
� ��
���
�
��
���
� ��
���
� ��
���
� ��
���
� ��
���
� ��
���
�
��
���
�
��
���
�
��
���
�
��
���
�
��
���
�
��
���
�
��
���
�
��
���
�
��
���
�
��
���
�
��
���
�
��
���
�
��
���
�
��
���
� ��
���
� ��
���
� ��
���
� ��
���
� ��
���
� ��
���
�
��
���
�
��
���
�
��
���
�
��
���
�
��
���
�
��
�����������������������������������������
������������������������
��������� �������� �����������������
�����������������������������
��� ������ ���� ���� ������������������
�������������������������������������������������������������������������������������

260 BIS Papers No 67

Table 4
�
Policy coordination at the operational level is also a critical dimension of monetary and fiscal
policy coordination. At the macro level, monetary programming frameworks can be
instrumental in preventing inconsistencies in the policy mix, whereas the coordination of
operations is of critical importance for the day-to-day implementation of monetary and fiscal
policies at the microeconomic level. As shown in Figure 8, the considerations of the daily
Treasury cash management shape the liquidity management of the BCRP to ensure
adequate liquidity for the desired closing liquidity demand of private banks. This is more
evident during the scheduled tax collection period, when private banks transfer liquidity to the
Treasury and the BCRP responds with open market operations to preserve the liquidity of the
system.
Figure 8
�����
�����
�����
����
�
���
����
����
� � � � � � � �� �� �� �� �� �� �� �� �� �� �� �� �� �� �� ��
��������������� ������ ���
��������� ��������
� ������������������� �������������������� ��
��������
� ����������������������������� �����
��������������������������
�����
�����
�����
�
����
����
����
����
�����
�����
���������������������������������������������� ��������������������
��������� ��������
� ���� ��������������������������������������
�����
� ����������������������������������������� ���������
�������������������������
���� ������ �������������
��������������������� ���
�����������������

Assets Liabilities
International reserves 1.52 Funding costs 1.45
Deposits of the public sector 2.35
In domestic currency 3.51
in foreign currency 0.11
Reserve requirements 0.27
In domestic currency 0.64
in foreign currency 0.10
Central bank instruments 3.92
CDs 3.88
Term deposits 4.00
Cash holdings 0.00
Other liabilities 0.26
Peru: Average yields and funding costs of the Central Bank balance sheet
(In percentages, November 30, 2011 figures)
������ ��������
������ ��������

BIS Papers No 67 261

Macroeconomic Policy Coordination
As part of the institutional framework governing the relationship between fiscal and monetary
policy in Peru, the BCRP is explicitly forbidden to finance the public sector with loans or
purchases of government securities. Also, the government is obligated to request BCRP
advice and publish it together with official fiscal forecasts. Finally, the BCRP is required to
inform the Ministry of Finance if a given policy affects the BCRP’s ability to fulfill its mandate.
The Fiscal Responsibility and Transparency Law (1999) includes a combination of a target
for the nominal fiscal deficit and a ceiling for the expansion for non-financial public sector
expenditure (Table 5). These targets are not adjusted for non-structural effects, and do not
necessarily ensure a countercyclical stance. Table 6 outlines the main goals and structure of
the Treasury Cash Management Committee (CMC).
Table 5
Macro-fiscal Rules

Deficit rule The annual deficit of the non-financial public sector (NFPS) cannot exceed 1% of GDP.
Expenditure rule
The annual increase in consumer spending of the central government
shall not exceed 4% in real terms. Consumer spending includes
spending on salaries, pensions, goods, and services.
Debt rule
NFPS debt must not be increased by more than the amount of deficit
corrected for the difference attributable to changes in currency parities,
issues of recognition bonds, changes in deposits and debt taken by the
NFPS.

Fiscal Stabilization Fund (FEF)

Funding sources
The FEF’s regular funding sources are the fiscal surpluses of the
Treasury obtained at the end of each year. The accumulated savings
cannot exceed 4% of GDP. Any additional earnings are used to reduce
debt. The FEF balance at December 31, 2011 was U.S. $ 5.6 billion.
Uses
The FEF can be used if the current revenue, in terms of GDP, falls
more than 0.3 percentage points below its average level of the last
3 years. In this event, the amount that exceeds the declining limit of
0.3% of GDP, and up to 40% of the current balance of the FEF, will be
used to cover poverty alleviation programs, as a priority.
Exceptions to the rules
Exceptional events
In the event of national emergency or international crisis, the Congress
may suspend up to a maximum of three years the implementation of
any of the fiscal rules outlined above. For instance, during 2009-2010
this exception to the rule was activated in order to meet the costs of the
international financial crisis.
�
�
�

262 BIS Papers No 67

Table 6
Cash Management Committee (CMC)

Functions The CMC evaluates and approves, on a monthly basis, the expenditure control and financing operations of the Treasury.
Treasury cash flow
(definition)
Flow of funds associated with the cash revenues and expenditure profiles
in domestic and foreign currencies from the national, regional, and local
government entities. The General Director of Treasury and Public Debt of
the Ministry of Economy is in charge of the management of the fund.

The cash flows are centralized and managed through the Treasury’s main
account at the Central Bank.
Members
The CMC is composed of 5 members, including the General Manager of
the Central Bank, the Vice Minister of Economy, the General Manager of
the Banco de la Nación, the General Director of Treasury and Public Debt,
and the General Director of the Public Budget. The importance of the
Treasury’s cash management for the monetary policy design justifies the
inclusion of a representative of the Central Bank on the CMC.

The charts in Figure 9 show the limits established by the Law of Fiscal Responsibility and
Transparency with respect to the deficit and expenditures for each year and the execution of
these variables. Since 2003, the deficit has been consistent with the rules, although in 2009
and 2010 it was necessary to establish waivers approved by the Congress as a result of the
global financial crisis. In 2005 and 2006, public spending growth was above the targets set
by the rule, as exceptions approved by the Congress. These deviations have given rise to
discussions about the need to replace the growth targets with limits to the structural balance.
BCRP independence includes the ability to establish its policy goals and decide which
instruments to use, and, as consequence, policy coordination with the Ministry of Finance is
mainly based on the consideration of the actions of the other body. This process includes the
publication of the Ministry of Finance and BCRP macroeconomic forecasts in the semi-
annually Multiannual Macroeconomic Framework Memoranda (MMM) and the quarterly
Inflation Report, respectively. Table 7 shows the sequence for the publication of these
forecasts during the year. As a result of this coordination in the forecasting process, the
Ministry of Finance and BCRP forecasts and final data tend to be quite similar.
Figure 9

���
����
���
���
���
���
���
���
���
����
���
���
����
����
����
���
���
���
���
���
����
����
���� ���� ���� ���� ���� ���� ���� ���� ���� ���� ���� ����
�����������������������������������
������� ��������
��������� ����

���
���
���
���
���
���
����
� ���
� ���
���
���
����
� ���
� ���
� ���
� ���
���
���
���
���
���
���� ���� ���� ���� ���� ���� ���� ���� ���� ���� ���� ����
� ����� ����������������������������������
�� �����
��������� ����
�������
�������

BIS Papers No 67 263

The Ministry of Finance’s statistics and figures on the budget policy goals are taken into
account by the BCRP to generate its own projections, but there could be forecasting
differences due to different assumptions in crucial variables like terms of trade and nominal
and real GDP growth. However, government expenditure figures tend to be similar, as they
reflect the annual budget’s policy goals.
Table 7
��� ��� � �� ��� � �� ��� ��� ��� ��� � �� � �� ���
� ��������
������
����������
�����������
���
����������
����������������
���
����������
����������������
���
����������
����������������
���
�������
������
� � � �����
����������
���
���������
� � � �����
�������������
��� ���������� ���������� ��������������

To assess the impact of the fiscal stance on aggregate demand, the BCRP calculates a
modified form of the indicator for the structural balance to take into account the effect of
export prices on tax revenues: the actual balance of the consolidated public sector is
adjusted not only for the effect of the output gap on tax revenues, but also for the effect of
the deviation of the average export price (relative to a long-run trend) on the income tax paid
by the mining sector.
Equation 1 shows how the structural balance (SBt) is calculated adjusting the public sector
balance (PBt) for the effect of the output gap (CEt) and export prices (PEt) on taxation. In
order to assess if the real intention of the fiscal policy is to contract or expand the economy,
the indicator used is the change of the structural balance, or the fiscal impulse (FIt). When
positive, FIt shows a policy expansion, and when negative a contraction.
tttt PECEPBSB �� (1)
� �1��� ttt SBSBFI (2)
Table 8 shows the evolution of the public sector actual and structural balances. It can be
verified that the unadjusted fiscal balance does not necessarily reflect the real fiscal stance.
For example, in years of apparent fiscal contraction like 2005, 2007, and 2010, the structural
balance showed a fiscal expansion.

264 BIS Papers No 67

Table 8
���������������� ���� ��� � � ���� �� ����������������� �
�� ������������������
���� ���� ���� ���� ���� ���� ���� ���� ���� ���� ����
�������������������������� ���� ���� ���� ���� ���� ��� ��� ��� ���� ���� ���
������������ ������ ��� ��� ��� ��� ���� ���� ���� ���� ���� ���� ����
������� �������������������� ��� ��� ��� ��� ��� ��� ���� ���� ��� ��� ����
������� ������������ ��������������� ��� ��� ��� ���� ���� ���� ���� ���� ���� ���� ����
� ������������ ���� ��� ��� ��� ��� ��� ��� ���� ���� ��� ���� ���
����������������������� ���� ���� ���� ���� ���� ��� ��� ���� ���� ���� ���
��������� ���������� ���� ���� ���� ���� ��� ���� ��� ��� ��� ��� ����
� �� ��
� ��������� ���� ���� ���� ���� ���� ��� ��� ��� ���� ��� ���
������������������ ����� ����� ����� ���� ���� ���� ���� ���� ���� ���� ����
������������������ ����� �
�
Figure 10
�
The BCRP’s Inflation Report presents estimates of the structural fiscal balance to show the
differences with the public sector balance (Figure 10). The BCRP identifies the size of the
temporary factors affecting public sector earnings and measures the fiscal impulse to
establish if fiscal policy is accommodative or tight.
The ability of fiscal policy to contribute to stabilizing macroeconomic activity can be assessed
estimating the multipliers of different components of the fiscal result. Using a structural vector
auto-regression (SVAR) model, with variables such as current income and expenditure,
capital expenditures, real GDP, terms of trade, and the balance of the monetary base, it can
be established that only capital expenditures have a multiplier2 generating an impact on GDP
greater than 1, and statistically different from zero (Table 9). This implies an important

2 We would like to acknowledge the assistance received from Mr. Guillermo Ferreyros, who carried out the
calculation of fiscal spending multipliers, Mr. Enrique Serrano, for computing the output gap response to the
fiscal impulse within the quarterly forecasting model of the Central Bank, and Mr. Luis Rizo Patron, for the
satellite estimations of the relationship between the fiscal impulse and the output gap.
����
����
����
����
����
���
���
���
����
����
���
���� ����
���� ���� ����
���
���
����
����
����
���
���� ���� ���� ���� ���� ���� ���� ���� ���� ���� ����
��������������������������������������
��������������������������������������
�����������������������
�������������� ������������������
����
����
����
����
���
����
���
���
���
���
����
���� ���� ���� ���� ���� ���� ���� ���� ���� ���� ����
��������� ����������������
�����������������������

BIS Papers No 67 265

difficulty, since capital expenditures take time to design and implement; in consequence,
using them to stabilize the economic cycle could have an untimely effect, producing an
involuntary pro-cyclical fiscal stance.
Table 9
Fiscal Multipliers
(in PEN Soles)
Effect on GDP of an increase of PEN S/.1.00 in:
� ��������
���������
���������
�������������
��������
�������������
�������������� ����� ���� �����
�������������������� ����� ���� �����
��������������������� ����� ���� �����
The difficulty of synchronizing fiscal policy with the need to reduce the output gap has been
given rise to periods of pro-cyclical fiscal policy. These events can be explained by the
tendency to maintain an expansionary (or a contractionary) fiscal stance once the recession
(or boom) is over. This has an important consequence: monetary policy has a greater
responsibility for fine-tuning aggregate demand during the business cycle. For example,
despite the recovery in 2010 from the external shock of 2009, the fiscal policy was
maintained in an expansionary mode. The forecasting process of the BCRP includes the
evaluation of a fiscal shock to aggregate demand, whose impact is estimated with a policy
parameter of 0.24 and a maximum policy lag of 6 quarters (Figure 11).
Figure 11
�
Figure 12 shows the sequence of fiscal policy decisions in Peru around the Lehman Brothers
bankruptcy, which demonstrates the lagged impact of those policy changes.
For example, between 2007 and 2008, and from 2009 to 2010, despite the improvement of
the output gap, the fiscal stance was maintained in an expansionary mode. Conversely, in
2008 and 2009 the fiscal impulse was of a similar magnitude in both years despite the
significant contraction in economic activity.
�������� � ������������������������������������� �����
��
�

���
���
����
�� �� ��� ��� ���

266 BIS Papers No 67

Figure 12

In order to contrast the fiscal policy decisions with some theoretical structural fiscal rule,
which could be identified with the fiscal impulse ( tFI ) as the policy reaction to the previous
level of the fiscal impulse ( 1tFI � ) and to yearly deviations of the output gap( tOG ), we can
check if the fiscal policy actions were systematically countercyclical according to the
following rule in equation 3:
0;1 �� � JJG ttt OGFIFI (3)�
The estimation of the relationship between the fiscal impulse and the output gap in equation
4 confirms that the difficulty in fine-tuning public investment around the business cycle in
recent years has led to the adoption of a fiscal stance that does not counteract the real
business cycle. In particular, the parameter of the output gap presents an opposite sign than
the expected, and without significant explanatory power. This result holds when we estimate
the equation looking for an optimal lag structure. This raises the question of the possibility of
adopting a structural rule for the public sector balance, based on structural fundamentals
such as the one presented in equation 3.
� � � �81548.054758.2
069.0365.0 1 ttt OGFIFI � � (4)
Conclusions
We have evaluated the financial and real links between fiscal and monetary policy in Peru,
and have shown that during the recent export commodity price boom, the public finances
supported the implementation of monetary policy. In this regard, the reduction of the net
public debt has been translated into greater Central Bank capability to sterilize its FOREX
interventions. Also, an active policy to enhance the development of the local capital market,
using the issuance of public bonds denominated in local currency as a benchmark, has
created the incentive to de-dollarize bank lending.

� ���
� ���
� ���
� ���
� ���
� ���
���
���
���
���
���
���
� ��� � ��� � ��� � ��� ���� ��� ��� �� � �� �
����������������������������������������
���
(C ontractionary)
��
(Expansionary)
���
(C ontractionary)
� ���������
��������� ������
���� ����
����
����
����
���� ����
����
����
����
(Expansionary)

BIS Papers No 67 267

On the other hand, difficulty in fine-tuning public investment around the business cycle in
recent years has led to periods of a fiscal stance that does not counteract the real business
cycle. This raises the question of the possibility of adopting a structural rule for the public
sector balance, based on structural fundamentals.
References
Banco Central de Reserva del Perú (2011), Reporte de Inflación, December.
Blinder, Alan (1982), Issues in the Coordination of Monetary and Fiscal Policy, in Federal
Reserve of Kansas City, Monetary Policy Issues for the 1980s.
Canzonieri, Mattew, Robert Cumby and Behzard Diba (2002), Should the European Central
Bank and the Federal Reserve be concerned about Fiscal Policy?, in Federal Reserve Bank
of Kansas City, Rethinking Stabilization Policy.
Departamento de Modelos Macroeconómicos (2009), Modelo de Proyección Trimestral del
BCRP”. Documento de Trabajo N° 2009-006- BCRP.
Laurens, Bernard and Enrique De La Piedra (1998), “Coordination of monetary and fiscal
policies”. IMF Working Paper WP/98/25, Washington DC.
Ministerio de Economía y Finanzas (2011), Marco Macroeconómico Multianual 2012-2014.
Moreno, E. and R. Lema (2008), “Metodología del Cálculo del Resultado Estructural”. Nota
de Estudio No.51- BCRP.
Rossini, Renzo, Zenón Quispe and Donita Rodríguez (2011), Capital Flows, Monetary Policy
and FX Intervention in Peru. BIS Papers chapters, in: Bank for International Settlements
(ed.), Capital flows, commodity price movements and foreign exchange intervention, volume
57, pp 261-274 BIS.
Taylor, John (1992), “Discretion Versus Policy Rules in Practice” Carnegie-Rochester
Conference Series on Public Policy 39.

BIS Papers No 67 269

Fiscal policy, public debt management and
government bond markets: the case for the Philippines
Diwa C Guinigundo1
Abstract
The fiscal health of the Philippines has improved significantly over the past decade.
Notwithstanding the dividends from reforms, challenges remain for the Philippines on the
fiscal side. Policy coordination, primarily through the Development Budget Coordinating
Committee, has helped to reduce the need for policy sterilisation. However, some concerns
have been raised by the Bangko Sentral ng Pilipinas (BSP) about the reduced issuance of
government securities as well as possible interest rate repression. Meanwhile, sufficient
liquidity in the domestic economy has ensured that the crowding out of private offerings is not
an immediate concern. Further reforms on public debt management are needed to promote
efficiency, further develop the capital market and enhance overall financial stability.

Keywords: Fiscal policy, public debt management, Philippines
JEL classification: E630, H063

1 Deputy Governor, Monetary Stability Sector, Bangko Sentral ng Pilipinas

270 BIS Papers No 67

1. Introduction
The fiscal health of the Philippines has improved significantly over the past decade. By 2005,
there was widespread recognition that the fiscal position of the national government had
become untenable. Subsequently, fiscal prudence was observed and new taxes were
enacted. As a quick result, the fiscal position had almost returned to balance by 2007. This
led to a reduction in the total outstanding government debt from a high of 74.4% of GDP in
2004 to a more manageable 52.4% of GDP by 2010. Along with improved fiscal balance
numbers and relatively robust economic performance, the Philippines has earned credit
rating upgrades and expects to do more in the near future. In recognition of the country’s
sustainable fiscal position, debt spreads have narrowed to levels better than those of higher-
rated sovereign bond issuers.
Notwithstanding the dividends from reforms, challenges remain for the Philippines on the
fiscal side. Weak revenue generation, enactment of revenue-eroding measures by the
Philippine Congress and recent underspending have generated concerns for the fiscal
authorities. While revenue shortfalls have been manageable, they may contribute to rising
deficits in the future. Should the fiscal stance become unsustainable, public expenditure may
again be constrained with a corresponding negative effect on economic growth.
With the improvement in the scale of government debt, the debt service burden has also
become less of a fiscal drag. From 85% of total government revenue in 2004, the debt ratio
fell to 57% by 2010. As a proportion of GDP, the debt service burden likewise dropped to
7.7 % in 2010 from a high of 13.6% in 2006. While the fiscal situation is currently under
control, the prospect of either lost opportunities for improved economic performance or future
instability requires further thought.
2. Potential for constraints on monetary policy from an
unsustainable path for public debt
2.1 Measurement of the fiscal policy stance and public debt
As determined by the fiscal authorities, the fiscal policy stance is designed to deliver sound
public financing including a commitment to medium-term objectives2 combined with the
flexibility to respond to changing economic conditions in the short term. Its measurement
takes into consideration cyclical movements in the economy and contingent liabilities over
the medium term. By cyclically adjusting the fiscal policy stance, important fiscal variables
are scaled to GDP to provide some insight into cyclical patterns in the economy.
Notwithstanding these measures, impulse responses generated from a vector error
correction model (VECM)3 show that public spending has been cyclical and needs to adopt a
more countercyclical stance to support the economy against countercyclical spending shocks
(Figure 1).

2 The medium-term goals of the fiscal programme pertain to fiscal consolidation with a view to meeting a
targeted ratio of the fiscal deficit to GDP.
3 The VEC model examines the impact of policy measures adopted during the global financial crisis. Monetary
policy was proxied by M3 in real terms while fiscal policy was represented by government consumption. The
VEC model is taken from Redoblado (2011). Please see Annex A for details.

BIS Papers No 67 271

Figure 1
Response of national government expenditure to a GDP shock

Public debt measures used in formulating the fiscal stance cover the Local Government Units
(LGU), the 14 monitored Government Owned and/or Controlled Corporations (GOCCs), two
Government Financial Institutions (GFIs), and three Social Security Institutions. Through the
government’s fiscal risk management programme, contingent liabilities relating to pensions
and health care spending are included in the medium-term fiscal programme. These items
are periodically monitored but do not form part of the fiscal budget until assumed by the
government. The government’s gross debt is measured by netting out deposits placed with
the central bank. However, in the presentation of the consolidated public sector debt, the
intra-sector debt holdings are netted out. With respect to the measurement of public sector
assets when formulating the fiscal stance, these assets (eg central bank assets, state
pension funds) are not seen as an offset to gross debt. Otherwise, the ratio of debt to assets
would effectively be lower. The fiscal authorities, through the Development Budget
Coordinating Committee (DBCC), set fiscal targets such as the key tax and spending
priorities while avoiding an unsustainable rise in the burden of public debt.
2.2 Interaction between monetary and fiscal policy
Results from the same model suggest that the BSP and the national government have
coordinated their policy actions so that policy sterilisation has been avoided. Both impulse
response analysis and variance decomposition show that shocks to domestic liquidity allow
for higher spending by fiscal authorities. The reverse is also true as higher government
spending increases liquidity in the financial system. Consequently, policy measures
undertaken by monetary and fiscal authorities do not offset each other.4
Moreover, it was observed that GDP reacts more to monetary policy than fiscal policy. It also
seems that policy moves by monetary authorities take effect faster than fiscal policy action.5
Also, while GDP is more responsive to innovations in monetary policy than to fiscal policy
shocks, monetary policy tends to become countercyclical after roughly five quarters. This
may be construed as heading off the inflationary effects of output growth as the slack in the
economy is taken up by more economic activity.

4 This may of course be subject to some threshold which could be the basis for future research.
5 Variance decompositions indicate that monetary policy shocks account for the bulk of the response of GDP.
-.01
.00
.01
.02
.03
.04
.05
.06
1 2 3 4 5 6 7 8 9 10
Gross Domestic Product
Expenditures by the National Government
� � � � � ��
� � � �

272 BIS Papers No 67

While benign interaction between monetary and fiscal policy has been observed during the
global financial crisis and its aftermath, some concerns have been raised by the BSP on the
national government’s reduced issuance of government securities, as well as on the topic of
possible interest rate repression.
The recent underspending in 2011 has left the national government with sufficient funds for
its operations. Despite efforts to ramp up spending in the latter part of the year, the fiscal
deficit incurred by the national government has only amounted to 65.9% of the programmed
amount for the fiscal year (Table 1).

Table 1
National government fiscal performance
In billions of pesos

December January–December
2010 2011 Growth (%) 2010 2011
Growth
(%)
Q1–Q4
2011
Program
% to
Q1–Q4
2011
Program
Surplus/(Deficit) –44.6 –101.5 127.4 –314.5 –197.8 –37.1 –300.0 65.9
Revenues 103.2 110.2 6.8 1,207.9 1,359.9 12.6 1,411.3 96.4
Expenditures 147.8 211.7 43.2 1,522.4 1,557.7 2.3 1,711.3 91.0
Source: Bureau of the Treasury.

Combined with the high level of liquidity in the financial system, this has led to a pattern of
rejected bids in regular auctions of Treasury bills and Treasury bonds during the year. As a
result, the amount awarded has, at times, been less than the offer size. In the primary
auctions for Treasury bills, awards were below programmed offers on 13 out of 24 auction
dates.6 The Bureau of the Treasury (BTr) has cited the government’s comfortable cash
position and that the bids were deemed high as the reason for the rejections.
The BSP has expressed its concerns regarding the pattern of bid rejections. For one, the
cash management concerns of the national government have to be weighed against its
market-making role in the government securities market. Also, a steady and predictable
supply of government securities in primary markets is critical for the proper functioning of
credit markets. More importantly, the pattern of bid rejections leads to a higher differential
between the benchmark Treasury bill rates and the policy interest rates. It should be noted
that Treasury bill rates are used for pricing loans. Possible interest rate repression serves to
confuse signals on the price of funding for borrowers. Consequently, an impediment to the
efficient transmission of monetary policy has been artificially created.

6 In 2010, full awards were not made in 10 of 24 auction dates for Treasury bills.

BIS Papers No 67 273

3. Domestic currency public debt issues in local markets
3.1 Shift from international to domestic financial markets for public debt
Currently, there is sufficient liquidity in the domestic economy to obviate any concerns about
the crowding out of private offerings. The BSP has encouraged the national government to
access domestic financial markets and take advantage of this liquidity. It has even
encouraged the national government to access its foreign exchange needs through domestic
borrowing. This can be done in two ways. First, the national government can issue foreign
currency-denominated debt to residents because foreign exchange liquidity in the Philippine
financial system is also high. A second option is to issue domestic currency debt and then
exchange the proceeds with the BSP to meet the government’s foreign currency needs.
3.2 Implications for capital market development
Greater domestic borrowing would also promote the domestic capital markets. More
domestic issuance would create incentives for the development of market infrastructure. It
would also encourage more private firms to issue debt securities in domestic financial
markets. Furthermore, as the government is often one of the few safe issuers of long-term
debt, it would provide benchmarks that could then pave the way for private issuance of
longer-maturity debt securities. This could be a boon for financing long-gestation
(eg infrastructure) projects and could encourage greater private participation in infrastructure
development.
From the point of view of a central bank, additional resident-sourced sovereign debt would
also reduce the incentive of the government to inflate away its debt. With respect to external
debt management, domestic issuance would also reduce the currency risks faced by the
national government. Also, greater domestic issuance (especially in the domestic currency)
would also reduce financial stability concerns.
3.3 Lengthening maturity of domestic government bonds
Higher liquidity in domestic markets also provides the opportunity for stretching the maturity
of sovereign debt. The lengthening maturity of public debt would also reduce default risks for
borrowers since there would be less exposure to rollover risks or to liquidity risk. As a market
signal, issuance of debt with a longer maturity would signal relatively greater fiscal credibility
as demand for such debt paper would not be viable without fiscal credibility. Similarly, it
would also signal monetary credibility on the part of the central bank. However, the national
government should also ensure that its cash flows will meet future liabilities and it must avoid
bunching up on maturities. A bunching up on maturities poses risks for rolling over the debt
and may even lead to greater volatility in market interest rates.
3.4 Financial stability concerns
(a) Capital inflows and potential for faster transmission of external shocks
As most public debt is held by residents, the country’s susceptibility to the effects of a
sudden stop is reduced. Also, flows to domestic financial markets have been subdued
compared with those into other regional financial markets. Several factors such as the
relatively smaller size of domestic markets as compared to regional peers, limited offerings,
political uncertainty and risk aversion may explain this effect. As domestic financial markets
deepen, exposure to exogenous shocks may increase. However, since the distribution of
asset holdings in domestic financial markets has heavily favoured residents, the risk profile is
different than if the bulk were held by non-residents, who are more exposed to external
shocks.

274 BIS Papers No 67

(b) Reduced issuances as NG fiscal condition improved
Significantly, the dearth of liquidity in short-term Treasury bills has negative implications for
capital market development and long-term financial and macroeconomic stability. As the
supply of benchmark Treasury bills in primary markets dries up, interest rates fail to reflect
actual credit market conditions. Consequently, Treasury bills lose their usefulness as the
benchmark for market interest rates and as the basis for loan pricing. Furthermore, the bid
rejections and less-than-programmed award size make for higher volumes in succeeding
auctions, thereby feeding higher interest rate volatility. Lastly, as the Treasury bill rates are
repressed, investors seek higher returns in property and equities markets. These may feed
into asset bubbles as the lower rates contribute to mispricing.
4. BSP and public debt management
4.1 BSP’s issuance of its own securities
Under its charter, the BSP is not permitted to issue its own debt securities.7 Coordination with
the Department of Finance (DoF) on issuances of debt securities for the BSP has been
considered. However, concerns about such an arrangement have arisen from the
implications for the central bank’s independence, coordination difficulties and the potential
impact on the government’s credit ratings.
The national government’s issuance of debt securities on the BSP’s behalf may undermine
the central bank’s independence. It should be noted that the government takes on a debt
management perspective when it issues debt securities. The differing incentives arising from
the separate objectives of fiscal and monetary policy could pose conflicts of interest for the
Bureau of the Treasury, part of the Department of Finance, as the expected issuer.
Consequently, coordination may be difficult even if detailed agreements are made between
the monetary and fiscal authorities. Furthermore, as the national government operates in a
more politicised environment, the BSP could be forced to defend its operational decisions to
political forums. Such politicisation of the monetary policy implementation process represents
an unnecessary distraction in the conduct of monetary policy and imposes transaction costs
on the regular policy-setting process.
As capital inflows surge, the need for greater siphoning may ensue. This may require the
national government to increase its stock of debt by more than its programmed size.
Consequently, its willingness to continue issuing debt securities for the BSP may be
diminished. Should it continue to do so, the increase in its debt stock as well as the
corresponding negative effects on its debt ratios could trigger concerns on its credit ratings.
Based on the foregoing, it is deemed a better option for the BSP to pursue a proposed
amendment to its charter that would again authorise it to issue its own debt securities.8 If
granted, it would expand the scope of open market operations and enable better inflation
management especially in times of excess liquidity.

7 Section 92 (Issue and Negotiation of Bangko Sentral Obligations) of Republic Act no 7653 states that
“issuance of certificates of indebtedness shall be made only in cases of extraordinary movements in price
levels”.
8 The Charter of the old Central Bank of the Philippines (Republic Act no 265) granted it the authority to issue its
own debt securities.

BIS Papers No 67 275

4.2 BSP’s role in government debt management
Under Philippine law,9 all government borrowing, whether peso- or foreign currency-
denominated, require the approval of the Monetary Board. BSP staff examine the effects of
these borrowings on monetary aggregates, foreign exchange reserves, the balance of
payments and the sustainability of external debt.10 The implications of these borrowings for
monetary policy are also considered.
On a more direct basis, BSP representatives occupy two of the five seats in the auction
committee of the Bureau of the Treasury. Their participation in the Auction Committee affords
the BSP an inside view of primary markets for government securities and a unique vantage
point from which to monitor credit market trends and lending.
Beyond the opportunities afforded by its participation in the Auction Committee and by its
role in the approval of government borrowing, the BSP has not engaged in quasi-fiscal
operations and unconventional monetary policies as practised by a number of central banks
in advanced economies. Consequently, its balance sheet has not been exposed to shocks
arising from such practices.
However, the management of surges in capital flows has had a significant effect on its
balance sheet. As it accumulates foreign exchange reserves to manage the impact of capital
inflows on domestic liquidity and inflation as well as on the exchange rate, the BSP is
exposed to foreign exchange risk. Valuation losses from peso appreciation in the face of
strong FX inflows have negative implications for the BSP’s balance sheet and particularly for
its capitalisation.
4.3 Governance arrangements for the coordination of monetary policy and public
debt management
Within the BSP, the national government is able to coordinate monetary policy and public
debt management through a seat on the Monetary Board. The government representative on
the Monetary Board is currently the Secretary of the Department of Finance (DoF). In
instances that the DoF Secretary is unable to attend, the usual substitute has been the
Treasurer of the Philippines (ie the head of the agency that issues sovereign debt).
NEDA Board and related inter-agency committees11
Beyond the confines of the BSP, its participation in macroeconomic coordination is through
the Board of the National Economic and Development Authority NEDA Board).12 (Annex A
lists the composition of the NEDA Board.) To assist the NEDA Board in the performance of
its functions and duties, seven inter-agency cabinet level committees were formed. The BSP
is also involved in two of these seven committees. These are the Development Budget
Coordinating Committee (DBCC); and the Investment Coordination Committee (ICC).

9 Section 123 (Financial Advice on Official Credit Operations) of Republic Act no 7653
10 A ceiling on foreign currency-denominated debt is imposed to ensure sustainability of the external debt.
11 For further details, please see the website of the National Economic and Development Authority
(www.neda.gov.ph).
12 Macroeconomic planning and policy coordination is reposed in the National Economic and Development
Authority (NEDA). Its mandate is to formulate development plans and ensure their implementation in the
course of policymaking and policy coordination with other government agencies.

276 BIS Papers No 67

The Development Budget Coordinating Committee (DBCC)
The DBCC is composed of the Secretary, Department of Budget and Management as
chairperson; the Secretary, Department of Finance as co-chairperson; with the Executive
Secretary of the Cabinet and the Director General of NEDA as members. The DBCC is a
policymaking body which approves the macroeconomic assumptions and economic policy
directions for the preparation of the annual national government budget and for the
requirements of the government’s medium-term development plan. Specifically, the functions
of the DBCC are the following:
1. Recommend for presidential approval the level of the annual government
expenditure programme and the ceiling on government spending for social and
economic development, national defence, general government and debt service;
2. Recommend to the president the proper allocation of expenditures for development
activity between current operating expenditures and capital outlay; and
3. Recommend to the president the allocation for capital outlay under each
development activity for the various capital or infrastructure projects.
The BSP participates in the DBCC as a resource institution13 providing background
information on monetary and financial policy as well as perspectives on economic
developments.
The DBCC is the government body through which inflation targets are proposed by the BSP
and approved by the economic managers who comprise the Committee. In essence, while
the BSP enjoys operational and instrument independence, it is not fully independent in that
its inflation targets must be agreed by the DBCC. The central bank participates in the DBCC
both at the technical staff level and at the level of senior officials. At the technical staff level,
BSP staff provide input on the formulation of macroeconomic assumptions relating to inflation
rates, exchange rates, interest rates, oil prices and banking trends. Subject to these
constraints, the fiscal budget is computed. The output from the technical level is then
forwarded to senior officials who define policy priorities, finalise assumptions and make
recommendations to the President. Once finalised, the draft budget is presented to Congress
for enactment into the annual General Appropriations Act. The BSP attends budget
deliberations in Congress to brief the legislators on economic and financial developments as
well as explain the macroeconomic assumptions in its sphere of influence (eg interest rate
assumptions).
Under the DBCC is the Executive Technical Board (DBCC-ETB) which is responsible for
implementing the policy directions firmed up at the cabinet level. The DBCC-ETB is the
screening and review body for policies, measures and targets that are recommended to the
DBCC. It consists of undersecretaries and directors of the DBCC member agencies, which
include: the Department of Budget and Management (DBM), the Department of Finance
(DOF), the National Economic and Development Authority (NEDA) and the Office of the
President (OP). The DBCC-ETB is chaired by the DBM Undersecretary and receives
technical support from DBM that serves as the ETB Secretariat. The DBCC and the DBCC-
ETB work through the DBCC Secretariat, which is chaired by the Director of the Fiscal
Planning Bureau of the Department of Budget and Management. Similar to the DBCC, the
BSP participates as a resource agency providing input to its processes. During an October
2008 meeting of the ETB, the ETB Chairperson clarified the following rights and obligations
of the BSP as a resource institution in the DBCC and in its committees:

13 Per Executive Order no 232 dated 14 May 1970, the central bank was an original member of the Presidential
Development Budget Committee (PDBC) which was renamed DBCC in 1972. The Administrative Code of
1987 ordered the replacement of the central bank by the Executive Secretary in the DBCC membership.

BIS Papers No 67 277

• The BSP’s presence in the meetings is counted to establish a quorum;
• The BSP is not a voting member (ie “it does not sign resolutions”).
Furthermore, core functions are delegated to Technical Working Groups (TWGs) or Sub-
committees. The first of these is the Cash Programming and Monitoring Committee (CPMC).
This is tasked with closely monitoring the fiscal performance of the national government and
formulating fiscal policies for recommendation to the DBCC. It is chaired by the Treasurer of
the Philippines with the Bureau of the Treasury (BTr) as its secretariat. Then, there is the
Technical Working Group on Macroeconomy and Development Financing. Its job is to
develop macroeconomic models and other statistical tools for planning, forecasting and
policy analysis as well as to monitor macroeconomic performance and make economic
reports. It is run by an Assistant Director General from NEDA, and NEDA’s National Policy
and Planning Staff (NPPS) serves as its secretariat. There is also the Sub-committee on
Government-Owned and Controlled Corporations (GOCCs). Its task is to monitor the cash
flow of the government corporate sector and formulate policies affecting government-owned
and controlled corporations for recommendation to the DBCC. It is headed by an
undersecretary of the Department of Finance and has the DoF’s Corporate Affairs Group as
its secretariat. Lastly, the Technical Working Group on Program Loans participates in policy
formulation regarding external resource mobilisation. It is headed by an undersecretary from
the Department of Finance and has NEDA’s Public Investment Staff as its secretariat. The
organisational chart of the DBCC is given below (Figure 2).
Figure 2
Development Budget Coordination Committee Organisation

The Investment Coordination Committee (ICC)
The ICC is composed of the Secretary, Department of Finance as chairperson with the
Director General of NEDA as co-chairperson. Its members include the Executive Secretary of
the Cabinet, the Secretary, Department of Agriculture, the Secretary, Department of Trade
DBCC
Chair: DBM Secretary
ETB
Chair: DBM Undersecretary
CPMC
Chair: Treasurer of
the Philippines
TWG on Macroeconomy
and Devt Financing
Chair: NEDA ADG
Sub-Com on
GOCCs
Chair: DOF Usec
TWG on Program
Loans
Chair: DOF Usec
DBCC Secretariat
Chief Director, FPB (DBM)
Secretariat: BTr Secretariat: NEDA-NPPS Secretariat: DOF-CAG Secretariat: NEDA-PIS

278 BIS Papers No 67

and Industry, the Secretary, Department of Budget and Management and the Governor of
the BSP.14 Its functions under Philippine law are:
1. Evaluate the fiscal, monetary and balance of payments implications of major
national projects and recommend to the president the timetable of implementation of
these projects on a regular basis; and
2. Recommend to the president a domestic and foreign borrowing programme updated
each year, and subsequently, submit to the president a status of the fiscal, monetary
and balance of payments implications of major national projects
Aside from formal arrangements for policy coordination, the BSP Governor and the
Secretary, Department of Finance along with key officials from their respective agencies hold
informal meetings every month. These are usually scheduled every last Friday or Tuesday of
a given month. The meetings provide an additional venue for discussing fiscal performance
and its implications for fiscal and monetary policy coordination.
5. Conclusion
The fiscal health of the Philippines has improved significantly over the past decade.
Notwithstanding the dividends from reforms, challenges remain for the Philippines on the
fiscal side.
Econometric results suggest that the BSP and the national government have coordinated
their policy actions so that policy sterilisation has been avoided. While interaction between
monetary and fiscal policy has been productive during the financial crisis and its aftermath,
some concerns have been raised by the BSP on the national government’s reduced
issuance of government securities as well as possible interest rate repression.
With respect to the shift from international to domestic financial markets for public debt, it
should be noted that there is sufficient liquidity in the domestic economy to obviate concerns
about the crowding out of private offerings. The central bank is even urging the national
government to source more of its financing from domestic markets.
The higher liquidity in domestic markets also provides the opportunity for lengthening the
maturity of sovereign debt. However, the national government should also ensure that its
cash flows will meet future liabilities as they come and it must avoid bunching up on
maturities.
As this issue impinges on financial stability and especially on the attractiveness of public debt
to foreign investors, it should be noted that most public debt has historically been held by
residents. By the same token, the country’s susceptibility to the effects of a sudden and
sharp outflow of foreign currency has been limited. However, the dearth of liquidity for short-
term Treasury bills has negative implications for capital market development and long-term
financial and macroeconomic stability.
The BSP’s role in public debt management has been to examine the effect of public debt
issuance on the key macroeconomic variables under its purview. Under Philippine law, all
government borrowing, whether peso- or foreign currency-denominated, requires the
approval of the Monetary Board. On a more direct basis, representatives of the BSP occupy
two of the five seats in the Bureau of the Treasury’s auction committee.

14 The Governor is represented in the ICC by a senior member of the Monetary Board.

BIS Papers No 67 279

Ironically, the BSP is unable to directly participate in domestic capital markets, being
prohibited by its charter from issuing its own debt securities. The issuance of debt securities
for the BSP by means of a collaboration with the Department of Finance has been
considered. However, concerns about such an arrangement have arisen from the
implications for the central bank’s independence, coordination difficulties and the potential
effect on the government’s credit ratings.
Within the BSP, the national government coordinates monetary policy and public debt
management through a seat on the Monetary Board. Beyond the confines of the BSP, the
government participates in macroeconomic coordination through the Board of the National
Economic and Development Authority (NEDA Board).
The main body through which policy coordination has been conducted has been the
Development Budget Coordinating Committee. Aside from formal arrangements for policy
coordination, the BSP Governor and the Secretary for the Department of Finance, along with
key officials from their respective agencies, hold informal meetings every month. The
meetings provide an additional venue for discussing fiscal performance and its implications
for fiscal and monetary policy coordination.

280 BIS Papers No 67

Annex A:
Policy responses to the global financial crisis:
the Philippine Case
In 2010, the Southeast Asian Central Bank Training and Research Centre (SEACEN)
initiated an international research project on the “Relative effectiveness of policy choices
during the global financial crisis”. Individual country studies were conducted complete with
econometric modelling based on a vector error correction specification.
The Philippine case15 used an empirical model of the form
Yt = f(MPt, FPt, Zt)
where Yt is a measure of economic activity and Zt refers to other relevant variables while MPt
and FPt correspond to monetary and fiscal policy responses, respectively. For simplicity but
without loss of generality, the logarithmic form16 of real GDP (LGDP) was used as Yt; the log
of M3 or domestic liquidity (LM3) represented MPt, the log of government purchases of
goods and services (LGOVCONS) served as FPt and an indicator of financial markets
volatility, SQR_RPHISIX17 as Zt. With the exception of SQR_RPHISIX, all the data have
been seasonally adjusted. In the error correction specification used, two dummy variables
were created to represent the Asian financial crisis (AFC) and the global financial crisis
(GFC).18 For the AFC dummy variable, its value was equal to one from the third quarter of
1998 to the fourth quarter of 1999 and zero, otherwise. For the GFC dummy variable, its
value was equal to one from the third quarter of 2007 to the third quarter of 2009 and zero,
otherwise.
The model used data from the first quarter of 1995 to the second quarter of 2010. It was
recognised that, despite being a relatively robust and flexible model specification, the VECM
did not differentiate completely between the pre-crisis and post-crisis periods.
The key findings from econometric estimation are as follows:
1. There appears to be a weakening of long-run economic relationships19 arising from
the recent global financial crisis.
Time and again, tests for co-integration were negated by findings of co-integration
breakdown at the endpoints. Explicit inclusion of financial turbulence finally
generated a viable co-integration framework for analysis.

15 See Redoblado (2011).
16 The use of logarithmic form was necessary to account for the non-linearity arising from the crisis episodes in
the sample. Also, vector autoregressive (VAR) models and vector error correction models (VECM) are linear
approaches that do not necessarily mitigate non-linearities in the data.
17 The indicator SQR_RPHISIX was computed as the squared residual from an autoregressive model of the
Philippine Composite stock market index, PHISIX. SQR_RPHISIX was included with a view to incorporating
uncertainty in financial markets and providing information on general financial turbulence. Its inclusion was
deemed a necessity in rendering long-run economic relationships stable.
18 The dummy variables accounted for the duration of the crisis episodes. For a discussion of the methodology,
see Harding and Pagan (2002).
19 It was found that the link between output, fiscal and monetary policy variables had weakened since the second
quarter of 2007. Varying permutations of these variables were tested. These permutations varied in terms of
nominal versus real terms, seasonally adjusted versus those with seasonal data, ratios to GDP, differing
indicators for monetary policy and fiscal policy. Attempts to include other variables such as remittances were
also made.

BIS Papers No 67 281

2. The economy was more responsive to monetary policy action than to fiscal stimulus.
Significantly, monetary policy has a stronger and quicker stimulus effect than fiscal
policy. This is not surprising given that fiscal authorities have to tackle coordination
and planning difficulties in crafting and implementing fiscal policies.
3. There is econometric evidence that the BSP initiates monetary stimuli insofar as it
does not conflict with the price stability objective.
There is some evidence to show that, while GDP is more responsive to innovations
in monetary policy than to fiscal policy shocks, monetary policy becomes
countercyclical after roughly five quarters. This may be interpreted as heading off
the inflationary effects of output growth as the slack in the economy is reduced or
even eliminated.
4. Macroeconomic coordination has supported the policy response.
The impulse response analysis shows that the BSP and the national government
coordinated their policy actions so that policy sterilisation did not occur.

282 BIS Papers No 67

Annex B:
Composition of the NEDA Board
This mandate is exercised through the NEDA Board which is composed of the following:
1. President of the Republic of the Philippines (as chairperson)
2. Secretary of Socio-economic Planning and NEDA Director General (as vice
chairperson)
3. The Executive Secretary of the Cabinet
4. Secretary, Department of Finance
5. Secretary, Department of Trade and Industry
6. Secretary, Department of Agriculture
7. Secretary, Department of Environment and Natural Resources
8. Secretary, Department of Public Works and Highways
9. Secretary, Department of Budget and Management
10. Secretary, Department of Labor and Employment
11. Secretary, Department of the Interior and Local Government
Over the years, the following members of the cabinet were added:
1. Secretary, Department of Health
2. Secretary, Department of Agrarian Reform
3. Secretary, Department of Foreign Affairs
4. Secretary, Department of Science and Technology
5. Secretary, Department of Transportation and Communications
6. Secretary, Department of Energy
7. Deputy Governor, The Bangko Sentral ng Pilipinas20

20 This conforms to Section 124 (Representation on the National Economic and Development Authority) of the
BSP Charter.

BIS Papers No 67 283

References:
Harding, D and A Pagan (2002): “Dissecting the cycle: a methodological investigation”,
Journal of Monetary Economics, vol 49, pp 361–85.
Redoblado, J (2011): “Policy responses to the global financial crisis: the Philippine case”, in
Relative Effectiveness of Policy Choices During the Global Financial Crisis in SEACEN
Economies, by R Dheerasinghe (ed), pp 391–5, Kuala Lumpur: The South East Asian
Central Banks (SEACEN).
Executive Order no 232 dated 14 May 1970.
National Economic Development Authority (NEDA) website, www.neda.gov.ph.
Republic Act (RA) no 265, Old Central Bank Charter, 15 June 1948.
RA no 7653, The New Central Bank Act, 14 June 1993.

BIS Papers No 67 285

A framework for fiscal vulnerability assessment
and its application to Poland
7RPDV]�-ĊGU]HMRZLF]��:LWROG�.R]LĔVNL1
Abstract
The sharp worsening of fiscal positions in the aftermath of the global economic crisis has
brought the issues of fiscal sustainability to the fore of economic policy debate. This has
focused the attention of policymakers on the broader implications of unsustainable fiscal
positions, including the consequences for monetary policy and financial stability. As a result,
there is now an increased need for central banks to closely monitor risks to fiscal
vulnerability. This note proposes a framework for such an assessment, consisting of five
elements: (i) the level of public debt; (ii) the medium-term dynamics of public debt; (iii) long-
term sustainability of public debt; (iv) public debt management and the liquidity position of the
government; and (v) fiscal rules and institutions. The note also presents a brief assessment
of Poland’s fiscal vulnerability using the framework described above, and finds that Poland’s
vulnerability to fiscal risks is quite limited, although there is still a need to correct the fiscal
imbalances that could otherwise lead to a build-up of public debt.

Keywords: Public debt, fiscal vulnerability, fiscal sustainability
JEL classification: H63

1 Economic expert and Vice President, National Bank of Poland, respectively. The authors are grateful to
$QGU]HM� 6áDZLĔVNL� IRU� KLV� YDOXDEOH� FRPPHQWV�� $Q\� HUURUV� RU� RPLVVLRQV� DUH� WKH� VROH� UHVSRQVLELOLW\� RI� WKH�
authors. The views expressed in the paper are those of the authors and not necessarily those of the National
Bank of Poland.

286 BIS Papers No 67

Introduction
The global economic crisis has led to a sharp worsening of government finances all over the
world. Discretionary fiscal stimulus measures, automatic fiscal stabilisers, reversal of
extraordinary revenue windfalls associated with asset price bubbles and the cost of
government support to ailing financial institutions have all contributed to this worsening. As a
result, public debt in advanced economies has risen from 73% of GDP in 2007 to 104% of
GDP in 2011 and is projected to grow further.2 Fiscal problems have manifested themselves
with particular starkness in the form of a sovereign debt crisis in the euro area, where at least
one member state is now widely considered to be insolvent.
These worrisome developments have brought the issues of fiscal sustainability to the fore of
the economic policy debate. This has focused the attention of policymakers on the broader
implications of unsustainable fiscal positions, including the consequences for monetary policy
and financial stability. Blommestein and Turner (2011) argue that the current fiscal
environment has set the stage for a new period of fiscal dominance, undermining the
traditional division of labour between monetary, fiscal and public debt management
authorities. Central bank interventions in sovereign bond markets, in some cases on a
considerable scale, are one sign of this. Meanwhile, financial stability is being undermined by
the reassessment of risk associated with sovereign bonds held by the financial sector,
previously considered risk-free. As noted by Das et al (2011), the relationship between public
debt vulnerability and financial stability tends to be procyclical. During a downswing,
especially one triggered by financial sector dislocation, maintenance of the asset quality of
the government’s liabilities is much more critical in containing adverse developments in the
real and financial sector. In such a situation, any threat to government solvency will have a
negative impact on financial institutions’ balance sheets, incomes and capital reserves. This
may in turn result in the need for government support to bank resolution and restructuring,
ultimately leading to a vicious circle of deteriorating government and financial sector balance
sheets.
Another important recent development to be noted in this context is the increased aversion of
financial markets to sovereign risk. Schuknecht et al (2010) show that the strength of the
response of euro area government yield spreads to a higher public debt ratio increased
eightfold in the period following the collapse of Lehman Brothers in September 2008. This
increases the risk of a negative self-reinforcing feedback between fiscal sustainability risks
and the financial market perception thereof. As has been evident in the euro area sovereign
debt crisis, notably in the case of Italy, financial market concerns can aggravate fiscal
sustainability problems, as an increase in government bond yields translates into a higher
interest burden, thus increasing the size of the adjustment required to stabilise the debt ratio.
In this light, there is now a heightened need for close monitoring of risks to fiscal
vulnerability, a role that should be performed by central banks. This note presents a
framework for such an assessment and briefly describes its application to Poland.
1. Level of public debt
The headline ratio of public debt to GDP is the most commonly used measure of government
solvency. The government budget constraint states that the current level of public debt must
be repaid with future primary surpluses. Therefore, the higher the debt ratio, the more difficult
it will be for government to generate sufficient surpluses. A higher debt ratio also implies a

2 Source: IMF WEO database.

BIS Papers No 67 287

history of fiscal indiscipline, complicating the task of turning the situation around. However,
the debt ratio is by no means a comprehensive measure of government solvency. There is
also no clear consensus in the literature on what level of public debt may be considered safe.
Some guidance on critical levels of public debt is provided by literature examining public debt
developments in countries undergoing financial crises, including sovereign defaults.
However, as these events have been largely confined to emerging market countries, at least
in the period since the Second World War, empirical findings are also relevant for this group
of countries. Reinhart et al (2003) put forward the concept of “debt intolerance” and grouped
a sample of 53 developing and developed3 countries according to their measure of debt
intolerance. According to these authors, the level of debt intolerance may be explained by
the average level of long-term foreign indebtedness and an insolvency risk index. In the case
of countries with the highest debt intolerance, a foreign debt level of just 15% of GDP may
already indicate a risk of insolvency. Daniel et al (2004) have analysed insolvency episodes
in emerging countries over the past 30 years and found that 55% of them occurred when
public debt was below 60% of GDP, while 35% of the cases occurred in countries with a debt
ratio below 40% of GDP.
Daniel et al point out that the risk of insolvency is greater in emerging market countries due
to, among other factors, a lower and more variable ratio of government revenue to GDP and
a lower quality of institutions. Therefore, results obtained using a sample of emerging market
countries may not be directly translated into conclusions regarding advanced economies.
While there is no sample of defaults in advanced economies, there are a number of quite
recent studies that seek to determine the safe level of public debt from the viewpoint of its
effect on economic growth. The general finding of these studies is that the negative impact of
public debt on growth is non-linear and becomes significantly stronger once debt exceeds a
critical threshold of around 90–100% of GDP. Checherita-Westphal and Rother (2010) have
analysed this impact on a sample of 12 euro area countries and found that, in some of these
cases, the critical threshold could be as low as 70% of GDP. Furthermore, the negative
impact of debt on GDP growth appears earlier when there is higher volatility of inflation,
interest rates and government spending. Kumar and Woo (2010) obtained similar results
using a panel of advanced and emerging economies. Their results indicate, that during the
period analysed (1970–2007), an average increase in the debt ratio by 10 percentage points
of GDP slowed economic growth by 0.2 percentage points annually, although the effect was
slightly weaker (0.15) for advanced economies. After accounting for a potential non-linear
relationship between these variables, the authors found that the negative impact of rising
public debt on economic growth becomes statistically significant only once debt exceeds
90% of GDP. Reinhart and Rogoff (2010) also observed a similar relationship, in addition
indicating that in the case of emerging market economies, the critical level of external debt
(public and private) amounts to 60% of GDP.
2. Medium-term dynamics of public debt
The change in the debt-to-GDP ratio may be decomposed into the primary balance, interest
rate on government debt and the growth rate of the economy using the well-known equation.

where:

3 The sample used in the study dates back 200 years in some cases, thereby covering sovereign default
episodes in countries currently classified as developed.

288 BIS Papers No 67

d – debt-to-GDP ratio
r – interest rate
g – GDP growth
p – primary balance ratio to GDP
In order to obtain a true medium-term picture of public debt trends, one may apply cyclically
adjusted figures to the equation – ie the cyclically adjusted primary balance and a potential,
rather than actual, growth rate. However, at present, uncertainty regarding the cyclical
position of the economy is particularly high.
Medium-term prospects for the development of the debt-to-GDP ratio crucially depend on a
country’s primary balance, the growth prospects of the economy and the risk premium
attributed to the sovereign debt of the country in question. At the current juncture, the
majority of advanced economies (20 out of 27 for which the IMF WEO provides data) are
running a primary deficit, while potential growth estimates for this group of countries are
historically low. This implies that public debt ratios around the world are not only high, but
also face unfavourable medium-term trends.
3. Long-term sustainability of public debt
As noted by Baldacci et al (2011), fiscal solvency also depends on the extent to which long-
term demographic and economic trends will put pressure on the budget. The majority of
advanced economies are projected to face substantial fiscal pressures in the coming
decades due to population ageing. IMF (2009) calculations show that for advanced countries,
the net present value of the cost of ageing is about nine times higher than the estimated
fiscal burden of the global economic crisis. A fiscal vulnerability analysis should also take
these risks into account.
4. Public debt management and liquidity position of the government
An assessment of fiscal vulnerability also needs to incorporate the structure of the
government’s balance sheet. As noted by Das et al (2010), the structure of public debt may
become a channel or source of vulnerability to the real economy and the financial system.
Therefore, this structure should be designed in such a way as to mitigate risk both for the
government and for markets. Das et al point to two main sources of vulnerability – foreign
currency-denominated liabilities and short-term liabilities. The difficulty in shaping the optimal
debt structure lies in the trade-off between the cost of financing public debt and the risks of a
given structure. In the past, emerging market economies often relied on foreign-currency
borrowing because it was easier to obtain, particularly in the context of less developed local
financial markets. Similarly, the cost of short-term borrowing, particularly in less developed
financial markets, is also lower thanks to the lower risk premia. However, short maturities
entail high rollover and refinancing risk, making the government prone to confidence crises.
The risks associated with large shares of foreign currency and short-term borrowing are
illustrated in the financial crisis literature. Short-term debt, usually measured as a total of
short-term public and private debt, plays a particularly important role. Hemming et al (2003)
show that the share of short-term debt is usually considerably higher in periods directly
preceding financial crises. As shown by Furman and Stiglitz (1998), the example of Asian
crises indicates that excessive reliance on short-term funding leads to the risk of a self-
fulfilling sudden-stop crisis.
As shown by Attinasi et al (2011) using a sample of euro area countries, the structure of
public debt and other aspects of sovereign debt management have an impact on sovereign

BIS Papers No 67 289

yields in advanced countries too. Government bond spreads relative to German bunds are
shown to be higher with a lower residual maturity of public debt and lower with a higher share
of long-term outstanding debt. Spreads are also shown to be higher for countries with a lower
liquidity in their sovereign debt markets.
5. Fiscal rules and institutions
Governments worldwide face a dilemma: they need to consolidate their fiscal position while
safeguarding very fragile economic growth. A solution often suggested in the economic
debate is for governments to make a firm commitment to fiscal consolidation, but to extend
its implementation over a number of years, so as not to harm growth in the short run. The
problem with this proposal is the credibility of the commitment to consolidate public finances.
Once financial markets sense a threat of insolvency, no mere declaration about future
adjustment measures is likely to dispel these fears, due to the obvious political risks
surrounding such a declaration. A possible way to resolve this problem is through the
introduction of fiscal rules and institutions that might help to convince financial markets and
other economic agents that sound fiscal policies will be adhered to.
There is an extensive body of literature devoted to studying the impact of fiscal rules and
institutions, based on the experience of US states, as well as European governments. The
literature (see European Commission (2011) for a brief overview) generally finds that
stronger fiscal frameworks are correlated with better fiscal outcomes, although identification
of the channels through which this effect takes place has not been straightforward.
Since better fiscal frameworks have a positive impact on fiscal outcomes, while the latter are
known to be a key determinant of government bond yields, the implication would be that
strong fiscal rules and institutions should improve a government’s fiscal prospects as well as
the financial markets’ perception thereof. This relationship has been shown empirically to
hold in the US states, and, in more recent literature, in the EU countries. In particular, Iara
and Wolff (2010) show that strong fiscal rules are of great importance in containing sovereign
bond spreads, particularly in times of elevated market uncertainty. Under extreme
circumstances, stronger fiscal rules can reduce sovereign bond spreads between euro area
member states and Germany by as much as 80 to 100 basis points. The legal basis of the
fiscal rules in force is found to be of particular importance. The authors argue that national
fiscal rules have a beneficial effect by reducing the uncertainty of market expectations of
fiscal variables, which is particularly important in times of higher risk aversion.
Feld et al (2011) show similar findings based on a study using data on Swiss cantons. The
presence and the strength of fiscal rules in the cantons are deemed to contribute to lower
risk premia. These effects are quantitatively quite significant, as the introduction of a strong
fiscal rule may contribute to a decline in risk premia of more than 10 basis points.
In view of these findings regarding the impact of fiscal frameworks both on fiscal outcomes
and on the financial markets’ perception of sovereign risks, such frameworks may be
considered another element of the assessment of fiscal vulnerability. Strong and credible
rules clearly appear to be an important element in limiting fiscal risks.
Poland: an assessment of fiscal vulnerability
Fiscal deficits increased sizeably in Poland during the economic crisis. This was related to a
number of factors, but discretionary anti-crisis stimulus measures or government support for
the financial sector were not among them. Fiscal policy had been loosened, but this was
related to cuts in taxes and social contributions that had been put in place before the onset of
the crisis, as well as a considerable increase in public investment, partly related to the
co-financing of EU funds and largely taking place at the local government level. In addition,
while Poland is widely known to be the only country in the EU to escape recession in 2009,

290 BIS Papers No 67

the economy did slow considerably and automatic stabilisers were allowed to come into
effect. According to European Commission estimates, the output gap deteriorated by
2.9 percentage points of GDP in 2009, as compared with an average worsening of 5.2 points
in the EU.
Despite the increase in the deficit to 7.1% of GDP in 2009 and to 7.8% in 2010, public debt
increased only moderately, owing to continued economic growth and to privatisation receipts
that helped to offset the government’s high borrowing requirements. Debt in ESA95 terms
rose from a low of 45% of GDP in 2007 to 56.7% of GDP in 2011 (European Commission
estimate). This is a quite moderate level, even taking into account that Poland may not yet
have fully graduated from the emerging market into the advanced economy club.
The general government primary deficit rose to around 5% of GDP in 2009–10, causing a
sharp increase in the ratio of public debt to GDP. At the same time, the good performance of
the economy meant that, even in those two years, nominal GDP growth (5.3%) was on
average only slightly lower than the interest rate on public debt (5.7%). The outlook for public
debt dynamics may be expected to improve considerably in the coming years, notably in
connection with fiscal consolidation measures implemented in 2011–12. As a result of these
measures, the primary deficit is expected to decline to well below 1% of GDP in 2012.4
Although even this proportion may not yet be considered a safe level, the government’s
declarations point to continued fiscal consolidation. Assuming these plans are fulfilled and
that economic growth gradually recovers, the government is expecting the ESA95
debt-to-GDP ratio to decline steadily to around 50% of GDP in 2015.5

Chart 1 Chart 2
Maturity structure of domestic
public debt in Poland
Currency structure of government
debt* in Poland
(% of GDP)

Source: Ministry of Finance. * Public debt according to the domestic definition
Source: Ministry of Finance.

4 According to the latest IMF forecast (December 2011), the general government deficit is expected to fall to
3¼% of GDP in 2012. Interest payments may be expected to reach close to 3% of GDP.
5 According to “The Public Finance Sector debt management strategy in the years 2012–15” accompanying the
2012 budget draft, submitted to Parliament in December 2011.

BIS Papers No 67 291

The financial crisis has not affected the debt management strategy and the structure of
public debt to any large extent. As a result of the liquidity crisis in 2008, the debt managers
offered more Treasury bills that year, as shown in Chart 4. However, this was a temporary
development that did not materially impact the maturity structure of public debt – the average
maturity declined slightly in 2009, but stayed at a level well above that seen in 2004–07
(Chart 1). The currency structure of debt issuance changed more visibly during 2009–10, as
the share of bonds issued on international markets increased, driven partly by strong
demand.6 The overall increase in net borrowing requirements in 2009–10 has been financed
largely by higher foreign borrowing, the transfer of road construction financing to the National
Road Fund 7 and, particularly in 2010, increased privatisation receipts. Meanwhile, net
domestic issuance by the Finance Ministry debt managers has remained broadly stable at a
level of around 3% of GDP (Chart 3).

Chart 3 Chart 4
Financing of net borrowing
requirements of the
central government
(% of GDP)
Financing of gross borrowing
requirements of the
central government
(% of GDP)

Source: Ministry of Finance. Source: Ministry of Finance.

The increased share of foreign financing does imply an increase in exchange rate risk,
although this shift was quite moderate. Nonetheless, the elevated exchange rate volatility
during 2011 has been perceived by the markets as a factor that could lead to a large
increase in the debt-to-GDP ratio and even cause it to reach the prudential threshold of 55%
of GDP. In the end, these fears have proved to be unfounded, after NBP interventions have
helped to reduce the exchange rate volatility.
The higher share of foreign financing at the same time implies a lengthening of the average
maturity, as foreign debt is generally issued for longer periods. The overall (domestic and
foreign) average maturity of central government debt reached an all-time high of 5.45 years
at the end of the third quarter of 2011.

6 For example, in mid-July 2009, Poland issued benchmark bonds worth $2 billion in the US market but, in view
of exceptionally strong demand, the issue was reopened two weeks later and another $1.5 billion was issued.
7 The Fund obtained funding independently – from loans from international financial institutions (eg the EIB) and
through issuance of “road bonds” conducted by the state bank BGK.

292 BIS Papers No 67

Government financing is also exposed to rollover risk. In Poland’s case, the scale of this risk
is, inter alia, related to the share of non-resident investors on the domestic Treasury bond
market. This share is currently at one of the highest levels in the market’s history, having
risen to almost 32% in the autumn of 2011, up from a low of 15% in early 2009. If non-
resident investors were to suddenly withdraw, this could potentially imply liquidity problems
for public debt management. As a precautionary measure, the government maintains
sizeable cash reserves that averaged around 2.6% of GDP in 2011. For emergency use, the
government has also secured a Flexible Credit Line (FCL) from the IMF.
Poland’s population is currently projected to age more rapidly than most others in the EU –
the old-age dependency ratio being projected to increase from 21% in 2010 to 71% in 2060,
while in the EU on average it is set to rise from 28% to 58%.8 In spite of this, according to the
European Commission Sustainability Report 2009, the long-term impact of ageing on public
finances is among the lowest in the EU, as ageing-related expenditure is actually projected to
decline by 1.1 percentage points of GDP between 2010 and 2060.

Chart 5
Projected change in ageing-related public expenditure
between 2010 and 2060
(percentage points of GDP)

Source: European Commission Sustainability Report 2009.

The underlying reason is a far-reaching reform of the pension system introduced in 1999,
which transformed a defined-benefit pension system into a partly-funded defined-contribution
system. As a result, future pension benefits will be directly linked to an individual’s career
history and will take into account life expectancy. In addition, pensions will be partly financed
from a mandatory funded pillar, relieving the pressure on public finances.
Poland has a well established fiscal policy anchor, consisting of a constitutional ceiling on
public debt of 60% of GDP, accompanied by prudential thresholds of 50% and 55% of GDP.
These limits are set out in the public finance act, the breaching of which triggers fiscal
adjustment measures. The constitutional basis makes the rule an exceptionally strong one.
Besides Germany, Poland is the only country in the EU with a constitutional fiscal rule.9 The

8 Source: Eurostat. Dependency ratio calculated as ratio of number of persons aged 65+ to those aged 20–64.
9 Source: European Commission database on numerical fiscal rules. Spain adopted a constitutional fiscal rule in
2011 which is scheduled to go into effect in 2020.

BIS Papers No 67 293

Polish fiscal rule framework has been in force since 1998, meaning that there is already a
considerable body of evidence on its effects. In that time, the debt ratio has exceeded the
50% of GDP threshold and approached the 55% threshold on two occasions, in 2003–04 and
2011–12. In both these instances, the rule signalled the need for fiscal consolidation and
policymakers have acted upon this signal. In 2003–04, the government presented and partly
implemented the so-called Hausner plan of public expenditure reform, encompassing a
number of structural measures that will curb spending in the medium to long term. In
2011–12, the government has also implemented a large-scale fiscal consolidation
programme, yielding a cumulative fiscal adjustment of close to 4 percentage points of GDP.
Overall, the assessment presented above indicates that Poland’s vulnerability to fiscal risks
is quite limited. There is still room for improvement as regards the level of public debt and, in
particular, its medium-term dynamics, since there is a need to correct the general
government primary deficit. However, longer-term fiscal prospects appear to be quite
favourable, as a result of the pension reform introduced in 1999. In addition, Poland’s fiscal
rule framework may be viewed as an additional factor that will contribute to the soundness of
fiscal policies in the future.
Bibliography
Attinasi M.G., Checherita-Westphal C., Rodriguez-Vives M., Slavik M. (2011): Sovereign
debt management in the EU: recent trends and practices, presentation at the ECB Public
Finance Workshop, October 7, 2011
Baldacci E., McHugh J., Petrova I. (2011): Measuring Fiscal Vulnerability and Fiscal Stress:
A Proposed Set of Indicators, IMF Working Paper
Blommestein H.J., Turner P. (2011): Interactions between sovereign debt management and
monetary policy under fiscal dominance and financial instability, paper presented at the ECB
Public Finance Workshop, October 7, 2011
Checherita-Westphal C., Rother P., (2010): The impact of high and growing government debt
on economic growth: an empirical investigation for the euro area, ECB Working Paper
Daniel J.A., Callen T., Terrones M., Debrun X., Allard C., (2004): Public debt in emerging
markets: is it too high?, proceedings of the Banca d’Italia workshop on Public Debt
Das U.S., Papapioannou M., Pedras G., Ahmed F., Surti J. (2010): Managing Public Debt
and Its Financial Stability Implications, IMF Working Paper
European Commission (2009): Sustainability Report 2009
European Commission (2011): Public Finances in EMU 2011
Feld L., Kalb A., Moessinger M.D., Osterloh S. (2011): Sovereign bond market reactions to
fiscal rules and no-bailout clauses: the Swiss experience, paper presented at the European
Commission workshop on Public finances in times of severe economic stress: the role of
institutions
Furman J., Stiglitz J. (1998): Economic Crises: Evidence and Insights from East Asia,
Brookings Papers on Economic Activity: 2, Brookings Institution, pp. 1–135
Hemming R., Kell M., Schimmelpfennig A. (2003): Fiscal Vulnerability and Financial Crises in
Emerging Market Economies, IMF Occasional Paper
Iara A., Wolff G. (2010): Rules and risk in the euro area: does rules-based national fiscal
governance contain sovereign bond spreads? European Commission Economic Paper
International Monetary Fund (2009): Fiscal Implications of the Global Economic and
Financial Crisis, Staff Position Note

294 BIS Papers No 67

Kumar M.S., Woo, J., (2010): Public Debt and Growth, IMF Working Paper
Ministry of Finance (2010): 2009. Public debt in Poland. Annual Report
Reinhart C.M., Rogoff K. (2010): Growth in a Time of Debt, NBER Working Paper
Reinhart C.M., Rogoff K., Savastano M.A. (2003): Debt Intolerance, NBER Working Paper
Schuknecht L., von Hagen J., Wolswijk G. (2010): Government bond risk premiums in the EU
revisited. The impact of the financial crisis, ECB Working Paper.

BIS Papers No 67 295

Fiscal policy, public debt management and government
bond markets: issues for central banks
Elizaveta Danilova1
Abstract
This paper covers fiscal policy tools that promote sustainability and their influence on
monetary policy in Russia, the Bank of Russia’s role in public debt management and main
features of the domestic currency public debt market. At present Russia’s budgetary
circumstances are quite favourable, however, the budgetary system remains exposed to the
world financial markets conditions and, above all, to oil prices. To reduce the budget’s
dependence on oil and gas revenues and accumulate reserves in the event that oil prices
UHWUHDW� ɚ� VSHFLDO� PHFKDQLVP� IRU� WKH� XWLOLVDWLRQ� RI� RLO� DQG� JDV� UHYHQXHV� ZLWKLQ� WKH� IHGHUDO�
budget has been envisaged. Operations to finance the budget deficit affect the Bank of
Russia’s monetary policy, however, this effect is not significant. The Bank of Russia’s role in
public debt management is focused on consulting the Ministry of Finance on government
securities’ issuance and repayment schedules given the impact on banking system and
monetary policy priorities. The main funding source of the federal budget deficit is the
government securities’ market which has shown swift growth over the last decade. The
reforms conducted by the Ministry of Finance and the Bank of Russia including the
improvement of financial market infrastructure would promote further market development.

Keywords: Fiscal policy, public debt management, government bond markets, emerging
markets.
JEL classification: E62, H63

1 Head of Macroprudential Analysis Division, Financial Stability Department, Bank of Russia.

296 BIS Papers No 67

1. Fiscal policy: mechanisms that promote sustainability and their
influence on monetary policy in Russia
At present, Russia’s budgetary circumstances are quite favourable – in 2011, public debt-to-
GDP ratio stood at about 10.4% of GDP according to the Ministry of Economic Development.
(As of 1 December 2011, the public debt of the Russian Federation amounted to
$163.3 billion.) At the same time, the Russian budgetary system remains exposed to
conditions in the world financial markets and, above all, to oil prices.
In 2011, thanks to the strong price for Urals2 oil, the surplus of the Russian Federation
budget amounted to about $13.4 billion (0.8 % of GDP). High world prices for energy
resources should also contribute to a rather high level of predicted budget revenues in the
coming years. At the same time, growing budget expenditures increase the probability of a
structural deficit in the long run (the Federal Law on the federal budget envisages a budget
deficit for the period 2012–143). On this basis, public debt will reach about $380 billion, some
17% of GDP in 2014. But, despite the large public borrowing programme, the debt burden of
the Russian Federation remains moderate when compared to that of countries with similar
sovereign ratings.
Since 2008, Russia’s Budgetary Code has set out a special mechanism for the utilisation of
oil and gas revenues within the federal budget with a view to reducing the budget’s
dependence on oil and gas revenues, as well as to accumulating reserves in the event that
oil prices retreat. These revenues, the Code stipulates, should be applied to federal budget
expenditures and to the Reserve and the National Wealth Funds.
The Reserve Fund is intended to ensure the financing of the oil and gas transfer4 in the
event of a shortfall in oil and gas budget revenues. It absorbs any surplus in the federal
budget’s oil and gas revenues over and above the volume of the oil and gas transfer and the
Reserve Fund’s investment revenues. The prescribed size of the Reserve Fund is limited to
10% of the forecast GDP for the corresponding fiscal year.
The National Wealth Fund’s purpose is to co-finance the voluntary pension savings of
Russian citizens and to balance the Pension Fund’s budget. The National Wealth Fund
accumulates the oil and gas revenues of the federal budget that exceed the value of the oil
and gas transfer approved for the corresponding fiscal year after the Reserve Fund reaches
its prescribed size. It also accumulates the investment revenues of the National Wealth
Fund.
The Budgetary Code sets a limit on the non-oil-and-gas deficit of the federal budget5
(which should not exceed 4.7% of GDP). The non-oil-and-gas deficit is financed by the oil
and gas transfer (which is restricted to 3.7% of GDP) and from the deficit financing of the
federal budget.
The Reserve Fund was severely depleted in the 2008–09 crisis after the Urals oil price fell
from $140 per barrel to below $40 per barrel. In October 2008, the Reserve Fund had
reached a peak level of $140.98 billion, of which more than $80 billion was used to finance
the budget deficit.

2 In 2011, the average level of oil prices amounted to $110.
3 Base on the following forecast oil prices incorporated in the Federal Budget Law: $93 per barrel in 2012,
$95 per barrel in 2013, $97 per barrel in 2014.
4 Comprising the oil and gas revenues of the federal budget and the Reserve Fund.
5 The non-oil-and-gas deficit of the federal budget represents the difference between the revenues of the
federal budget without oil and gas revenues and the revenues from the management of the Reserve and the
National Wealth Funds, and federal budget expenditures in the corresponding financial year.

BIS Papers No 67 297

Graph 1
Evolution of the Reserve Fund and National Wealth Fund
�
��
��
��
��
���
���
���
���
�
��
���
���
���
���
��
��
��
��
��
��
��
��
��
��
��
��
��
��
��
��
��
��
��
��
��
��
��
��
��
��
��
��
��
��
��
��
��
��
��
��
��
��
��
��
��
��
��
��
��
��
��
��
��
��
��
��
��
��
��
��
��
��
��
��
��
��
��
��
��
��
��
��
��
��
��
��
��
��
��
��
��
��
��
��
�
��
��
��
��
��
���
��
��
��
��
��
��
��
��
��
��
�
��
��
��
���
��
�
���������� �����
����
������������
�����������
������������

Source: Ministry of Finance of Russia; Bloomberg.

In 2010, following heavy anti-crisis spending within the federal budget, it was decided to
suspend the use of the above-mentioned mechanism for oil and gas revenues and use them
to finance the budget deficit.
In 2011, the non-oil-and-gas deficit dropped to 9.7% of GDP (in 2010, it reached 12.6% of
GDP), while oil and gas revenues totalled nearly $174 billion. As a result the 2011 federal
budget generated a surplus of 0.8% of GDP, allowing $31 billion in oil and gas revenues to
be allocated to the planned replenishment of the Reserve Fund.

Table 1
International reserves of the Russian Federation vis-à-vis public debt
Date
Accumulated public
debt, $ bn
Including foreign
debt, $ bn
International
reserves, $ bn
2006 92.4 52.0 303.7
2007 97.9 44.9 478.8
2008 91.6 40.6 426.3
2009 107.0 37.6 439.5
2010 136.9 40.0 479.4
1 December 2011 163.3 35.8 510.9
1 January 2012 n/a n/a 498.6

298 BIS Papers No 67

Russia’s public debt is completely covered by the Federation’s international reserves,6 which
include the Reserve Fund and the National Wealth Fund (see table below). The aggregate
assets of the Reserve and National Wealth Funds amounted to $112.4 billion as of
1 December 2011.
Operations to finance the budget deficit affect the Bank of Russia’s monetary policy because
allocations of debt securities on the primary market lead to liquidity outflows from the banking
sector into general government accounts at the Bank of Russia.
However, this effect is not significant for the following reasons:
1. The volume of government securities is relatively small (at the end of 2011, total
outstandings in the OFZ federal loan bonds market was an estimated $78.4 billion).
2. Most of the borrowed Treasury debt market funds eventually return to the banking
system in the form of budgetary expenditure or through allocation of temporarily
available budgetary funds to bank deposits (the amount of deposits of the general
government in commercial banks totalled approximately $48.8 billion on
1 December 2011).
3. Credit institutions use the purchased securities as collateral for refinancing
operations of the Bank of Russia.
2. The Bank of Russia’s role in public debt management
The Bank of Russia supports the government in realising its fiscal policy. In particular, the
Bank acts as the agent of the Ministry of Finance in conducting operations with domestic
debt securities, taking into account the impact of these operations on the banking system
and its own monetary policy priorities. The Bank of Russia submits recommendations to the
Ministry of Finance, which sets the key direction for debt policy and issues advice on the
issuance, distribution and redemption of government securities on domestic and foreign
markets.
The Bank of Russia does not conduct monetary operations of a quasi-fiscal or non-traditional
nature. The Bank of Russia is prohibited by law7 from financing the federal budget deficit and
from buying government securities in the primary market.
In recent years, the Bank of Russia has not frequently been involved in purchasing
government securities in the secondary market (other than when fulfilling obligations in
repurchase agreements). The Bank of Russia does not make such purchases in order to
influence the yield curve. Rather, short-term lending operations are used for this purpose.
Nor do these operations constitute any part of the central bank’s anti-crisis policies. During
the 1990s, the Bank’s purchases of government bonds in the secondary market were the
main instrument of liquidity provision to commercial banks (indeed, the volume of
government securities held by the Bank of Russia at times amounted to more than twice the
monetary base). However, in 2008–09, it was principally credit operations that were used for
this purpose. Over the same period, the total amount of purchases of OFZ and foreign
currency-denominated government securities by the Bank of Russia amounted to only about

6 The international reserves comprise highly liquid financial assets that are held by the Bank of Russia and the
Russian Government, including foreign exchange reserves and monetary gold, as well as the foreign currency
holdings of the Reserve Fund and the National Welfare Fund.
7 Federal Law no 86-FZ dated 10 July 2002: “On the Central Bank of the Russian Federation (Bank of Russia)”.

BIS Papers No 67 299

$5 billion. At the same time, the Bank’s claims on credit organisations at the peak of crisis
exceeded $120 billion (for instance, as of 1 February 2009).
The Bank of Russia is also empowered to issue its own bonds. These short-term securities
are used to manage excessive liquidity in the banking sector. Between February 2007 and
August 2010, the Bank of Russia regularly issued short-term bonds with a maturity of six
months. After November 2010, the maturity of this issuance was reduced to three months.
Since mid-October 2011, the Bank of Russia has suspended bond issuance due to a
shortage of market liquidity.
3. Main features of the domestic currency public debt market
At present, the Government of Russia borrows mainly by issuing debt securities on the
domestic market. (Domestically issued securities account for up to 78% of internal public
debt in the form of debt securities.)
Before the 2008 crisis (when the government budget was in surplus) domestic public
borrowing, in the form of OFZ, was relatively low (ranging between $5.3 billion and
$7.8 billion annually). Since 2009, however, the budget deficit has been financed mainly via
the domestic public debt market. Over the last decade, Russia’s rouble-denominated public
debt market has increased in volume (doubling over the last two and a half years to
$78.4 billion). At present, federal loan obligations (OFZ) account for a significant share (37%)
of the entire Russian rouble-denominated debt market.
The expansion of the public debt market has contributed to the development of the internal
money market. For instance, repo operations are developing rapidly. As of end-December
2011, public securities accounted for 26% of domestic repo operations (Graph 2).
Government securities (OFZ) are issued with maturities of up to 30 years. However, the most
liquid (benchmark) issues have maturities of up to 10 years. New OFZ issues from the
Ministry of Finance have maturities of 10 years or less. The maturity of public debt
obligations is not increasing; longer-term issues imply higher borrowing costs under current
market conditions.
Despite these positive trends, the Russian public debt market remains underdeveloped. This
is due not only to Russia’s long period in budget surplus, during which there was no need for
large-scale borrowing but also to perceived access problems for international participants in
the Russian market. International investors are ready to take credit risk on rouble-
denominated debt obligations, but they would require high premiums to compensate them for
the risks associated with Russian financial market infrastructure.
A very limited number of state-associated banks and the Pension Fund of the Russian
Federation are the main investors in the public debt market; the share of non-residents in
trading is minimal (Graph 3). Thus, the Russian public debt market lacks institutional
investors with long-term investment strategies.
In this connection, the Ministry of Finance and the Bank of Russia are seeking to liberalise
the public debt market. Until recently, there was only one trading platform for Russian
government bonds – a special section of MICEX. Since January 2012, however, government
securities can also be traded on the securities section of the combined MICEX-RTS
exchange, as well as on the OTC market. The Bank of Russia expects that these reforms will
help broaden the investor base to international participants, as well as improve the market’s
liquidity and capacity.
The Federal Law of 7 December 2011 no 414-FZ “On the Central Depository” provides that,
from 1 July 2012, central depositories active in the Russian market will be able to open
custody accounts for foreign nominee holders of international centralised depositories or
securities settlement systems as well as national depositories, securities settlement systems

300 BIS Papers No 67

and clearing organisations. This provision is expected to increase the attractiveness of the
Russian government securities market to foreign investors by mitigating the risks faced by
foreign investors who use such financial market infrastructures (FMIs) in their jurisdictions. At
present, foreign FMIs can only open accounts at Russian depositories, which exposes
foreign investors to risks in the event of financial problems at foreign FMIs.

Graph 2 Graph 3
�����
�����
�����
������������������������������� ���������������������
������������������� �����������
��������
������� ���������
���������������
�������������
� ��������������

Source: Bank of Russia

As non-residents account for a minimal level of Russian debt holdings at present, the
reorientation of public borrowing towards the domestic market will not expose the market to a
higher degree of risk from external shocks and does not present any additional risk to
financial stability.

BIS Papers No 67 301

Aspects of fiscal/debt management and monetary policy
interaction: the recent experience of Saudi Arabia
Abdulrahman Al-Hamidy1
Abstract
As Saudi Arabia’s oil export revenues constitute about 90% of its budget, its fiscal policy is
largely a function of developments in the oil market. Over the years, a countercyclical fiscal
stance has been used to reduce the volatility of domestic growth against the background of
vacillating oil revenues. Given the structure of the economy, the need is to continue to
encourage the private sector to assume a greater role in the country’s diversification efforts.
Reflecting the dominance of fiscal policy in Saudi Arabia and its impact on economic growth,
the primary aim of monetary policy is to assure exchange rate stability with a view to
providing an environment that is conducive to financial stability and sustainable growth. The
interaction between monetary and fiscal policy has changed in the current global economic
and financial climate. Greater cooperation between the fiscal and monetary authorities has
become indispensable in meeting macroeconomic objectives. It is envisaged that policy
interaction can return to normal modes of operation as the global economic recovery takes
hold.

Keywords: Saudi Arabia; experiences; fiscal; monetary; policy interactions; countercyclical;
exchange rate; stability.
JEL classification: E32, E44, E52, E63, F31

1 Vice Governor, Saudi Arabian Monetary Agency.

302 BIS Papers No 67

1. Introduction
The current euro zone sovereign debt crisis and worsening debt dynamics in major industrial
economies reflect elevated risks in sovereign overindebtedness with adverse consequences
for economic growth in affected countries. In an age of fiscal austerity and overstretched
monetary policy accommodation, the conventional interaction appears to be losing its
effectiveness in reviving the economy.
2. Saudi Arabia’s experience
(i) Relevance of export revenues in the oil sector
As Saudi Arabia is a resource-based economy, with oil export revenues constituting about
90% of its budget, its fiscal policy is largely a function of developments in the oil market. Over
the years, a countercyclical fiscal stance has been used to reduce the volatility of domestic
growth against the background of vacillating oil revenues. With the improvement in the oil
market since 2003, fiscal spending has risen and there have been enlarged fiscal surpluses.
In fact, part of these cumulative budget surpluses has been used to redeem government
debt, which stood at about 10% of GDP by the end of 2010, down from its peak of over 100%
in 1999. As is the case in most emerging market economies, fiscal policy remains dominant
in stimulating private sector growth.

(ii) Policy challenges for the Saudi government
In Saudi Arabia, all oil revenues accrue to the government. While higher oil revenues are
beneficial, they also entail many challenges. The three major and competing considerations
are: (a) cyclical, ie containing inflation, which calls for fiscal restraint; (b) secular, ie ensuring

BIS Papers No 67 303

that oil revenues are distributed through investments in value-added sectors with a focus on
projects that are job-creating and socially valuable; and (c) the preservation of
intergenerational equity with the twin plans of accumulating foreign assets and increasing
economic diversification by investing in physical and social infrastructure. The government
budget prioritises spending in health, education and infrastructure. Putting a high weight on
ensuring intergenerational equity and long-term fiscal sustainability is consistent with short-
run cyclical considerations when inflationary pressure is high (ie consistent with fiscal
restraint). Creating a favourable environment for long-term economic diversification via public
investment is consistent with a more expansionary fiscal policy. During the past decade, the
diversification motive has been particularly strong in Saudi Arabia, as reflected in massive
investment programmes.
(iii) Monetary policy
Saudi Arabia’s monetary regime is effectively its exchange rate regime. SAMA’s monetary
policy objectives are to maintain the dollar/riyal exchange rate within the framework of the
pegged exchange rate regime with a view to pursuing price stability and safeguarding
financial stability.
In a fixed exchange rate regime, interest rate policy is largely influenced by monetary
developments in the anchor currency country. Notwithstanding this limitation, there are other
policy options, such as the application of reserve requirements and prudential guidelines on
bank credit to steer monetary policy in the desired direction.
Higher interest rates and exchange rate appreciation have limited impact on curbing inflation
that is driven by supply shocks such as higher food prices, as consumption of such goods
tends not to be credit-financed. Since demand for food is price-inelastic, higher interest rates
can constrain overall demand but will have little effect on food prices. In the recent past,
Saudi Arabia’s inflation has been largely driven by supply shocks, which cannot be
successfully addressed by higher interest rates. Indeed, rate hikes are more likely to soften
growth than dampen prices. But inflation has moderated recently in line with falling food
prices and softening rent charges.

304 BIS Papers No 67

(iv) The monetary process and causative factors for the money supply
The monetary process in an oil-based economy differs from that in non-oil economies. In
Saudi Arabia, the receipt of oil revenues adds to government deposits with no immediate
impact on domestic liquidity. When the government injects these revenues into the domestic
income stream through its domestic expenditure, the inflow of foreign exchange is translated
into domestic liquidity. Similarly, external government transactions have no impact on
domestic liquidity. It is the private sector’s transactions with the rest of the world that affect
domestic liquidity. Given a relatively limited home production base and an open economy,
the private sector’s payments for imports and other external transactions far exceed its
receipts from abroad. Private sector imports are highly influenced by government spending,
which remains the driving force in the economy. Hence, the government’s net domestic
expenditure and the private sector’s balance of payments deficit control domestic liquidity
and provide the basis for constructing causative factors for changes in broad money M3.
The following table for monetary developments in 2005 and 2006 is quite revealing due to
monetary policy transition during this period. There was a marked acceleration in M3 growth
to 19.3% in 2006 from 11.6% in 2005. SAMA’s prudential measures slowed down the
expansion in bank claims on the private sector (SAR 40.1 billion in 2006 vs SAR 122 billion
in 2005). At the same time, the deficit in the private sector’s balance of payments continued
to grow, owing to the sustained increase in imports (SAR 289.4 billion in 2006 vs
SAR 238.3 billion in 2005). These two factors had the effect of slowing M3 growth, but they
were substantially offset by the government’s increased net domestic expenditure, which
amounted to SAR 270.2 billion in 2006 compared with SAR 159 billion in 2005.
Miscellaneous factors also exerted a larger expansionary influence on broad money in 2006
as compared to 2005. In short, the government’s net domestic expenditure is the major
driving factor behind M3 growth.

Causative factors for changes in broad money (M3)
Billions of riyals
2005 2006
Change in M3 (19.3% in 2006 vs 11.6% in 2005) 57.6 106.9
Causative factors
Net domestic expenditure of the government* 159.0 270.2
Change in bank claims on the private sector 122.0 40.1
Change in bank claims on non-financial public sector enterprises 2.5 3.3
Private sector balance of payments deficit –238.3 –289.4
Other items (net) 12.4 82.7
Total 57.6 106.9
* Domestic expenditure less domestic revenue of the government.
Source: SAMA 43rd Annual Report.

(v) Interaction of fiscal and monetary policy
Fiscal policy is dominant in Saudi Arabia and affects monetary policy through direct and
indirect channels.

BIS Papers No 67 305

An expansionary fiscal policy would warrant a restrictive monetary policy in normal times to
curb inflationary pressures and vice versa (direct channel).
A more recent analysis regarding the latest developments in advanced economies is that
expectations of continuing large budget deficits may trigger a lack of confidence in economic
prospects, warranting an expansionary monetary policy to support the financial system
(indirect channel).
A third area where fiscal and monetary policies come together is the development of financial
markets. On the fiscal side, liquid markets that facilitate deficit funding are crucial for
economic development and growth and they also enable the central bank to conduct market-
based operations.
Before the crisis, a standard measure during times of buoyant economic growth was to
rebalance the macroeconomic policy mix by tightening monetary policy. Following the global
crisis in 2008, both fiscal and monetary policies in Saudi Arabia were set in the same
expansionary direction to support demand and ensure continued growth and prosperity. This
is an example of how the government has been able to pursue a countercyclical policy
against a comfortable cushion, built in good times, in the form of state reserves from
cumulative budget surpluses. SAMA’s accommodative monetary policy was not adopted
primarily to support the banking system, which has shown its resiliency during the crisis. In
fact, the capital adequacy ratio of Saudi banks has averaged 16.5% in the last three years.
Inflation has been contained and credit conditions have remained favourable, allowing SAMA
to keep monetary conditions easy.
(vi) Development of the financial system
The Saudi financial system has developed and grown significantly over the past decade. In
2003, a Capital Market Authority was established under the CMA Law to regulate and
supervise the securities sector. Since 2003, the CMA has licensed many investment
companies, asset management firms, brokers and financial advisors. The emergence of
these firms has contributed to the expansion in market activities.
Regulatory coverage of the financial sector was further broadened to include insurance
through the enactment of the Cooperative Insurance Companies Control Law of 2003, which
entrusted SAMA with the responsibility of supervising Saudi Arabia’s insurance sector. At the
time of the law’s enactment, there was only one licensed company, operating in the country,
namely the National Cooperative Company for Insurance (NCCI). Other participants in the
insurance sector were branches and agencies of foreign companies that were registered with
the Ministry of Commerce. The Insurance Law prescribes that companies wishing to do
insurance business in the Kingdom must obtain a license from the authorities. At the end of
2010, some 34 licensed insurance companies were operating in the Saudi insurance market.
The banking system has also expanded phenomenally. The number of banks operating in
the Kingdom more than doubled from 11 in December 2000 to 23 by September 2011. Of
these banks, 11 are branches of foreign banks and four represent Saudi-foreign joint
ventures.
Under the supervision and guidance of the Saudi Arabian Monetary Agency (SAMA), the
central bank, the Saudi banking system has become a sound, stable and dynamic banking
system. The banks offer a wide range of products and services that can be increasingly
accessed through ATMs, internet and phone banking. Over the decades, the Saudi banking
system has adopted best international practices in corporate governance, risk management,
risk disclosure and transparency. Saudi banks were among the leaders in fully implementing
the Basel I framework in 1992 and the Basel II framework in 2008. The banking system is
now preparing for the implementation of Basel III on the timelines proposed by the Basel
Committee. Over the years, the Saudi banking system has been characterised by strong
capital adequacy, ample liquidity and high levels of loan loss coverage. In fact, SAMA has

306 BIS Papers No 67

encouraged banks to raise capital and provisioning levels on a countercyclical basis. Given a
prudent and conservative supervisory stance, there has been no incidence of a bank failure
in Saudi Arabia and Saudi banks have shown resilience in the face of recent large regional
and global shocks.
In recent years, Saudi banks have remained profitable with the average banking system
ROA and ROE at end-December 2010 improving to 1.9% and 13.2% respectively.
Furthermore, the banks are highly liquid (with a liquid assets-to-deposits ratio of 36%) and
well provisioned against loan losses. In short, Saudi banks are poised to strengthen their
performance in coming years as their risk management practices are set to improve further
as a result of the shift to Basel III.
For its part, SAMA stands ready to ensure that the banking system is endowed with
adequate liquidity to meet the genuine credit needs of the economy. When required, it
provides liquidity to banks through its repo facility, foreign exchange swaps and placement of
deposits on behalf of autonomous government institutions.
Since the establishment of the Capital Market Authority in 2003, there has been a significant
expansion in capital market activity. By the end of December 2010, the number of listed
companies had almost doubled to 148 and market capitalisation had reached about
SAR 1.3 trillion. By the end of 2010, the number of transactions had reached 19.5 million
worth a total of SAR 759 billion. In the asset management sector, the number of investment
funds had reached 243 with total assets under management of SAR 94 billion. There is also
a nascent sukuk and bond market where total issuance since the market’s start-up stands at
SAR 43 billion (as at December 2010).
Also of some importance in the Saudi financial system are the five specialised credit
institutions that have played an important role in providing long-term credit to vital sectors
such as industry, agriculture and real estate. These institutions are funded from loan
repayments and the central government’s budget. Total disbursements by these institutions
since their inception to end-December 2010 amounted to about $106 billion.
There are also many other non-bank financial institutions, including leasing companies,
insurance companies and licensed money changers. These account for only a small share of
total financial system assets but their transaction volumes and quality are on the rise.

Performance indicators of the banking sector
End of period
2000 2005 2010
Number of licensed banks 11 15 23
Total assets (percent of GDP) 65.0 64.7 85.0
Total private sector loans (percentage of private sector GDP) 61 121 154
Risk-weighted capital/assets ratio 21.0 17.8 17.1
Liquid assets/customer deposits 56.0 34.0 35.7
Loans-to-deposits ratio 50.1 67.5 74.0
Leverage ratio 11.2 12.7 14.4

BIS Papers No 67 307

Conclusion
The challenges confronting Saudi Arabia’s fiscal policymakers arise from the finite, volatile
and uncertain nature of the oil revenues on which the state budget depends. Oil revenues
are also influenced by Saudi Arabia’s systemic role as a leading producer in stabilising the oil
market. This of course means that, on an intergenerational basis, Saudi Arabia is swapping
oil revenues for monetary reserves in order to develop its economic base and encourage the
private sector to assume a greater role in the country’s economic diversification efforts.
Given the dominance of fiscal policy in Saudi Arabia and its impact on economic growth,
monetary policy aims primarily at exchange rate stability with a view to providing a monetary
environment that is conducive to financial stability and sustainable growth over time.
The interaction of fiscal and monetary policy has changed in the current global economic and
financial climate. Greater cooperation between fiscal and monetary authorities has become
indispensable in meeting macroeconomic objectives. It is envisaged that policy interaction
can return to normal modes of operation as the global economic recovery takes hold.

BIS Papers No 67 309

Development of the government bond market and
public debt management in Singapore
Monetary Authority of Singapore
Abstract
This paper describes the growth of the Singapore Government Securities (SGS) market. It
elaborates on the balanced budget policy of the Singapore government, explains how SGS
are issued unrelated to fiscal needs and describes how a liquid SGS market is used to
establish a robust government yield curve for the pricing and development of the domestic
corporate debt market. Recent developments in the SGS market, the issuance of central
bank debt and future challenges are included in the paper.

Keywords: National Government Expenditures and Related Policies, National Budget, Deficit
and Debt, National Debt, Debt Management, Macroeconomic Policy, Macroeconomic
Aspects of Public Finance and General Outlook
JEL classification: H5, H6, H63, E6

310 BIS Papers No 67

(A) Singapore Government Securities (SGS)
Overview of the bond market
The Singapore bond market has grown considerably in terms of size, depth and liquidity in
the last decade. As an indication, market capitalization in the government bond (SGS)
market has increased over three-fold from S$43.3 billion in 2000 to S$138.5 billion in 2011
(see Chart 1). The sovereign curve was also extended twice during this period, starting with
a 15-year issuance in 2001 followed by a 20-year issuance in early 2007. As an indication of
the market’s maturity and demand for even longer benchmarks, the curve was further
extended with the introduction of a 30-year issuance in April 2012, marking another critical
milestone in the growth of the government bond market.
Chart 1
Growth of SGS (2000–2011) – in S$ billions
�
��
��
��
��
���
���
���
���� ���� ���� ���� ���� ���� ���� ���� ���� ���� ���� ����
���������
���������

In line with its growth, the SGS has been gradually included in widely followed bond indices
during this period. As an example, Singapore was entered in Citigroup World Government
Bond Index (WGBI) in January 2005, and the SGS is similarly included in the Barclays
Capital Global Aggregate Index and JP Morgan World Government Bond Index, amongst
others.
(B) Singapore’s fiscal strength and motivation for SGS issuance
Strong fiscal position
The Singapore government operates on a balanced budget policy and does not need to
finance its expenditures via the issuance of government bonds. It has enjoyed healthy
budget surpluses over terms of government in the past decades and does not have any
external debt. The main focus of the government’s expenditure is on the delivery of essential
public goods and services to residents, with national security, education, public housing,

BIS Papers No 67 311

health care and economic infrastructure development forming the key areas in the national
budget.
Funding for government expenditure programmes is primarily dependent on Singapore’s tax
policies, which are among the most competitive in the world and are designed to enhance
Singapore’s economic competitiveness and promote long-term economic growth. This
combination of fair tax policies and prudent expenditure programmes has made up the
government’s successful fiscal policy over the years. Embedded within this set of fiscal
policies is the belief that the private sector is the engine of growth, with the role of the
government being to provide a conducive and stable environment for it to thrive. Also, all tax
and expenditure policies are justified on microeconomic grounds and focus on supply-side
issues.
Against this backdrop, the government was able to ensure no build-up of net debt1 while
growing the pool of reserves to a state that allows it to be tapped in a sustainable manner to
supplement the government’s budget needs. As an indication of the government’s fiscal
rectitude, the Protection of Reserves Framework in the Constitution further stipulates that
only reserves accumulated during each term of government can be spent, hence further
ensuring that the strength of the government’s fiscal position is not easily eroded.
The availability of a significant reserve pool also strengthens Singapore’s financial standing,
while giving the central bank the capacity to intervene in the FX market in support of the
Singapore dollar (SGD). This is particularly important for Singapore given that its monetary
policy is centered on the management of the trade-weighted basket of exchange rates of its
major export competitors and sources of imports. Hence, rather than being constrained in
any manner, Singapore’s fiscal strength has actually helped the MAS in the management of
the SGD in monetary policy implementation.
Catalyzing growth of the domestic bond market
SGS are marketable debt instruments of the government of Singapore, comprising
short-term Treasury Bills (T-bills) as well as longer-term SGS bonds. As the fiscal agent of
the Singapore government, the MAS is empowered to undertake the issuance and
management of SGS on its behalf. The issuance of T-bills and SGS bonds is governed by
the Local Treasury Bills Act (LTBA) and the Government Securities Act (GSA), respectively,
with separate debt ceilings set by resolutions in Parliament and approved by the President.
As of April 2012, the authorised borrowing limits (representing amounts that the Minister of
Finance is able to borrow at any point) for T-bills and bonds were S$60 billion and
S$490 billion, respectively. In accordance with the GSA, all proceeds from securities
issuance, and any investment returns derived from the proceeds, are paid into the
Government Securities Fund (GSF). Payments from this fund are limited to the payment of
interest and repayment of principal, and are a statutory obligation. This framework ensures
that the government borrowing is not used to fund the government’s expenditures.
Given the strong fiscal position, the Singapore government is in a unique position where
domestic debt securities are issued unrelated to fiscal needs. In fact, the SGS was originally
issued to satisfy banks’ needs for risk-free assets in their liquid asset portfolios; the focus
subsequently shifted to developing the domestic debt market after the Asian financial crisis.
A focused issuance program was introduced aimed at building large and liquid benchmark
bonds, primarily through larger issuance of new SGS bonds and reopening of existing issues
to enlarge the free float and re-channel liquidity from off-the-run issues to benchmark bonds,

1 The Singapore government has assets well in excess of its liabilities, which includes investments in the
Government of Singapore Investment Corporation and Temasek Holdings.

312 BIS Papers No 67

which helped catalysed the growth of the market. With the improvement of liquidity to the
market, the provision of a robust government yield curve was possible, thus stimulating the
growth of the domestic corporate debt market which was able to price off the benchmark
curve. This is important as growth of the corporate debt market will facilitate greater financial
disintermediation towards direct financing (via the financial markets) instead of indirect
financing (via bank credit), hence providing an avenue of matching savers and those in need
of capital outside the banking system and lowering the borrowing costs for domestic
corporates.
In view of these considerations, Singapore is able to continue enjoying short- and long-term
credit ratings from international credit rating agencies, despite the significant growth in
government debt over a short period of time. The three main international rating agencies
(Moody’s, Standard and Poor’s, and Fitch) continue to accord Singapore the highest, AAA,
credit rating.
(C) Recent development in the SGS market and issuance of central
bank debt
Global capital flows and their impact on SGS yields
Towards the end of 2011, Singapore saw strong capital inflows, similarly to Japan and
Switzerland, largely motivated by the global risk aversion from the uncertainty over the debt
crisis and policy uncertainties in both the US and the EU. As these flows end up largely
invested in highly rated safe haven assets such as the SGS, yields have declined sharply
since the start of 2011 with a bull-flattening bias (see Chart 2). This is also broadly in line
with price movements in global safe haven assets, which highlights sustained confidence in
the resilience of Singapore’s economy and the government’s fiscal discipline.
Chart 2
SGS Benchmark Yields since Jan 2010
2Y
5Y
10Y
15Y
20Y
0
0.5
1
1.5
2
2.5
3
3.5
4
Ja
n-
10
Fe
b-
10
M
ar
-1
0
A
pr
-1
0
M
ay
-1
0
Ju
n-
10
Ju
l-1
0
A
ug
-1
0
S
ep
-1
0
O
ct
-1
0
N
ov
-1
0
D
ec
-1
0
Ja
n-
11
Fe
b-
11
M
ar
-1
1
A
pr
-1
1
M
ay
-1
1
Ju
n-
11
Ju
l-1
1
A
ug
-1
1
S
ep
-1
1
O
ct
-1
1
N
ov
-1
1
D
ec
-1
1
�

Looking ahead, increased volatility in domestic financial markets cannot be ruled out if
economic and financial conditions in the advanced economies become markedly weaker,

BIS Papers No 67 313

resulting in contagion shocks. However, the impact of any sudden capital outflows from the
SGS market would likely be mitigated since the SGS remains one of the few highly rated
sovereign assets and local institutions continue to be dominant investors in the SGS market.
Central bank debt security – MAS bills
In addition, since April 2011 the MAS has issued MAS bills, which are central bank bills for
money market operations aimed primarily at financial institutions to help increase the
availability of high quality liquid assets and manage banking system liquidity. While similar to
the T-bills in many ways, these central bank bills are essentially money market instruments,
with shorter tenors ranging from four weeks to three months.
MAS bills are negotiable, so banks needing liquidity can sell them or pledge them as
collateral in the interbank repo market or enter into repo transactions with the MAS through
the Intraday Liquidity Facility and the Standing Facility. The initial issuance will be kept to
S$20 billion to start, which can be increased based on subsequent sterilization requirements.
The development of MAS bills is particularly apt in the face of the changing regulatory
landscape, which has seen greater demand for government and central bank debt securities
along with a growing banking system and higher liquidity requirements. In this regard, MAS
bills will help to meet the needs of banks in Singapore for more regulatory and liquid assets.
(D) Challenges and key policy considerations
While the government bond market has grown significantly in the last decade, developments
in the recent global environment and regulatory landscape have also highlighted a couple of
areas for review and policy considerations. Firstly, given the current low yield environment,
which is likely to persist, the implication of negative yield auctions for short-dated SGS is a
particular cause for concern, in terms of both its implications and its potential impact on
institutional as well as retail investors. Another area is how the use of the central bank debt
(MAS bills) can be further expanded as an instrument for sterilization requirements over the
use of T-bills, which will alleviate the need for the government to raise more debt in
managing capital inflows. This is likely to be explored gradually in the next few years as the
issuance of MAS bills continue to grow and reach a steady state.

BIS Papers No 67 315

Fiscal policy, public debt management and government
bond markets: issues for central banks
Lesetja Kganyago1
Abstract
To reinforce the long-term sustainability of public finances, the South African National
Treasury has proposed a set of fiscal guidelines informed by three principles, namely: a
counter-cyclical fiscal stance, long-term debt sustainability and inter-generational equity.
Owing to the sound management of the fiscus during the six years of strong economic
growth (2002–2007), budget surpluses and fiscal space were created.
The South African economy entered the 2008–2009 recession with healthy public finances
and comparatively low levels of debt. The issuance of domestic government bonds and
Treasury bills alongside government cash balances remains the government’s primary
source of financing. This reflects the healthy, liquid and deep domestic bond market in South
Africa. Domestic government debt accounted for 90.1 per cent of total gross debt of the
national government in 2010/11. The government’s total gross loan debt (comprising
domestic and foreign debt) increased from R990.6 billion in the 2010/11 fiscal year to an
estimated R1.2 trillion in 2011/12. As a ratio of gross domestic product, the national
government’s total gross loan debt increased from 36.0 per cent to an estimated 40.1 per
cent during this period and is expected to plateau at just over 42 per cent in 2014/15, which
should allay concerns over debt sustainability.
The financing of fiscal deficits recently had no adverse impact on monetary policy, despite
the shift from international to domestic markets for public debt. The build-up of government
deposits with the central bank reflects the government’s funding of foreign exchange
purchases for purposes of foreign reserve accumulation. The growth in the central bank’s
balance sheet over the past six years reflects the accumulation of foreign reserves and was
not due to any sort of quantitative easing policies pursued by the central bank. The central
bank issues its own debt paper. The central bank is a full participant on the National
Treasury’s debt management committee.

Keywords: Sustainability of public finances, government debt-to-GDP ratio, government debt
maturity profile, liquid, deep and transparent government debt markets, debt management
coordination, yield curve for government debt, capital flows into domestic debt markets,
public debt and monetary policy
JEL classification: E52, E62, H63

1 Deputy Governor, South African Reserve Bank.

316 BIS Papers No 67

A. Is monetary policy constrained by unsustainable paths of public
debt?
To reinforce the long-term sustainability of public finances, the South African Minister of
Finance has proposed a set of fiscal guidelines informed by three principles, namely, a
counter-cyclical fiscal stance, long-term debt sustainability and inter-generational equity
i. The fiscal stance
The National Treasury (NT) considers a host of measures when determining the fiscal
stance. These include the budget balance, the primary balance (which is a key driver of
fiscal sustainability), the current balance (to determine how much borrowing is being
undertaken for consumption expenditure), and the cyclically adjusted budget balance (to
account for the effects of the business cycle on revenue). All of these measures are at the
consolidated level, which covers the national and provincial governments, social security
funds, and selected entities, and are projected over a 3-year rolling time frame and updated
bi-annually.
South Africa publishes information on its stock of debt on both a gross and net basis. Net
loan debt consists of total domestic and foreign debt, less the cash balances of the National
Revenue Fund. These cash balances consist of deposits in rand and foreign currency.
Owing to sound management of the fiscus during the preceding six years of strong economic
growth (2002–2007), the economy was doing well, with budget surpluses recorded in the
2006/07 and 2007/08 fiscal years, which helped to create fiscal space. The South African
economy entered the 2008–2009 recession with healthy public finances and comparatively
low levels of debt.
The issuance of domestic government bonds and Treasury bills (TBs) alongside government
cash balances remained the government’s primary source of financing. This reflects the
healthy, liquid and deep domestic bond market of South Africa. Domestic government debt
accounted for 90.1 per cent of total gross loan debt of the national government in 2010/11,
with foreign debt accounting for the balance.
The national government’s total gross loan debt, which comprises domestic and foreign
government debt, increased from R990.6 billion in the 2010/11 fiscal year to an estimated
R1.2 trillion in 2011/12. As a ratio of gross domestic product, the national government’s total
gross loan debt increased from 36.0 per cent to an estimated 40.1 per cent during this
period. The government debt-to-GDP ratio is projected to plateau at just over 42 per cent in
2014/15, which should allay concerns over debt sustainability.
Retirees in South Africa have two main sources of income: the means-tested state old age
pension and private pensions. There is no statutory obligation to contribute to either. In some
firms, participation in a pension fund is a condition of employment, but there is no obligation
for workers to preserve these savings if they lose or change jobs. Millions of South Africans
are unable to save adequately for retirement and rely on the old age pension, even though it
might provide an income well below average career earnings.
South Africans obtain medical care either through the public health system or through
contributory medical schemes. In pursuing a more equitable and effective health system, the
government recognises the complementary role of public and private health services.
Proposals for a national health insurance (NHI) system are currently under review, along with
other elements of a 10-point strategy for revitalisation of health services. Health insurance is
a way of paying in advance for some or all of the costs of health care.
Local government finances are monitored by the Intergovernmental Relations Division of the
NT, and published in the annual Local Government Budgets and Expenditure Review.
Monthly finance reports are published on the NT website.

BIS Papers No 67 317

The NT also projects national gross loan debt and net loan debt (ie gross debt less cash
balances) using a variety of methods, in order to determine whether debt is rising indefinitely
over time. More recently, the NT has begun publishing estimates of provisions, contingent
liabilities, and the overall public sector borrowing requirement. Three-year national debt
projections are published bi-annually, but the NT also considers longer-term projections
when setting the budget.
The NT publishes medium-term estimates of the costs of the health system, the social grants
system and the balances of social security funds. The NT is preparing a long-term fiscal
report, to be released in 2012, which will discuss the long-term pressures on the fiscus of
population changes.
ii. Stock of public sector assets
Central bank assets
At the end of March 2012, foreign assets amounted to approximately 90 per cent of the
South African Reserve Bank’s (Bank) total assets of R457 billion. The NT has funded
approximately 31 per cent of the official gross gold and foreign exchange reserve
accumulation. These purchases were funded out of excess income of the NT and therefore
do not reflect additional government debt. The official foreign reserves are owned by the
Bank, and cannot be seen as off-setting government debt. The Bank has a portfolio of rand-
denominated government bonds amounting to R8.6 billion at the end of March 2012, or
1.9 per cent of total assets. These bonds were not acquired in the primary market, but were
obtained from the NT as settlement of realised currency valuation losses on foreign
exchange holdings and forward contracts. In terms of the Reserve Bank Act, local currency
valuation profits/losses of the official gold and foreign exchange reserves are for the account
of the NT.
State pension funds
State pension funds are funds administered by the NT, Transnet, Telkom and the Post
Office. The value of government bonds held by these funds increased from R191 billion in
the first quarter of 2007 to R221 billion in the fourth quarter of 2011. As a percentage of total
assets, holdings of government bonds declined from 25 per cent to 20 per cent over the
same period. This indicates that the financing of government debt by state pension funds has
declined somewhat since 2007. These funds seemed to have switched from government to
public corporation bonds, as the ratio of the latter to total assets increased by 5 percentage
points. The government has recently provided guarantees on some bonds issued by public
corporations such as Eskom and Sanral. State pension funds have also increased
investment in bonds issued by the private sector.

318 BIS Papers No 67

iii. Financing of fiscal deficits and monetary policy
Normally, the financing of fiscal deficits would not affect monetary policy. However, financing
pressures increased due to higher deficits, and this could have a monetary impact, especially
if the Bank has to issue securities to manage money market liquidity. Government assistance
to the central bank to accumulate foreign reserves would be curtailed as funds would be
applied to more pressing priorities. Deficit financing pressures could compel the government
to broaden funding sources and borrow offshore for use in the domestic market. The
government would be faced with selling such funds to the central bank, but this would have
unintended consequences in that liquidity would be injected into the domestic money market
which would need to be sterilised. In order to overcome this problem, the government could
opt to sell foreign currency in the domestic market, but this would result in the appreciation of
the local currency.
Recently, the government has opted to place foreign exchange deposits with the Bank to
avoid the money market liquidity impact of foreign exchange accumulation.
B. Domestic currency public debt issued in local markets
iv. Shift from international to domestic markets for public debt
The domestic capital markets remain the primary source of funding for the government’s
gross borrowing requirements. The purpose of borrowing in the international capital markets
is to finance the government foreign currency commitments and to establish benchmarks for
local public entities to borrow in the international market. The country’s foreign debt as a
percentage of gross loan debt is estimated to decrease from 9.9 per cent in 2010/11 to
5.9 per cent in 2014/15.
v. Money market development
Short-term borrowing consists of TB issuance. Provinces and some public entities are
required to invest their surplus cash with the Corporation for Public Deposits (CPD), and the
government borrows from the Corporation for its financing activities. The government TB
portfolio has been diversified from 91-day and 182-day bills to also include 273-day and
364-day maturities. The short-term debt portion of the government’s total gross loan debt,

BIS Papers No 67 319

mainly TBs, increased from 6.4 per cent in 2000/2001 to 13.7 per cent in 2010/11. It is
expected to decrease to 13.0 per cent in 2014/15.
In 2010, the money market migrated to the dematerialised Money Market Settlement System
(MMSS) of Strate. Initially only new TB issuances since 26 February 2010 were issued,
cleared and settled electronically. By February 2011, all TBs were dematerialised. Beneficial
ownership is recorded and updated in Strate’s Securities Ownership Register (SOR) and
ownership information is provided to all the issuers.
Government deposits with the Bank increased considerably from levels of around R70 billion
at the end of 2009 to approximately R132 billion at the end of March 2012. These deposits
largely sterilised liquidity injected into the money market, emanating from foreign exchange
transactions by the Bank.
The increase in government deposits with the Bank since 2009 was mainly the result of a
sharp rise in foreign currency-denominated deposits, which increased from R2.9 billion in
December 2009 to R30.9 billion in December 2010 and R67.6 billion in March 2012.
vi. Lengthening maturity of domestic government bonds and developing yield
curves
Holdings of government bonds by the Bank remained broadly on the same level for the past
six years, with generally the same bonds being held in its portfolio.
The NT switched approximately R15 billion of short-term government bonds maturing
between one and two years into longer-term bonds during 2011/12 to manage the
refinancing risk of the government debt portfolio. The accompanying table indicates the
average, original and remaining maturity in years of all national government bonds, weighted
by the outstanding amounts in issue. Further switch auctions are planned for 2012/13.

From the beginning of 2007 to July 2008, bond yields increased in response to inflationary
concerns arising from record-high prices of oil and food. Bond yields then declined sharply
until December 2008, before increasing again up to July 2009 in reaction to, among other
things, the notable increase in the supply of government bonds. From July 2009 to
November 2010, bond yields decreased as a result of the appreciation in the exchange value
of the rand, the reductions in the repurchase rate, the release of better-than-expected
consumer inflation data and strong non-resident demand for domestic bonds, before
fluctuating higher up to the beginning of 2011. After experiencing an inverted yield curve from
the end of 2006, the yield curve normalised in mid-2009 and has since remained positive
sloping.

320 BIS Papers No 67

Except for the short end, which remained anchored to the unchanged repurchase rate, the
level of the yield curve across the rest of the maturity spectrum declined from March 2011 to
September 2011, before increasing up to November 2011 following the depreciation in the
exchange value of the rand. The unusually flat yield curve from the middle area of the curve
to the long end steepened from September 2011 to November, as the issuances of longer-
term bonds were more significant and as a result of the switches. Since September 2011, the
yield curve has moved marginally downwards.
As yields of longer-term bonds increased more pronouncedly than the shorter-term bonds
from September to November 2011, the yield gap, measured as the difference between the
yields at the extreme long and short ends of the curve, widened from 264 basis points on
6 September 2011 to 372 basis points on 23 November, but has narrowed to 334 basis
points on 18 April 2012.

6
7
8
9
10
11
Oct-07 Oct-08 Oct-09 Oct-10 Oct-11
Per cent
Government bond yields
R157 (2015) R209 (2036) R208 (2021)
5
6
7
8
9
10
0 5 10 15 20 25 30
Years
Per cent
Yield curve
2010/11/05 2011/03/11 2011/09/06 2011/11/23 2012/03/31

BIS Papers No 67 321

vii. Financial stability risks related to deeper domestic financial markets
Larger capital inflows, more assets and liquidity to support speculative activity
In the aftermath of the global financial and economic crisis, interventions by authorities to
stabilise financial systems introduced cyclical imbalances that run the risk of becoming
structural in nature. In this way, historically low interest rates and high levels of liquidity in
advanced economies caused investors to borrow in these countries and invest in high-
yielding assets in EMEs (carry trade). Furthermore, EMEs recovered much quicker from the
effects of the global financial and economic crisis, making them attractive destinations for
international investment flows. These imbalances created the risk of excessive capital flows
to emerging markets, increasing the risk of asset-price bubbles followed by collapses in
prices. These risks arise when capital flows are not matched by economies’ ability to absorb
the flows productively.
Despite the relatively lower economic growth rate in the latter part of 2010 compared to the
first half, EMEs remained an important driver of global economic growth and recorded a
7.5 per cent economic growth rate for 2010 as a whole. Countries in developing Asia grew
the most rapidly of all EMEs, reaching an average economic growth rate of 9.7 per cent
during 2010. Economic growth in EMEs is estimated to have slowed to 6.2 per cent in 2011,
as their developed trading partners experienced a significant loss of momentum. In its latest
edition of the World Economic Outlook (WEO),2 the IMF forecasts that EME economic
growth will moderate somewhat further in 2012, before reaccelerating mildly in 2013, to
6.0 per cent. The IMF, however, still points to lingering downside risks. Key risks for EMEs
include a rapid rise of inflation pressures and overheating pressures, partly driven by capital
inflows.
Faster transmission of external shocks
The South African bond and foreign exchange markets are extremely liquid. With no
exchange controls applicable to non-residents, the domestic markets are popular for the
hedging of exposures in other markets. Moreover, these deep and liquid domestic markets
are convenient for the outright selling/buying of foreign exchange and other securities. The
domestic bond market is also a favourable destination for carry trades out of the US, the UK
and Japan. The experiences of 2010 and 2011 (year-to-date), however, demonstrated that
due to the high level of liquidity, the domestic markets, especially the foreign exchange
market, can be extremely volatile, with the exchange rate serving as a key transition channel
of global instability into our market. Recent volatility in the rand was therefore the result of
global instability and negative sentiment which triggered risk on/risk off trading scenarios,
and not really due to domestic circumstances.
The monetary policy transmission mechanism in the domestic market is still seen as being
predominantly via the credit channel, with credit priced off the prime lending rate. The pricing
of credit from both the money and bond market yield curves has grown significantly. The
short end of the bond yield curve is mainly affected by expectations regarding the central
bank’s monetary policy stance, while the longer end is driven mainly by inflation
expectations, explaining the relatively long transmission period between short- and long-term
interest rates.

2 April 2012.

322 BIS Papers No 67

C. Central banks and public debt management
viii. Central bank debt paper
The Bank issues SARB debentures in terms of section 10(1))(i) of the South African Reserve
Bank Act, 1989 (Act No. 90 of 1989). The SARB debentures are issued solely for liquidity
management purposes in the domestic money market, that is, to drain excess liquidity from
the market. They are issued with maturities of 7, 14, 28 and 56 days. The outstanding
amount of SARB debentures in the market was R31.2 billion at the end of November 2011,
and R21.0 billion at the end of March 2012.
The Bank issues SARB debentures in the shorter end of the money market curve while the
government issues in the longer end. Furthermore, TBs are at times issued at interest rates
above the repurchase (repo) rate, whereas debentures are issued at rates at or below the
repo rate. The Bank occasionally encounters difficulties in issuing its own securities.
The NT issues TBs for cash management purposes. These are money market instruments
with maturities of 91, 182, 273 and 364 days. The outstanding amounts were R54.8 billion,
R35.8 billion, R38.0 billion and R31.4 billion, respectively, at the end of November 2011, and
R46.5 billion, R36.3 billion, R38.7 billion and R34.8 billion, respectively, on 13 April 2012.
NT auctions are on different days to those of the SARB debentures. Furthermore, the
NT coordinates the issuance programme of the broader public sector.
ix. Short-term vs long-term public debt
The central bank’s involvement in government debt management is on an agency basis to
conduct auctions for primary issuance of bonds and TBs.
Historically, the government has co-operated with the central bank when raising currency
debt. Initially, proceeds from foreign currency loans were applied against the Bank’s oversold
net open position. Since then, the government has continued to support the Bank’s efforts in
foreign exchange reserve accumulation.
x. Domestic vs foreign currency debt
The current arrangement between the Bank and the NT is that the proceeds of the
government’s offshore borrowing are deposited with the Bank. Furthermore, the NT has
funded a substantial portion of the accumulation of foreign reserves in recent years. The
purchases are funded out of revenue overruns. However, foreign exchange swaps are also
extensively utilised to fund foreign exchange purchases. Outstanding foreign exchange
swaps conducted for this purpose amounted to USD6.6 billion at the end of November 2011
and USD7.0 billion at the end of March 2012. These swaps will also eventually be funded by
the NT.
xi. Central bank balance sheets and by quasi-fiscal operations
The Bank is not involved in quasi-fiscal operations or unconventional monetary policies. The
growth in the size of the Bank’s balance sheet over the past five years was therefore due to
the policy of the monetary authorities to accumulate foreign reserves in order to reduce the
country’s external vulnerabilities, and not to unconventional monetary policies (or quantitative
easing).
xii. Governance arrangement for the coordination of monetary policy and public
debt management
A Memorandum of Understanding (MOU) between the Bank and the government sets out a
framework for a consultative process. This MOU also sets out a framework for the formation

BIS Papers No 67 323

of Standing Committees to oversee macro-economic, banking and financial market, financial
and regulatory, and international relation issues. The Standing Committee on Banking and
Financial Markets contributes, among other things, to the primary objective of debt
management policy of minimising debt costs within acceptable risk levels. The Bank also sits
as a full participant on the Debt Management Committee chaired by the NT.

BIS Papers No 67 325

Fiscal policy and its implication for central banks
Suchada Kirakul1
Abstract
Over the past decade, prudent fiscal management has served Thailand well in cushioning
the impact of the global financial turbulence. However, going forward, fiscal risks which
include a weakened global economic outlook, unbalanced fiscal structure and growing
contingent liabilities may have implications for fiscal debt sustainability in the medium term.
Since the credibility of fiscal policy greatly influences the conduct and effectiveness of
monetary policy, central banks have an incentive to monitor fiscal positions closely. At the
same time, it is important to preserve central bank independence and credibility to ensure
that the central bank can carry out its primary mandate. In relation to this, policy coordination
between the central bank and the government is crucial both in terms of policy stance and
public debt management, which will also help foster bond market development and promote
financial stability.

Keywords: Thailand public debt, fiscal and monetary policy, fiscal risk
JEL classification: E52 E61 E62 H63

1 Bank of Thailand.

326 BIS Papers No 67

1. Introduction
The credibility of fiscal policy greatly influences the conduct and effectiveness of monetary
policy. Monetary policy is more effective when the private sector trusts that the government
will not resort to inflationary deficit financing. Therefore, central banks have an incentive to
monitor fiscal positions closely as: (1) the governments may be tempted to call on central
banks for debt financing, which would then directly damage the central bank’s credibility; and
(2) fiscal policy can have a significant impact on the economy as well as the financial
markets.
Analysis of fiscal policy and public debt sustainability often hinges on the expected path of
public debt. Underlying the projected debt dynamics are factors such as the ability of the
government to raise revenues or limit expenditures, medium-term growth prospects and
market sentiment that may influence the cost and availability of financing. Rising contingent
liabilities, especially those that are less transparent and implicit in nature, may result in a
surprise public debt overshoot. The next section of this country paper will offer a brief review
of Thailand’s public debt development. The paper will then highlight a number of concerns
regarding fiscal policy and public debt, followed by important implications for the central
bank.
2. Country experience: Thailand
2.1 Development of the fiscal position and public debt
Over the past 25 years, Thailand’s public debt path has reflected prudent fiscal management
overall – never exceeding the fiscal sustainability guideline of 60 percent of GDP even during
the 1997 Asian crisis. Below is a short discussion of two notable episodes in the evolution of
Thailand’s public debt.
1. The pre- and post-Asian crisis: 1990 to 2007
As depicted by Figure 1, the public debt-to-GDP ratio was on a declining trend from 1990 up
to the onset of the Asian crisis, bottoming out at 13.5 percent of GDP in 1996 after nine
consecutive years of fiscal surpluses. Then, during 1997–2000, there was a sharp rise in
public debt to a peak of 57.8 percent of GDP in 2000, as a result of the financial sector
bailout during the crisis and the countercyclical role of the public sector to restore the
economy2. Efforts to reduce the public debt level to achieve fiscal consolidation then
followed, and when coupled with the strengthening economy, led to a continuous decline in
ratio of public debt to GDP until 2007.

2 It should be noted that the total amount of public debt already incorporates the outstanding debt of the
Financial Institutions Development Fund (FIDF).

BIS Papers No 67 327

2. Cushioning the impact of the global financial crisis: 2007 onwards
As in many countries, the global financial crisis necessitated a fiscal expansion in Thailand.
Besides running sizable budget deficits during the past five fiscal years (FY2007–2011), the
government also engaged itself in borrowing to mobilize funds for additional investment
projects under the “Strong Thailand” project (FY2009–2012), totaling around 3.5 percent of
GDP. Moreover, with many upcoming government initiatives to maintain the incumbent
government’s popularity, coupled with extraordinary spending to mitigate the negative impact
of the recent flood disaster (e.g. urgent relief expenses to assist adversely affected people,
spending to restore confidence, and the medium- and long-term investment projects in water
management system infrastructure of preliminarily around 3.5 percent of GDP), the
government’s expenditure is likely to expand substantially in the periods ahead. Though
these efforts have been vital to counter a deep and long recession and propel the economy
forward, the fiscal consolidation plan will inevitably be delayed.

At present, the current public debt level is roughly 40 percent of GDP – this figure already
incorporates the FIDF debt totaling 1.14 trillion baht at the end of 2011. This level is
considered manageable and has some room for fiscal deficit; however, the debt level will
likely continue rising in the medium term. Should the global economy be weaker than
expected, fiscal consolidation could be further delayed and the debt-to-GDP figures could be
even higher and closer to the threshold. There are also a few concerns regarding fiscal risks
in the medium term, as discussed in the next sub-section.
����
����
����
����
�
��
��
��
��
��
��
��
19
90
19
91
19
92
19
93
19
94
19
95
19
96
19
97
19
98
19
99
20
00
20
01
20
02
20
03
20
04
20
05
20
06
20
07
20
08
20
09
20
10
20
11
� �������
��������
�����������������������������������
������ ����
�������������������������� �� ��������� �������
��������������������������������
�������������������������������� ������ ����� �����
���
��
�
�
��
19
90
19
91
19
92
19
93
19
94
19
95
19
96
19
97
19
98
19
99
20
00
20
01
20
02
20
03
20
04
20
05
20
06
20
07
20
08
20
09
20
10
20
11
������������
��
��
��
��
19
90
19
91
19
92
19
93
19
94
19
95
19
96
19
97
19
98
19
99
20
00
20
01
20
02
20
03
20
04
20
05
20
06
20
07
20
08
20
09
20
10
20
11
������� �����������
� ������� ������������������������������������
���������������������������������� �������� ������������������

328 BIS Papers No 67

2.2 Fiscal risks in the medium term
1. The weakened global economic outlook will likely weigh on domestic growth
prospects, implying the possibility of a greater need for fiscal stimulus in the
periods ahead. While fiscal stimulus can be quite effective in smoothing economic
cycles, caution must be exercised in ensuring not to over-stretch the fiscal room. In
particular, as we head into a period of heightened uncertainties, some insurance,
i.e. retaining some fiscal buffers for emergency, may be needed, though this seems
difficult to achieve given politicians’ short policy horizon. The key to lessening this
risk is to better align the length of the policy horizons of politicians and institutions
charged with the economic and financial stability mandate.
2. The unbalanced fiscal structure may limit fiscal room, resulting in greater
reliance on debt financing and declining debt service ability. Currently,
Thailand’s fiscal structure is unbalanced in two ways. First, expenditures have been
growing on average twice as fast as revenues during the past five years, owing to a
rapid increase in expenditures associated with social welfare and education, and the
continued expansion of government initiatives. On the other hand, revenue
collection, averaging about 17 percent of GDP, is modest compared to other
emerging countries, owing to a narrow tax base and numerous tax exemptions and
deductions. Second, the expenditure side is fairly skewed towards current spending.
During the past 10 years, capital expenditure has declined from 24 percent to
17 percent of total budget in 2011. Taking into account infrastructural investment
through Public-Private Partnerships (PPPs), the amount is still low. The low public
investment is a concern in terms of the country’s medium-term growth potential that
will come back to affect revenue collection, increase the budget deficit, and further
fuel the uptrend in public debt. Furthermore, with GDP growing at a slower pace
while public debt continues to rise, the debt-to-GDP ratio may edge up further and
risk becoming unsustainable in the medium term. To address the problem of an
unbalanced fiscal structure, there is a need to implement fiscal reforms, especially
tax reform and current expenditure cut-back with a view to ensuring fiscal
sustainability.

3. Growing contingent liabilities may reduce balance sheet transparency and
increase the possibility of a debt surprise that could be particularly negative
for the financial markets and financing costs of the government. While explicit
government guarantees are already included as part of public debt, the number of
government initiatives implemented through government/state owned institutions, for
example Specialized Financial Institutions, has increased in recent years. Though
����������� ��������� ���������
����������� ��� ���
��������� ��������� ��� ����
����������� ��� �����
�� ���������������
���������������������������������������� ����
��� ���
� ���������� ��� ���
� �� �������� ��� ���
��� ����� �� �������������������������������������������������������������������������������
�� ���������������� ������� �������������������������������� �����������������
����������������������� �����������������
����������� ��������� ���������
����������� ��� ���
��������� ��������� ��� ����
����������� ��� �����
�� ���������������
���������������������������������������� ����
��� ���
� ���������� ��� ���
� �� �������� ��� ���
��� ����� �� �������������������������������������������������������������������������������
�� ���������������� ������� �������������������������������� �����������������
����������������������� �����������������
����� ���������� ����������������������������������������
����������������������������������������
������������������������ �������������������������

BIS Papers No 67 329

these activities do not instantaneously create liabilities for the government, in the
event that a systemic risk concern arises from significant losses from such activities,
the government will be called upon to rescue the institutions in order to maintain
confidence. From this perspective, this represents an implicit contingent liability for
the government, and efforts should be made to evaluate its fiscal burden. There are
several ways to reduce this risk, including (1) bringing such government initiatives
on balance sheet; and (2) putting in place a mechanism to ensure effective risk
management of the institutions concerned. There is also a risk of creating market
distortions. While there are reasons to support the government’s role in addressing
market failures, care must be taken to not introduce market inefficiencies, for
example favorable treatment for certain segments, distorted pricing, and monopoly.
To reduce this risk, it is important to strategize the method and timing of exit policies
of some government initiatives, involving the private sector as appropriate.
Taken together, these fiscal risks may have important implications for Thailand’s fiscal debt
sustainability in the medium term, thus deserving serious attention. Moreover, growth in
public debt will also imply a higher financing burden, which during an economic expansion
may lead to the crowding out of private sector access to sources of funds, inhibiting private
sector activities. In addition, with higher indebtedness, the sovereign credit rating may be
affected, resulting in a higher cost of financing. The bottom line is that fiscal adjustments will
still be required in Thailand to mitigate risks from the global economy and to raise the
country’s growth potential. At the same time, efforts must be made to ensure that any stimulus
measures are based on longer-term economic stability considerations with a view to limiting
the risk of fiscal dominance in monetary policy, which is addressed in the final part of this
paper.
3. Implications of fiscal risk for the central bank and challenges
ahead
The fiscal risks highlighted in section 2 and their impact on public debt sustainability will likely
have implications for the central bank in the following ways.
1. Closer coordination between the central bank and the Ministry of Finance on
fiscal and monetary policies is essential. The recent European debt crisis has
underscored the multifaceted nature of the interrelations between fiscal policy,
monetary policy and overall financial stability. Though Thailand’s fiscal position is
still strong and the country is nowhere near on the verge of a debt crisis, addressing
this institutional arrangement issue early on is worthwhile. As the main public
institutions responsible for the country’s macroeconomic policy, coordination
between the central bank and Ministry of Finance is therefore crucial. Information
sharing, cooperation and coordination between the two institutions need to be
ensured in order to put in place sound and coherent macroeconomic policy as well
as to better align the length of the policy horizons of the government and the central
bank so that everyone is on the same page regarding the country’s long-term
prospects and risks. In Thailand, the country’s main economic institutions meet
yearly to discuss the appropriate government budget framework for each fiscal year.
The discussion focuses on the outlook for economic and monetary conditions,
revenue collection, issuance of government bonds and the sustainable level of
public debt, to name but a few. The challenge, however, remains in finding ways to
further strengthen and improve this coordination mechanism to ensure greater
effectiveness and further stimulate open and candid policy discussions.
2. All non-standard measures taken by the central bank must be restricted to
extraordinary circumstances and be temporary, with a clear exit strategy. This
is to ensure that the central bank can carry out its primary mandate of price stability.

330 BIS Papers No 67

It has often been the case that emerging economy central banks are required to
take on a number of development functions, pursuing “quasi-fiscal operations” such
as extending credit to priority industries, e.g. a coordination mechanism between the
central bank and banks to extend credit – soft loans – to flood-affected SMEs. Such
practice entails a risk of impairing monetary policy effectiveness, i.e. interest rate
pass-through, while the problem of market failures (both credit access and
availability) prevail. In this case, the central bank needs to closely gauge the impact
of such activities on the money market and conduct appropriate monetary
operations – a costly action, but nevertheless important to ensure that monetary
policy transmission is not impaired. An additional challenge is how to eventually
phase out this role, despite public expectations and political pressure. The exit plan
should be well thought through and clearly spelled out.
When economic and financial development takes hold, the central bank needs to
return to more normal modes of operation, minimizing the fiscal implications of
monetary policy operation, in order to achieve the medium-term goal of price
stability. While this challenge may be more easily addressed in advanced
economies due to greater clarity in terms of the institutional setup which allows a
clear exit strategy and loss bearing responsibilities to be devised, developing
economies may face a more difficult time due to the likelihood of greater fiscal
dominance on the central bank’s operations. Thus, preserving central bank
independence and credibility is all the more important at times like this.
3. Political pressures on the central bank to monetize the deficit may intensify.
This would complicate the work of the central bank and damage its credibility, which
would ultimately undermine the achievement of its primary objective of price
stability.
4. The roles of public debt management and central bank bond
issuance in fostering bond market development and promoting
financial stability
4.1 The government and the Bank of Thailand have made continuous efforts to
strengthen and deepen the domestic bond market to create a more resilient
financial sector
1. The local bond market has grown by more than double over the past decade.
Financial disintermediation since the 1997 crisis has highlighted the importance of
financial markets as viable alternative sources of funds to bank loans. The domestic
bond market has doubled in size from 33.2 percent of GDP in 2000 to 68.9 percent
of GDP in 2010, while the equity market has more than tripled from 26.0 percent of
GDP to 82.5 percent of GDP. On the other hand, bank loans have declined from
93.2 to 85.0 percent of GDP.

BIS Papers No 67 331

2. The increasing role of domestic bond market as a source of financing has
prompted the need for a deep and liquid bond market. The government bond
market has grown in breadth and depth with regular issuance of benchmark bonds,
a broad investor base, diverse and innovative products, and sound market
institutions and infrastructure. Establishing a substantial enough issuance volume
has contributed to a more reliable yield curve, thus providing a benchmark for the
private sector’s bond issuance and enhancing liquidity in the secondary bond market
trading. Savings bonds, floating-rate bonds, and inflation-linked bonds are issued to
broaden the investor base and provide investors with diversified products as well as
risk-hedging instruments.

3. The BOT has also become an important issuer in the Thai bond market as its
sterilization needs grow. Since 2003, the BOT has issued its own debt paper on a
regular basis as another monetary instrument used to absorb excess liquidity in the
system. As of end-2011, total outstanding BOT bills and bonds stood at
approximately 2.642 trillion baht, compared with 2.627 trillion baht outstanding of
government bonds and treasury bills.
4. The BOT coordinates closely with the Public Debt Management Office (PDMO)
to ensure the best possible outcome for the bond market as a whole. The
presence of two “sovereign” issuers may not be ideal as it could cause market
segmentation and make both securities less liquid. To mitigate this potential
negative impact, the BOT and the PDMO coordinate closely and continually
������ ������������������������������������ �
������������������������ ����� ����
�
��
��
��
��
���
���� ���� ���� ���� ���� ���� ���� ���� ���� ���� ����
���������� �������� ��� �����������
� �������
������ ����� ������������� ��������������������� ����������
Million THB
������������������������

332 BIS Papers No 67

throughout the year, both formally and informally. The PDMO holds quarterly
meetings for the debt management committees, in which BOT representatives take
part, to discuss and plan the issuance calendar. To avoid competing with each
other, the timing and maturities of issuance are carefully designed for both issuers to
complement one another. The government primarily issues longer-term benchmark
bonds (e.g. 5-, 7- and 10-year maturities) with the longest maturity of 50 years, while
the BOT issues its securities with a maximum maturity of 4 years. Their auctions are
also held on different days of the week. In addition, the BOT also facilitated the first
issuance of government inflation-linked bonds (ILBs) by providing technical support
in the calculation of real yields and pricing along with supportive registration system
and helped resolve related regulatory issues to ensure its smooth launch.
4.2 Nevertheless, as cross-border capital flows surge, it is critical to strike a
balance between promoting market development and ensuring financial
stability
1. Foreign participation has enhanced liquidity in the bond market, but large and
volatile capital flows have rendered domestic bond markets susceptible to
sudden price movements and greater disruption. In the early stage of bond
market development, non-resident investors were exempted from withholding tax in
an effort to attract foreign investors into the domestic debt market, in order to
enhance liquidity. However, volatile global market sentiments have heightened
volatility in the domestic bond market and occasionally became the primary factor
affecting the yield curve – both in the short and the long end. In October 2010,
Thailand reintroduced the withholding tax on non-resident investors, in response to
a surge in short-term capital inflows which seemed speculative in nature and caused
excessive exchange rate volatility. Thus, the role of foreign investment in the
domestic debt market needs to be carefully considered in order to reap its benefit
without making the country more vulnerable.

2. The impact of capital flows on bond yields may hinder monetary policy
transmission. From time to time, the movement in the short end of the yield curve
has misaligned with that of the policy rate due to large capital inflows. As bond
yields factor into the pricing behavior of commercial banks and impact retail rates,
such misalignment, in effect, may moderate the degree of interest rate pass-through
�����
�����
�����
�
����
����
����
����
������ � ������ � ������ � ������ � ������ � ������ � ������ �
������������������������������ ����������������������
Million USD
������������������������

BIS Papers No 67 333

from the policy rate to retail rates.3 A large proportion of the change in long-term
yields in Asia over the last decade can be explained by global factors,4 which may
subject the cost of long-term funding (via the bond market) to global factors rather
than domestic financial conditions. This may, consequently, hinder monetary policy
transmission.

3 Pongsaparn R, S Wongwaisiriwat, P Chotewattanakul and S Vimolchalao (2011), “Challenges to Monetary
Operations in a Small Inflation-Targeting Economy: Living with Foreign Exchange Flows”, European Central
Bank Workshop on The post-crisis design of the operational framework for the implementation of monetary
policy, Frankfurt am Main (October 2011).
4 International Monetary Fund (2011), Regional Economic Outlook: Asia and Pacific (Washington, April).

BIS Papers No 67 335

Globalisation of the interaction between
fiscal and monetary policy
0HKPHW�<|U�NR÷OX�DQG�0XVWDID�.ÕOÕQo1 Abstract The interaction between fiscal and monetary policies evolves over time and differs from country to country. In this study, we first present the case of Turkey. During the 1990s, the country’s fiscal deficits and public debt ballooned. Monetary policy was severely constrained by the resulting high-risk outlook for the economy, combined with the underdevelopment of domestic financial markets. In the 2000s, however, a significant fiscal consolidation has allowed fiscal policymakers to move from a procyclical to a countercyclical stance, increasing the effectiveness of monetary policy. In the second part of the paper, we discuss the implications of globalisation for the interaction between fiscal and monetary policy. One possible channel comes from the interplay of the inflation rates, policy rates and real exchange rates between emerging and advanced countries. Structural factors such as differences in consumer baskets and quality measurement error, or convergence processes might lead to higher inflation rates and currency appreciation in emerging countries. It might be desirable to smooth this appreciation and contain excessive exchange rate volatility. In this regard, monetary policy in emerging countries might be constrained by inflation differentials and the low level of policy rates in developed countries. In this case, a possible policy option would be to use fiscal consolidation, a strategy that has been observed in emerging countries over the past decade. Keywords: Fiscal policy, monetary policy, globalisation, real exchange rates JEL classification: E52, E62, F42, F31 1 Central Bank of the Republic of Turkey. 336 BIS Papers No 67 1. Introduction The global financial crisis of 2008 and Europe’s sovereign debt problems have highlighted the importance of the interaction between fiscal and monetary policy. Fiscal imprudence can significantly constrain monetary policymakers by forcing them to take into account additional concerns, such as borrowing spreads. It also weakens the transmission of monetary policy by, for instance, hampering the development of domestic currency markets. In contrast, a sound fiscal policy can help monetary policy both in the short term, by allowing fiscal policy to be conducted in a countercyclical way, and in the long run by strengthening the transmission of monetary policy. A stable fiscal stance improves perceptions of country risk, helps develop the domestic currency markets, and mitigates structural problems such as maturity and currency mismatches. Turkey presents an excellent case study of how fiscal policy, which significantly constrained monetary policy in the 1990s, evolved into a sound framework that supported monetary policy over the business cycle and strengthened the transmission of monetary policy in the 2000s. Turkey’s experience is in line with that of most other emerging countries as presented by Frankel et al (2011). In that paper, the authors show that, over the past decade, many developing countries have moved from a procyclical to a countercyclical fiscal policy and that stronger institutions were partly responsible for this change. In this paper we first analyse in detail the evolution of fiscal policy in Turkey. We show the implications for financial markets of increasing public debt and high budget deficits, the development of domestic currency financing and dollarisation, and the effectiveness of central bank policies. We note that fiscal dominance arising from increasing public debt tended to “crowd out” the availability of financial funds to the private sector, and hampered the development of domestic currency financial markets. As a result, there was a high degree of dollarisation in deposit/credit markets as well as in public debt instruments. This high level of dollarisation, coupled with the partial dependence of fiscal policy on direct central bank advances, significantly constrained the effectiveness of monetary policy. Exchange rate pass-through was very high and the credit channel was weak. As a result, Turkey experienced very high and volatile inflation in the 1990s. After the banking and currency crisis in 2001, Turkey implemented important reforms in the monetary, fiscal and financial areas. A significant fiscal consolidation has been achieved and the public debt–to- GDP ratio has steadily fallen. Monetary policy was granted independence in 2001, and inflation has since been brought down from above 60% to below 10%. Substantial reforms in financial markets have also improved the health of the banking sector. Overall, in the 2000s Turkey has achieved a solid fiscal consolidation, successful disinflation, the development of domestic currency financial markets and de-dollarisation. Emergence of a strong credit channel along with a lower level of exchange rate pass-through has significantly strengthened monetary policy transmission. During the 2008 global financial crisis, the banking sector’s robustness and the strong countercyclical reactions of both fiscal and monetary policies helped the economy to recover quickly from the crisis. At this time, strong fiscal balances contributed to the good risk perceptions of the Turkish economy, and provided a wide operational space for the monetary policy response. In the second part of the paper, we look at the global dimensions of fiscal and monetary policy interactions. We show that there are persistent differentials in inflation rates between advanced and developing countries, ie over the last decade inflation has been consistently higher in developing countries than advanced countries. Some part of this difference might come from measurement issues between advanced and emerging countries, differences in WKH�&3,�EDVNHW�ZHLJKWV�DQG�WKH�FRQYHUJHQFH�SURFHVV�RI�GHYHORSLQJ�FRXQWULHV��$V�<|U�NR÷OX� (2010) demonstrates, measurement issues and basket differences can make developing country inflation look significantly higher than advanced country inflation. This difference puts upward pressure on both policy rates and the real exchange rates of the developing countries. However, faced with the persistent inflation differential and the resulting steady BIS Papers No 67 337 upward pressure on real exchange rates, developing countries might find it optimal to reduce that pressure by consolidating their fiscal balances and applying a tight fiscal policy. For advanced countries, this mechanism works in reverse, in the sense that low inflation along with a depreciating domestic currency helps to loosen their fiscal policy. In this part, we present the relevant data and elaborate on how the global interplay of inflation, policy rates and exchange rates can affect the interaction between fiscal and monetary policy. In the rest of the paper, Section 2 outlines the fiscal and monetary policy interactions experienced in Turkey, Section 3 outlines the global dimensions of this interaction and Section 4 concludes. 2. Evolution of fiscal-monetary policy interaction in Turkey from the 1990s to the 2010s Figure 1 presents the evolution of Turkey’s public debt and budget balance as a ratio to GDP between 1990 and 2010. As the left-hand panel shows, the gross debt-to-GDP ratio almost tripled from 25% in 1990 to 74% in 2001, mostly due to the domestic debt component. In the same period, the consolidated budget deficit moved from 2.3% of GDP in 1990 to 12.4% of GDP in 2001. This unsustainable path for the fiscal variables greatly increased the riskiness of the Turkish economy as observed in sovereign bond spreads of more than 10 percentage points in 2001. During this period, government finances relied mostly on the domestic credit markets and partly on central bank advances, therefore limiting the development of domestic currency financial markets and hence constraining monetary policy. After the 2001 crisis, significant reforms followed in fiscal, monetary and financial policies. During the reform process, budget deficits fell quickly below 3% and debt stock decreased below 40% as shown in Figure 1. This fiscal consolidation helped the development of financial markets and supported the operational framework for monetary policy. Figure 1 Public Sector Debt and Budget Balance in Turkey Source: Central Bank of the Republic of Turkey. 338 BIS Papers No 67 2.1. Fiscal dominance and constrained monetary policy in the 1990s Turkey’s deteriorating fiscal position during the 1990s was financed mainly by the domestic banking sector. If we consider the government and the private sector as the two main users of banking assets, we see in the right-hand panel of Figure 2 that the government’s share of banking assets increased significantly. By mid-1996, the ratio of public sector claims to private sector credit in banking assets was around 0.45 and this ratio increased to 2.4 by mid-2002. This unprecedented increase in the government use of banking funds was a real challenge for the development of domestic financial markets in that heavy public borrowing can easily crowd out private activity in the banking sector. This moves banks away from their traditional role of credit supplier to the private sector. As a result, private agents may respond by moving to foreign currency funding, as was the case in Turkey. Another dimension of high public debt levels is that some part of this debt was financed by central bank advances. As seen in the left-hand panel of Figure 2, central bank advances increased to around 20% of domestic debt stock in 1993 before going to zero in 1998. The side effects of fiscal deterioration showed themselves in the low level of financial development, a high degree of dollarisation, high exchange rate pass-through, and the restricted effectiveness of monetary policy. The outcome was high and volatile inflation in the 1990s. Given the low level of financial development in the 1990s (the private credit-to-GDP ratio fluctuated between 15% and 25% from 1990 to 2002), the heavy dependence of the public sector on domestic financial markets created a high level of dollarisation in both public finances and private transactions. The FX share in gross public debt, as shown in the left- hand panel of Figure 3, was close to 60% by the beginning of 2003; and the dollarisation levels for deposits and credit were around 50% from mid-1996 until end-2002, as shown in the right-hand panel of Figure 3. This high level of dollarisation manifested itself as a high level of exchange rate pass-through to domeVWLF�SULFHV��$V�HVWLPDWHG�E\�.DUD�DQG�g÷�Qo� (2008), the cumulative pass-through to core consumer price inflation in nine months was 48% in the February 1995–April 2001 period (Figure 4, left-hand panel). Figure 2 Central bank advances to government and fiscal dominance in financial markets in Turkey Source: Central Bank of the Republic of Turkey. Monetary policy was significantly constrained, and in several dimensions, by the resulting fiscal dominance in financial markets, the low level of financial development and the high level of exchange rate pass-through. The underdevelopment of domestic credit markets severely hampered the effectiveness of the credit channel in demand management. Dollarisation also played a part in undermining the effectiveness of this channel. Moreover, BIS Papers No 67 339 given the high level of dollarisation, exchange rate movements had a strong effect on inflation, and thus became the main determinant of domestic price developments. Exchange rates were largely determined by risk perceptions that were conditioned by public policy in particular and the Turkish economy in general, with the result that monetary policy had to respond mainly to changes in risk appetite and the resulting exchange rate fluctuations. Thus, monetary policy was focused mainly on containing the negative effects of underdeveloped credit markets and high currency mismatches in the domestic economy rather than on directly managing aggregate demand and containing inflation. Turkey’s twin banking and currency crisis in 2001 demonstrated how monetary policy can be tightly constrained by fiscal policy. During the crisis, public debt and the budget deficit reached record levels, aggravating risk concerns for the economy, in a similar way to that seen in the current European debt crisis. EMBI spreads for Turkey increased above 10 percentage points, and the economy experienced an output loss of more than 5%. At the same time, the Turkish lira depreciated significantly. In contrast to a conventional monetary policy during a crisis (similar to the one in developed countries where monetary policy reacts in countercyclical way by decreasing the policy rates to support the economy), monetary policy in Turkey reacted in a procyclical way and the monetary stance was tightened in 2001. This policy action was taken mainly to contain the adverse balance sheet effects coming from the large depreciation of the Turkish lira. In this way, monetary policy was significantly constrained by the imprudent fiscal policy and its consequences such as underdeveloped financial markets and a high degree of currency mismatch in the economy. In addition, fiscal policy had to tighten significantly after the crisis to improve risk perceptions towards the country. Turkey’s experience is comparable with that of some heavily indebted European countries in 2011. Continued budget deficits drove public debt to very high levels, worsening risk perceptions and raising concerns about fiscal sustainability. This led to sharp increases in risk spreads. To contain this deterioration in risk perceptions, some European countries resorted to tight fiscal policies even before the end of the 2008–09 global financial crisis. In the process, monetary policy was severely constrained either by the zero lower bound or by the risk concerns that dominated financial markets. Figure 3 Dollarisation in public debt and in financial markets in Turkey Source: Central Bank of the Republic of Turkey. 340 BIS Papers No 67 2.2. Reforms, fiscal consolidation and independent monetary policy in the 2000s After the 2001 crisis, Turkey implemented significant reforms in fiscal, monetary and financial policies. On the fiscal side, significant fiscal consolidation has resulted in lower levels of public debt and smaller budget deficits as seen in Figure 1. On the financial side, there has been more prudent regulation and tighter supervision of the banking sector, firms and households. The net foreign asset positions of banks have been curbed and foreign currency borrowing by firms and households has been regulated. On the monetary policy side, central bank advances were reduced in 1998, and the central bank was granted independence in 2001. Figure 4 Cumulative exchange rate pass-through to core CPI inflation (in months): VAR evidence Source: Kara, H. and F. Öðünç (2008). The improved fiscal stance has steadily reduced the degree of fiscal dominance in financial markets. As shown in the right-hand panel of Figure 2, the ratio of public sector claims to private sector credit in banking assets fell from 2.4 in 2002 to 0.6 in the end of 2010. This development was accompanied by the rapid development of financial markets, where the private sector credit-to-GDP ratio rose from around 15% in 2001 to around 50% in 2010. In the meantime, the currency decomposition of government finance has also changed significantly. As seen in the left-hand panel of Figure 3, the share of foreign currency in gross public debt fell from 57.8% in the beginning of 2003 to 26.7% by the end of 2010. Both the reduced fiscal dominance in financial markets and the falling share of foreign currency in public debt instruments have helped the development of domestic currency financial markets, thereby strengthening monetary policy transmission. Financial market reforms have also contributed to the sound development of the banking sector. The banking sector’s net foreign asset position of the banking sector has been curbed. Meanwhile, corporate foreign currency borrowing has been restricted to exporting firms and foreign currency borrowing by households has been prohibited. So that currency mismatch risks can be better hedged, the development of hedging techniques has been encouraged in the financial markets. As a result of these reforms, the level of dollarisation in credit and deposits fell from about 50% in 2000 to below 30% at the end of 2010 (Figure 3, right-hand panel). In addition to improvements in the structural factors underlying fiscal variables, fiscal policymaking has also improved. In Table 1, we see cyclicality as measured by the contemporaneous correlation with GDP for the main government variables. Before 2001, total government expenditures were procyclical with a correlation coefficient of 0.19 and they became countercyclical after 2001 with a correlation coefficient of –0.38. Regarding the BIS Papers No 67 341 composition, government consumption as a ratio to GDP was countercyclical but government investment with all its subcomponents was strongly procyclical. After 2001, consumption has become more countercyclical, and all components of government investment except machinery investment have changed from procyclical to countercyclical. This movement of fiscal policy towards a more countercyclical stance has been also observed in other emerging countries as shown by Frankel et al (2011). A procyclical fiscal policy works against monetary policy. During a demand boom with inflationary pressure, monetary policy would ideally tighten so as to curb demand and ward off inflationary pressure. However, a procyclical fiscal policy puts extra demand pressure on the economy and reduces the effectiveness of monetary policy. Therefore, a change in fiscal policy from a procyclical to countercyclical stance in Turkey and in most emerging countries has been an important change in the interaction of monetary and fiscal policies. Table 1 Cyclicality of public policy : contemporaneous correlations with GDP 1987q1– 2001q4 2002q1– 2007q3 Government Consumption and Investment over GDP 0.19 –0.38 Government Consumption over GDP –0.48 –0.57 Government Consumption-Wages over GDP –0.92 –0.90 Government Consumption-Other over GDP 0.09 –0.40 Government Investment over GDP 0.34 –0.20 Government Machinery Investment over GDP 0.41 0.20 Government Construction (Building) Investment over GDP 0.27 –0.39 Government Construction (Other Building) Investment over GDP 0.10 –0.10 Source: Central Bank of the Republic of Turkey. All series are seasonally adjusted and HP-filtered. Statistics are for cyclical components. Important monetary policy reforms were also made in the 2000s. The central bank was made independent in 2001 and price stability was defined as the Bank’s main responsibility. As for the exchange rate, a market-based flexible exchange rate policy was adopted to support monetary policy. In 2006, the central bank started implementing a fully fledged inflation targeting policy thereby increasing the transparency and predictability of monetary policy. Favourable fiscal consolidation and the development of the domestic currency credit markets along with policy independence significantly widened the operational space for monetary policy and improved its effectiveness. As seen in the right-hand panel of Figure 4, the exchange rate pass-through to core consumer price inflation in first nine months fell to 21% in the May 2001–September 2004 period. Thanks to a stronger credit channel and the easing of concerns about fiscal risks, monetary policy has been able to focus mainly on the Bank’s main responsibility of price stability. As a result, a significant disinflation was achieved during the early 2000s. As shown in Figure 5, inflation fell from above 60% in mid-2002 to below 10% at end-2004 and stayed very stable afterwards. A similar improvement was also observed in the volatility of inflation. Moreover, the lengthening maturity of domestic public debt allowed healthier yield curves to develop, together with a stronger transmission mechanism for monetary policy. 342 BIS Papers No 67 Figure 5 Inflation and volatility in Turkey Source: Central Bank of the Republic of Turkey. The global financial crisis of 2008 demonstrated, in a Turkish context, the benefits of a sound fiscal stance and an independent monetary policy. In contrast to the 2001 Turkish crisis, in which economic policies were constrained by heavy public debt and high levels of dollarisation both fiscal and monetary policies responded during the 2008 crisis in a countercyclical fashion to support the economy. Benefiting from the strong stance of fiscal policy and the relatively low level of dollarisation, monetary policymakers were able to ease significantly by cutting policy rates by more than 10 percentage points. These measures were instrumental in the fast recovery of the Turkish economy from the crisis. In the meantime, strong GDP growth and Turkey’s favourable fiscal position led to a surge of volatile short-term capital flows into the Turkish economy, as seen in other emerging countries. During the crisis, an unconstrained and independent monetary policy also proved very useful in countering the ill-effects of financial volatility on the economy. The central bank has devised a new policy mix consisting of an interest rate corridor and reserve requirement ratios as the main tools. These policies have proved to be very useful in containing excessive exchange rate movements and in moderating domestic credit growth. This experience has also underlined the importance of an unconstrained monetary policy in effectively supporting the macrofinancial environment. 3. The global dimensions of fiscal-monetary policy interaction Economic activity has become substantially globalised in the last two decades. During this period, the mutual integration of the advanced and developing countries has proceeded rapidly and, as a result, the policy actions of one group of countries exert a considerable effect on those of the other. One prime example of such policy spillovers was the quantitative easing policies of advanced countries during and after the global financial crisis of 2008. The resulting surge of short-term capital flows into developing countries has confronted their policymakers with substantial macroeconomic challenges. The globalisation process also has possible implications for the interaction between fiscal and monetary policies. One such channel might come from the sustained differences in BIS Papers No 67 343 certain macro variables between advanced and emerging countries. A sustained difference in inflation and policy rates deriving from structural factors in the two country groups would have implications for real exchange rates with the result that the fiscal stance of advanced and developing countries might endogenously differ from each other as observed in the data. 3.1. The inflation differential between advanced and developing countries The last decade has witnessed a persistent difference of inflation between advanced and emerging countries. As shown in the left-hand panel of Figure 6, consumer price inflation averaged around 6% in developing countries and 2% in advanced countries during the 2000s. Figure 6 Inflation and policy rate differentials between advanced and developing countries Source: IMF. <|U�NR÷OX��������GLVFXVVHV�VHYHUDO�VWUXFWXUDO�IDFWRUV�WKDW�PLJKW�EH�SDUWLDOO\�UHVSRQVLEOH�IRU� the difference in inflation rates. One factor is the differing weights of inflation baskets. The weight of food and energy items is larger and the weight of technology items is smaller in the inflation baskets of developing countries than in those of advanced countries. Food and energy goods usually have high demand elasticity and low supply elasticity. So, during sustained growth periods, the price of food and energy goods increases substantially, as observed in the 2000s. For technological goods, inflationary pressures tend to be dampened by rapid improvements in technology. Therefore, even if inflation trends at the sector-based level are very similar for different countries, differences in the weightings can cause persistently higher inflation rates in developing countries. To conceptualise the argument, assume that price indices in advanced (A) and emerging (E) countries consist of food plus energy prices (FE) and other goods prices (G) with geometric weights, ie ( ) ( )a a-= 1, ,i ii FE Gt i t i tP P P , i=A, E. For simplicity, assume that prices of FE and G are determined worldwide and they are the same in all countries. Then, the inflation differential between countries would be ( ) ( )p p a a p p- = - -E A FE Gt t E A t t where p t is the inflation at time t. This expression implies that, even though inflation rates at the sector-based levels of FE and G are the same, as long as the food and energy (FE) 344 BIS Papers No 67 goods inflation rates are higher than those of other goods ( )p p>FE Gt t and the weight of FE is
higher in emerging countries ( )a a>E A , then emerging country inflation would be higher than
in advanced countries ( )p p>E At t
Another factor underlying the inflation differential is the measurement bias in inflation. As
RXWOLQHG� E\� <|U�NR÷OX� �������� WKUHH� SRVVLEOH� ELDVHV� DIIHFW� WKH� PHDVXUHPHQW� RI� LQIODWLRQ�� namely the quality bias, the new goods bias, and the outlet substitution bias. All these biases are expected to be larger in developing countries. For example, Bils and Klenow (2001) estimate the quality bias for the United States to be 2.2% per year. The evidence for developing countries is scarcer, but Filho and Chamon (2008) for Mexico and Brazil, and $UVODQ�DQG�&HULWR÷OX��������IRU�7XUNH\�HVWLPDWH�WKH�TXDOLW\�PHDVXUHPHQW�ELDV�DV�DURXQG���� per year. In the process of technological catch-up, convergence and the urbanisation of developing countries, the measurement biases in inflation are expected to be higher in these countries. These structural factors may also help to explain why the inflation targets of inflation targeting developing countries are higher than those of inflation targeting advanced countries. 3.2. Implications for real exchange rates, monetary policy and fiscal policy One implication of the persistent inflation differential between advanced and developing countries is the sustained appreciation of real exchange rates.2 As the left-hand panel of Figure 7 shows, in the 2000s the real exchange rates of developing countries have steadily appreciated, a process that has been only occasionally interrupted by crises. In contrast, advanced country real exchange rates have been mostly stable in this period. Several reasons might account for the appreciation of real exchange rates in developing countries. One of the more fundamental mechanisms would be the so-called Balassa- Samuelson effect arising from the differences in technological growth between advanced and developing countries. If, in the process of convergence, technological growth happens to be larger in developing countries, then prices increase faster in developing countries and, as a result, the real exchange rate appreciates. But there might be other reasons for this appreciation. As discussed above, biases coming from the measurement of inflation would also put an appreciation pressure on the measured real exchange rates of developing countries. Also, when we look at the period between 2003 and 2008, we see that the real exchange rates of developing countries appreciated by around 30%. This period was a period of abundant global liquidity and significant capital flows to developing countries. Overall, some fundamental long-term factors and some short-term factors underlay the appreciation of developing country real exchange rates in the 2000s. The fundamental factors that drive currency appreciation are the result of deep economic forces. Thus, policies intended to constrain the resulting appreciation would be neither effective in the long run nor in the interests of society. However, temporary factors might increase the volatility of real exchange rates and accelerate their appreciation. As real exchange rates are also an asset price, they at times react strongly to news and expectations of future variables. Such fast and volatile exchange rate movements stand in contrast to real variables that adjust slowly in response to price signals. Real exchange rates are significant as relative prices, and sound signals deriving from real exchange rate movements are vital for the proper adjustment of the economy. 2 In this period, nominal exchange rates have been largely stable, implying that the inflation rate differential has been transformed into real exchange rate appreciation for developing countries. BIS Papers No 67 345 Policymakers face the task of containing the adverse effects of volatile and rapidly appreciating real exchange rates. Recent research also supports the idea that policies which lean against real exchange rate misalignments are welfare-improving, as shown by Corsetti et al (2011). The question is how to design the optimal policy framework. Monetary policy and macroprudential policies could be seen as the usual policy fronts against misalignments. But for monetary policy there are some possible tradeoffs in an open economy when containing the effects of exchange rate appreciation. Usually, exchange rate appreciations are associated with capital inflows to the economy and strong domestic growth. If policy rates are lowered to slow the appreciation of the domestic currency, then monetary policy might be too loose to be appropriate for domestic inflation, output and credit conditions. Since there is already a differential between the inflation rates of advanced and developing countries, there is an implied differential for policy rates also. Therefore, monetary policy in developing countries can be thought as being restricted by the inflation differentials and the policy rates of advanced countries. Nevertheless, some innovative ways of using monetary policy remain in such circumstances. For example, it is possible to reduce the lower bound of the interest rate corridor as was done in Turkey during the first half of 2011, thereby increasing the volatility of interest rates with a view to deterring very short-term or carry trade capital flows (Central Bank of the Republic of Turkey (2011)). Figure 7 Real exchange rates and public debt in advanced and developing countries Source: IMF. Fiscal policy is another possible tool for addressing the related problems of rapidly appreciating real exchange rates and their volatility. Fiscal policy cannot usually respond as quickly as monetary policy, but a change in its direction can exert a substantial effect on real exchange rates. One way of actively using fiscal policy in response to external factors would be fiscal tightening or fiscal consolidation. Over time, the fiscal authority might reduce debt levels by cutting budget deficits. Lower budget deficits and public debt imply lower public demand for domestic output, as explained in Frankel and Razin (1992). Some part of this lower demand would fall on the non-tradable part of output and, as a result, the real exchange rate would come under downward pressure. As a measure of fiscal consolidation in developing countries, we can look at the ratio of public debt as a percent of GDP in the right-hand panel of Figure 7. This ratio fell from 52% in 2002 to 32% in 2008 for developing countries. This 20 percentage-point reduction in the gross–debt-to-GDP ratio means a significant fiscal consolidation for these developing countries, with possible effects on external factors such as the net foreign asset position and real exchange rates. 346 BIS Papers No 67 There is considerable theoretical and empirical literature about the effects of fiscal policy on real exchange rates.3 On the theoretical side, both real business cycle and new Keynesian models predict that, in response to an expansionary fiscal policy shock, the real exchange rate will appreciate (Monacelli and Perotti (2010); Ravn et al (2007)). In open economy models, a rise in government spending erodes household net wealth and, as a result, consumption falls. Because of the strong risk-sharing in these models, the real exchange rate appreciates in support of the fall in consumption. On the empirical side, the evidence is rather mixed. A number of studies find that expansionary fiscal shocks lead to depreciation (Kim and Roubini (2008); Monacelli and Perotti (2010); Ravn et al (2007)), whereas other studies report an appreciation of the real exchange rate (Penati (1986); Beetsma et al (2008); Benetrix and Lane (2009)).4 Evidence for the effects of fiscal policy on real exchange rates in developing countries is scarcer. For example, Agenor et al (1997) show with a structural VAR analysis that in Turkey an increase in government expenditure over GDP leads to a real appreciation of the Turkish lira. So a fiscal consolidation could also lead to a depreciation of the currency. This result is consistent with the predictions of the standard models. One crucial element in these studies is whether these fiscal policy shocks are anticipated. Usually fiscal policy actions entail legislative and implementation lags, so that there is a news effect when a fiscal action is first announced. Ramey (2011) studies in detail the importance of the timing of government expenditure shocks for the United States and notes that it is crucial to control for the timing to get reliable results.5 This timing or anticipation explanation is also important for developing countries. For example, in Turkey, the government publishes three-year projections of fiscal variables such as revenues, expenditure, primary and general budget balances and the level of public debt. Consistent with the theoretical and empirical literature, a policy of announcing and implementing a fiscal consolidation in the coming years would put a downward pressure on the domestic currency. Therefore the strong fiscal consolidation in developing countries during the 2002–08 period (Figure 7, right-hand panel) can be seen both as a way of strengthening the fundamentals in these countries as well as the endogenous response of policymakers to the large appreciation of their currencies.6 Overall, the structural differences between advanced and developing countries can create persistently higher inflation in developing countries than in developed countries. This difference can put a strong upward pressure on the currencies of developing countries. Using monetary policy to ameliorate this pressure would create tradeoffs because, in an open economy, the monetary policy of a developing country might be constrained by inflation differentials, developed country policy rates and financial stability issues. As an alternative policy aimed at relieving upward pressure on the exchange rate, developing countries can resort to fiscal consolidation as 3 Hebous (2011) reviews the related theoretical and empirical literature about the effects of discretionary fiscal policy on macroeconomic variables including real exchange rates. 4 This evidence is mostly for advanced countries and the evidence for developing countries is very scarce owing to limited data. 5 This paper shows that, in the United States, once timing is taken into account, consumption of services increases and that of all other consumption items decreases in response to fiscal shocks. If we take these as relative demand changes for services and other goods, one might expect that prices of services (a proxy for non-tradable goods) would increase relative to prices of other goods (a proxy for tradable goods) and then the real exchange rate would appreciate. 6 Another related motive might be to restrict the current account deficit in developing countries by fiscal consolidation. Usually, high appreciation periods are associated with strong growth and a rising current account deficit in developing countries. Then fiscal consolidation can be seen as a way of both reducing the appreciation pressure and containing the current account deficit (see, for example, Kim and Roubini (2008) and Kumhof and Laxton (2009)). BIS Papers No 67 ��� VHHQ�LQ�WKH�����V��)LJXUHV���DQG�����7KLV�LQWHUSOD\�RI�LQIODWLRQ��UHDO�H[FKDQJH�UDWHV��LQWHUHVW� UDWHV� DQG� ILVFDO� SROLF\� VKRZV� KRZ� ILVFDO-PRQHWDU\� SROLF\� LQWHUDFWLRQV� PLJKW� KDYH� EHHQ� JOREDOLVHG�� 3.3. Fiscal policy and real exchange rates in Turkey :H�XQGHUWDNH�D�VLPSOH�HPSLULFDO�H[HUFLVH�WR�WUDFN�WKH�HIIHFWV�RI�JRYHUQPHQW�H[SHQGLWXUHV�RQ� reaO�H[FKDQJH�UDWHV�LQ�7XUNH\��,Q�RXU�PHWKRGRORJ\��ZH�FORVHO\�IROORZ�5DYQ�HW�DO��������DQG� HVWLPDWH�D�YHFWRU�DXWRUHJUHVVLRQ�PRGHO�RI�WKH�IRUP� l l � n n � � l l � n m H � � � � � ª º ª º « » « » « » « » « » « » �« » « » « » « » « » « » « » « » « » « »¬ ¼ ¬ ¼ 1 1 1 1 1 t t t t t t t t t tt g g y y A c B L c nxy nxy rer rer ZKHUH� l tg LV�JRYHUQPHQW�H[SHQGLWXUH��RU�LWV�VXEFRPSRQHQW��RYHU�*'3�� l ty LV�UHDO�*'3�� � tc is SULYDWH� FRQVXPSWLRQ� RYHU� *'3�� n tnxy LV� WKH� WUDGH� EDODQFH� RYHU� *'3� DQG� m trer is the FRQVXPHU�SULFH�LQGH[-EDVHG�UHDO�H[FKDQJH�UDWH��$OO�YDULDEOHV�DUH�IURP�WKH�&HQWUDO�%DQN�RI� WKH�5HSXEOLF�RI�7XUNH\��:H�ILUVW�VHDVRQDOO\�DGMXVW�DOO�WKH�VHULHV�DQG�WKHQ�DSSO\�DQ�+3�ILOWHU�WR� JHW�WKH�F\FOLFDO�FRPSRQHQWV��)RU�*'3�DQG�WKH�UHDO�H[FKDQJH�UDWH��ZH�WDNH�WKH�SHUFHQWDJH� GHYLDWLRQ�IURP�WUHQG�DV�WKH�F\FOLFDO�FRPSRQHQW�DQG�IRU�RWKHU�YDULDEOHV�ZH�WDNH�WKH�DEVROXWH� GHYLDWLRQ�IURP�WUHQG�DV�WKH�F\FOLFDO�FRPSRQHQW��EHFDXVH�WKH\�DUH�DOUHDG\�LQ�WKH�UDWLR�WR�*'3� IRUP���+HUH�DQ�LQFUHDVH�LQ�WKH�UHDO�H[FKDQJH�UDWH�LV�GHILQHG�DV�DQ�DSSUHFLDWLRQ�RI�GRPHVWLF� FXUUHQF\��Ht LV�WKH�YHFWRU�RI�GLVWXUEDQFHV�DQG�L LV�WKH�ODJ�RSHUDWRU�� :H�HVWLPDWH�WKH�YHFWRU�DXWRUHJUHVVLRQ�PRGHO�ZLWK�TXDUWHUO\�GDWD�IURP��4������WR��4������� )RU�LGHQWLILFDWLRQ��ZH�IROORZ�D�&KROHVN\�GHFRPSRVLWLRQ�VXFK�WKDW�JRYHUQPHQW�H[SHQGLWXUHV� RQO\� UHVSRQG� WR� LWV� RZQ� LQQRYDWLRQV� LQ� WKH� VDPH� TXDUWHU�� )LJXUH� �� SUHVHQWV� WKH� LPSXOVH� UHVSRQVH�IXQFWLRQ�RI�UHDO�H[FKDQJH�UDWHV�WR�RQH�VWDQGDUG�GHYLDWLRQ�VKRFNV�LQ�JRYHUQPHQW� H[SHQGLWXUHV�� :H� SUHVHQW� DOO� LPSXOVH� UHVSRQVHV� IRU� WKH� VXEFRPSRQHQWV� RI� JRYHUQPHQW� H[SHQGLWXUHV��:H�VHH�WKDW�LQ�QR�FDVH�GRHV�DQ�LQFUHDVH�LQ�JRYHUQPHQW�H[SHQGLWXUH�OHDG�WR�D� GHSUHFLDWLRQ�� 0RUHRYHU�� LQ� WKH� FDVH� RI� WRWDO� JRYHUQPHQW� H[SHQGLWXUH�� JRYHUQPHQW� FRQVXPSWLRQ� H[SHQGLWXUH� RWKHU� WKDQ� ZDJHV�� JRYHUQPHQW� LQYHVWPHQW� DQG� JRYHUQPHQW� LQYHVWPHQW�RI�PDFKLQHU\��ZH�VHH�WKDW�D�VSHQGLQJ�LQFUHDVH�OHDGV�WR�D�VLJQLILFDQW�DSSUHFLDWLRQ� LQ�WKH�FXUUHQF\��� :LWK� WKHVH� UHVXOWV�� ZH� VHH� WKDW� ERWK� D� ILVFDO� ZRUVHQLQJ� �DV� LQ� DQ� LQFUHDVH� RI� 7XUNH\¶V� JRYHUQPHQW� H[SHQGLWXUH� RYHU� *'3� IURP� ������ LQ� 4�� ����� WR� ������ LQ� �4� ����� RU� DQ� LQFUHDVH� RI� SXEOLF� GHEW� RYHU� *'3� IURP� ������ LQ� ����� WR� ������ LQ� ������ RU� D� ILVFDO� FRQVROLGDWLRQ��VXFK�DV�WKH�IDOO�LQ�JRYHUQPHQW�H[SHQGLWXUHV�RYHU�*3'�IURP�������LQ�4������� WR�������LQ�4�������RU�D�GHFUHDVH�LQ�SXEOLF�GHEW�RYHU�*'3�IURP������ LQ������WR�������LQ� ������ZRXOG�KDYH�VLJQLILFDQW�HIIHFWV�RQ�UHDO�H[FKDQJH�UDWHV��7KLV�UHODWLRQVKLS�EHWZHHQ�ILVFDO� SROLF\� DQG� UHDO� H[FKDQJH� UDWHV� SUHVHQWV� DQ� H[WUD� GLPHQVLRQ� WR� WKH� LQWHUDFWLRQ� EHWZHHQ� PRQHWDU\�DQG�ILVFDO�SROLFLHV��:KHQ�PRQHWDU\�SROLF\�IDFHV�VHULRXV�WUDGH-RIIV�UHJDUGLQJ�WKH� PRYHPHQWV�RI�FXUUHQF\��DQ�DFWLYH�ILVFDO�SROLF\�PLJKW�SURYLGH�DGGLWLRQDO�RSHUDWLRQDO�VSDFH�IRU� PRQHWDU\�SROLF\�E\�FRQVWUDLQLQJ�H[FKDQJH�UDWH�PRYHPHQWV� 348 BIS Papers No 67 Figure 8 Impulse response of real exchange rates (RER) to one standard deviation innovation in government expenditures -2 -1 0 1 2 3 4 1 2 3 4 5 6 7 8 9 10 Response of RER to Cholesky One S.D. GOV_EXPENDITURE Innovation -2 -1 0 1 2 3 1 2 3 4 5 6 7 8 9 10 Response of RER to Cholesky One S.D. GOV_CONS Innovation -2 -1 0 1 2 3 1 2 3 4 5 6 7 8 9 10 Response of RER to Cholesky One S.D. GOV_CONS_WAGES Innovation -2 -1 0 1 2 3 4 1 2 3 4 5 6 7 8 9 10 Response of RER to Cholesky One S.D. GOV_CONS_OTHER Innovation -1 0 1 2 3 4 1 2 3 4 5 6 7 8 9 10 Response of RER to Cholesky One S.D. GOV_INV Innovation -2 -1 0 1 2 3 4 5 1 2 3 4 5 6 7 8 9 10 Response of RER to Cholesky One S.D. GOV_INV_MACHINERY Innovation -2 -1 0 1 2 3 1 2 3 4 5 6 7 8 9 10 Response of RER to Cholesky One S.D. GOV_INV_CONST_BUILD Innovation -2 -1 0 1 2 3 1 2 3 4 5 6 7 8 9 10 Response of RER to Cholesky One S.D. GOV_INV_CONST_NONBUILD Innovation BIS Papers No 67 349 4. Conclusion Fiscal policy is an important determinant of the effectiveness of monetary policy. A strong fiscal stance can increase the effectiveness of monetary policy by promoting the development of domestic financial markets and of longer maturity yield curves, thus reinforcing the economy’s risk structure and reducing structural weaknesses such as mismatches and pass-through in the economy. A weak fiscal position severely restricts the scope for monetary policy because it increases the potential for problems in the above- mentioned channels. This relationship between fiscal and monetary policy is dynamic, changing over time both at the country and the global level. The case of Turkey, like that of most other developing countries, shows how fiscal policy evolved during the 1990s within the context of a severely constraining monetary policy into a strong fiscal stance in the 2000s that greatly improved the effectiveness of monetary policy. Another dimension of fiscal-monetary policy interaction might come from the globalisation of the world economy. Structural factors create inflation differentials between advanced and emerging countries that in turn put steady upward pressure on developing country currencies. However, monetary policies in these countries are constrained by inflation differentials, policy rates in developed countries and financial stability concerns. Hence they cannot effectively soften the appreciation pressures. This might lead to the active use of fiscal consolidation in developing countries to help lessen the appreciation pressures, therefore presenting a new dimension in fiscal-monetary policy interaction at a global level. References Agenor, P, C McDermont and M Üçer (1997): “Fiscal imbalances, capital inflows, and the real exchange rate: the case of Turkey”, European Economic Review, 41, pp 819–25. $UVODQ��<�DQG�(�&HULWR÷OX���������³4XDOLW\�JURZWK�YHUVXV�LQIODWLRQ�LQ�7XUNH\´��&HQWUDO�%DQN�RI� the Republic of Turkey, Working Paper, no 11/21. Beetsma, R, N Giuliodori and F Klaassen (2008) “The effects of public spending shocks on trade balances and budget deficits in the European Union”, Journal of the European Economic Association, 6, pp 414–23. Bénétrix, A and P Lane (2009): “Fiscal shocks and the real exchange rate”, IIIS Discussion Paper, no 286. Bils, M and P Klenow (2001): “Quantifying quality growth”, American Economic Review, 91 (4), pp 1006–30. Central Bank of the Republic of Turkey (2011): Inflation Report I, January. Corsetti, G, L Dedola and S Leduc (2011): “Demand imbalances, exchange rate misalignments and monetary policy”, mimeo, University of Cambridge. Filho, C and M Chaman (2008): “The myth of post-reform income stagnation: evidence from Brazil and Mexico”, IMF Working Paper, no 08/197. Frankel, J. and A Razin (1992): Fiscal policies and the world economy, MIT Press. Frankel, J, C Vegh and G Vuletin (2011): “On graduation from fiscal procyclicality”, mimeo, University of Maryland. Hebous, S (2011): “The effects of discretionary fiscal policy on macroeconomic aggregates: a reappraisal”, Journal of Economic Surveys, 25(4), pp 674–707. .DUD�� +� DQG� )� g÷�Qo� �������� ³,QIODWLRQ� WDUJHWLQJ� DQG� H[FKDQJH� UDWH� SDVV-through: the Turkish experience”, Emerging Markets Finance and Trade, 44, no 6, pp 52–66. 350 BIS Papers No 67 Kim, S and N Roubini (2008): “Twin deficit or twin divergence? Fiscal policy, current account, and real exchange rate in the U.S.”, Journal of International Economics, 74, pp 362–83. Kumhof, M and D Laxton (2009): “Fiscal deficits and current account deficits”, IMF Working Paper, 09/237. Monacelli, T and R Perotti (2010): “Fiscal policy, the real exchange rate, and traded goods”, Economic Journal, 120, pp 437–61. Penati, A (1986): “Government expenditure, real exchange rate and the ‘decline’ of the manufacturing sector”, some empirical evidence”, Economics Letters, 21, pp 365–70. Ramey, V (2011): “Identifying government spending shocks: it’s all in the timing”, The Quarterly Journal of Economics, 126(1), pp 1–50. Ravn, M, S Schmitt-Grohe and M Uribe (2007): “Examining the effects of government spending shocks on consumption and the real exchange rate”, NBER Working Paper, no 13328. <|U�NR÷OX��0���������³'LIILFXOWLHV�LQ�LQIODWLRQ�PHDVXUHPHQW�DQG�PRQHWDU\�SROLF\�LQ�HPHUJLQJ� market economies”, BIS Paper, no 49. BIS Papers No 67 351 List of participants Central Bank of Argentina Miguel Angel Pesce Deputy Governor Norberto Pagani Senior Manager of International Relations Central Bank of Brazil Carlos Hamilton Vasconcelos Araújo Deputy Governor Central Bank of Chile Sebastián Claro Board Member People’s Bank of China Xin Wang Chief Representative, Representative Office in Frankfurt Bank of the Republic (Colombia) Hernando Vargas Deputy Governor Czech National Bank Vladimír Tomšik Vice-Governor Hong Kong Monetary Authority Dong He Executive Director of Research Magyar Nemzeti Bank (the central bank of Hungary) Ferenc Karvalits Deputy Governor Reserve Bank of India Rama Subramaniam Gandhi Executive Director Bank Indonesia Mr Hendar Director, Monetary Management Erwin Gunawan Hutapea Senior Economist, Monetary Management Bank of Israel Karnit Flug Deputy Governor Bank of Korea Jun Il Kim Deputy Governor Central Bank of Malaysia Sukhdave Singh Assistant Governor Bank of Mexico José Julián Sidaoui Board Member Central Reserve Bank of Peru Renzo Rossini General Manager Bangko Sentral ng Pilipinas Diwa C Guinigundo Deputy Governor, Monetary Stability Sector Bank of Russia Alexander V Kashturov Director of Market Operations 352 BIS Papers No 67 Saudi Arabian Monetary Agency Abdulrahman Al-Hamidy Vice Governor Suliman Al Saeed Director, Public Relations Fahad Al Dossari Director General, Research and Statistics Monetary Authority of Singapore Chong Tee Ong Deputy Managing Director South African Reserve Bank Elias Lesetja Kganyago Deputy Governor Bank of Thailand Suchada Kirakul Deputy Governor, Monetary Stability Punpilas Ruangwisut Team Executive, Monetary Policy Office Central Bank of the Republic of Turkey 0HKPHW�<|U�NR÷OX Deputy Governor Mustafa Kilinc Economist, Research and Monetary Policy Bank for International Settlements Stephen Cecchetti Economic Adviser and Head of the Monetary and Economic Department Philip Turner Deputy Head of the Monetary and Economic Department and Director, Policy Coordination and Administration Madhusudan Mohanty Head, Emerging Markets Fiscal policy, public debt and monetary policy in emerging market economies Contents M S Mohanty: Fiscal policy, public debt and monetary policy in EMEs: an overview 1. Introduction 2. Fiscal constraints on monetary policy Shift to countercyclical fiscal and monetary policy Fiscal policy and interest rates Future fiscal risks 3. Local currency bond markets and central bank policies Implications for the conduct of monetary policy Implications for the monetary transmission mechanisms 4. Central banks and public debt management References Carlos Montoro, Elöd Takáts and James Yetman: Is monetary policy constrained by fiscal policy 1. Introduction 2. Factors influencing the relationship between fiscal and monetary policy Countercyclicality of fiscal policy Fiscal sustainability Contingent liabilities of the government 3. Consequences for monetary policy Monetary and fiscal stabilisation Fiscal deficits and government debt: effects on interest rates The inflation effects of fiscal deficit 4. Conclusions Pension liabilities and demographics Sovereign wealth funds Fiscal rules References Aaron Mehrotra, Ken Miyajima and Agustín Villar: Developments of domestic government bond markets in EMEs and their implications 1. Introduction 2. How far have domestic government bond markets developed in EMEs? (i) Size (ii) Composition Maturity Type Investor base (iii) Market liquidity 3. What factors have contributed to bond market development? 4. Implications for the conduct of monetary policy 5. Impact on financial stability Diversification of credit risk Foreign holdings and derivatives markets References Andrew Filardo, Madhusudan Mohanty and Ramon Moreno: Central bank and government debt management-issues for monetary policy Introduction 1. The size and maturity structure of government and central bank debt The size and maturity of government debt The size and the maturity structure of central bank debt 2. The consolidated official debt held by the public Short-term debt and money 3. Implications for the monetary transmission mechanism The short end of the yield curve The curvature of the yield curve Short-term debt and bank credit 4. Conclusion and comments on policy coordination Information coordination Centralised debt management units in central banks and explicit prohibitions References Miguel Angel Pesce: Fiscal policy, public debt management and government bond markets: issues for central banks Abstract 1. Introduction 2. Some macroeconomic thoughts 3. Aspects of the regulatory framework in relation to sovereign debt 4. Institutional arrangements: interaction between central banks and Treasuries during periods of crisis 5. Final comments Carlos Hamilton Araújo, Cyntia Azevedo and Silvio Costa: Fiscal consolidation and macroeconomic challenges in Brazil 1. Introduction 2. The Brazilian fiscal framework 3. Effects of fiscal policy on inflation and output 4. Final remarks References Sebastián Claro and Claudio Soto: Macro policies and public debt in Chile I. Introduction II. The evolution of public debt in Chile III. Fiscal policy and public debt IV. Monetary policy and public debt People's Bank of China: Monetary policy, fiscal policy and public debt management I. Public debt sustainability and monetary policy (1) When measuring the fiscal stance, the following factors should be taken into consideration: (2) Public sector assets as offset to gross debt (3) The influence of fiscal deficit financing on monetary policy II. Domestic currency-denominated public debt in China’s domestic market (1) Money market development (2) Maturity and yield curves of domestic government bonds (3) Deepening domestic financial markets and financial stability III. Central bank and public debt management (1) Whether the central bank can issue its own debt paper, and its coordination with the Ministry of Finance (2) Involvement of central banks in public debt management (3) Institutional arrangements for coordinating monetary policy and public debt management Hernando Vargas, Andrés González and Ignacio Lozano: Macroeconomic effects of structural fiscal policy changes in Colombia 1. Introduction 2. Fiscal policy in Colombia 3. Public debt management in Colombia 4. The macroeconomic effects of the fiscal policy changes a. Effects on the sovereign risk premium b. Effects on the short-run response of output to government expenditure shocks c. Effects on the transmission of monetary policy shocks to market interest rates 5. Conclusion References Vladimír Tomsík: Some insights into monetary and fiscal policy interactions in the Czech Republic 1. Introduction 2. The effects of fiscal policy on monetary policy transmission 3. Two alternative methods for cyclical adjustment of fiscal balance 4. Fiscal policy in the CNB forecast 5. Summary and conclusions References Hong Kong Monetary Authority: The importance of fiscal prudence under the Linked Exchange Rate System in Hong Kong SAR Fiscal position of the Hong Kong government The need for strong fiscal reserves to cope with shocks The role of fiscal reserves in fortifying the Linked Exchange Rate system Gergely Baksay, Ferenc Karvalits and Zsolt Kuti: The impact of public debt on foreign exchange reserves and central bank profitability: the case of Hungary Introduction The role of foreign currency-denominated debt in Hungary’s public finances The effect of foreign currency-denominated public debt on foreign reserve dynamics Foreign currency debt issuance can contribute significantly to the growth of foreign exchange reserves Foreign currency issuances can fall short in times of market stress, resulting in a lower-than-expected level of foreign exchange reserves Hedging foreign exchange risk can cause relatively large swings in foreign currency reserves The effect of foreign currency-denominated public debt on the required level of foreign reserves The total and relative amounts of public debt matter for the reserve requirement Heavy public indebtedness can cause significant interest rate differentials which can push up carry trader’s demand for short-term assets Flows from margin calls can also contribute to the volatility of the reserve requirement Long-term debt, usually considered to be stable, can also be a source of instability when liquidity needs suddenly increase during sudden sell-offs The effect of foreign currency-denominated public debt on the central bank’s profit and loss Lessons from the viewpoint of foreign reserve management Growing public debt tends to increase the level of uncertainty in the dynamics of foreign exchange reserves The central bank needs to have a buffer above necessary level of reserves The central bank’s autonomous instruments may not be sufficient to counteract volatility in reserve levels and reserve requirement Efficient coordination between government agencies is vital Additional sources of foreign currency liquidity could play an important role The costs and benefits of foreign currency debt should be considered on a consolidated basis Conclusion References R Gandhi: Sovereign debt management in India: interaction with monetary policy Public debt management framework Debt management experience during the crisis Interaction with monetary policy Further issues References Mr Hendar: Fiscal policy, dublic debt management and government bond markets in Indonesia Fiscal policy overview Bonds market, money market and monetary policy implementation The coordination of monetary policy and fiscal policy Kobi Braude and Karnit Flug: The interaction between monetary and fiscal policy: insights from two business cycles in Israel Introduction Two recessions – different circumstances A different policy response Why was the policy response so different? An illustrative estimate of the effect of the different policy responses Looking ahead – confronting the looming crisis Concluding remarks References Dr Geum Wha Oh: Public debt and monetary policy in Korea I. Introduction II. Fiscal policy and public debt management: an overview Institutional setup for fiscal policy Fiscal policy during crisis periods III. Sovereign debt market Sovereign debt market structure Debt instruments of the central bank IV. Capital flows and monetary policy V. Further considerations Long-run fiscal challenges Financial risks to the Bank of Korea Household debt and monetary policy José Sidaoui, Julio Santaella and Javier Pérez: Banco de México and recent developments in domestic public debt markets 1. Introduction 2. Policy aimed at developing local debt markets 2.1 Sound macroeconomic policy 2.2 Minimal market intervention 2.3 Pension system reforms 2.4 Improved securities clearing and settlement systems 2.5 Completing the market’s information set 2.6 The use of government securities as monetary policy instruments 2.7 The issuance of warrants 2.8 Additional measures to enhance the liquidity of government securities 3. The development of derivatives markets 4. The advantages of a well developed government securities market 4.1 Lower currency risk exposure 4.2 Lower interest-rate risk exposure 4.3 Lower refinancing risk for public and private issuers 4.4 Higher liquidity 4.5 Higher diversification of the investor base for domestic public debt 5. The stability of recent capital inflows 6. Concluding remarks 7. References Renzo Rossini, Zenón Quispe and Jorge Loyola: Fiscal policy considerations in the design of monetary policy in Peru Introduction Financial Policy Coordination Macroeconomic Policy Coordination Conclusions References Diwa C Guinigundo: Fiscal policy, public debt management and government bond markets: the case for the Philippines 1. Introduction 2. Potential for constraints on monetary policy from an unsustainable path for public debt 2.1 Measurement of the fiscal policy stance and public debt 2.2 Interaction between monetary and fiscal policy 3. Domestic currency public debt issues in local markets 3.1 Shift from international to domestic financial markets for public debt 3.2 Implications for capital market development 3.3 Lengthening maturity of domestic government bonds 3.4 Financial stability concerns 4. BSP and public debt management 4.1 BSP’s issuance of its own securities 4.2 BSP’s role in government debt management 4.3 Governance arrangements for the coordination of monetary policy and public debt management NEDA Board and related inter-agency committees The Development Budget Coordinating Committee (DBCC) The Investment Coordination Committee (ICC) 5. Conclusion References: Tomasz Jedrzejowicz, Witold Kozinski: A framework for fiscal vulnerability assessment and its application to Poland Introduction 1. Level of public debt 2. Medium-term dynamics of public debt 3. Long-term sustainability of public debt 4. Public debt management and liquidity position of the government 5. Fiscal rules and institutions Poland: an assessment of fiscal vulnerability Bibliography Elizaveta Danilova: Fiscal policy, public debt management and government bond markets: issues for central banks 1. Fiscal policy: mechanisms that promote sustainability and their influence on monetary policy in Russia 2. The Bank of Russia’s role in public debt management 3. Main features of the domestic currency public debt market Abdulrahman Al-Hamidy: Aspects of fiscal/debt management and monetary policy interaction: the recent experience of Saudi Arabia 1. Introduction 2. Saudi Arabia’s experience (i) Relevance of export revenues in the oil sector (ii) Policy challenges for the Saudi government (iii) Monetary policy (iv) The monetary process and causative factors for the money supply (v) Interaction of fiscal and monetary policy (vi) Development of the financial system Conclusion Monetary Authority of Singapore: Development of the government bond market and public debt management in Singapore (A) Singapore Government Securities (SGS) Overview of the bond market (B) Singapore’s fiscal strength and motivation for SGS issuance Strong fiscal position Catalyzing growth of the domestic bond market (C) Recent development in the SGS market and issuance of central bank debt Global capital flows and their impact on SGS yields Central bank debt security – MAS bills (D) Challenges and key policy considerations Lesetja Kganyago: Fiscal policy, public debt management and government bond markets: issues for central banks A. Is monetary policy constrained by unsustainable paths of public debt? i. The fiscal stance ii. Stock of public sector assets Central bank assets State pension funds iii. Financing of fiscal deficits and monetary policy B. Domestic currency public debt issued in local markets iv. Shift from international to domestic markets for public debt v. Money market development vi. Lengthening maturity of domestic government bonds and developing yield curves vii. Financial stability risks related to deeper domestic financial markets Larger capital inflows, more assets and liquidity to support speculative activity Faster transmission of external shocks C. Central banks and public debt management viii. Central bank debt paper ix. Short-term vs long-term public debt x. Domestic vs foreign currency debt xi. Central bank balance sheets and by quasi-fiscal operations xii. Governance arrangement for the coordination of monetary policy and public debt management Suchada Kirakul: Fiscal policy and its implication for central banks 1. Introduction 2. Country experience: Thailand 2.1 Development of the fiscal position and public debt 1. The pre- and post-Asian crisis: 1990 to 2007 2. Cushioning the impact of the global financial crisis: 2007 onwards 2.2 Fiscal risks in the medium term 3. Implications of fiscal risk for the central bank and challenges ahead 4. The roles of public debt management and central bank bond issuance in fostering bond market development and promoting financial stability 4.1 The government and the Bank of Thailand have made continuous efforts to strengthen and deepen the domestic bond market to create a more resilient financial sector 4.2 Nevertheless, as cross-border capital flows surge, it is critical to strike a balance between promoting market development and ensuring financial stability Mehmet Yörükoğlu and Mustafa Kılınç: Globalisation of the interaction between fiscal and monetary policy 1. Introduction 2. Evolution of fiscal-monetary policy interaction in Turkey from the 1990s to the 2010s 2.1. Fiscal dominance and constrained monetary policy in the 1990s 2.2. Reforms, fiscal consolidation and independent monetary policy in the 2000s 3. The global dimensions of fiscal-monetary policy interaction 3.1. The inflation differential between advanced and developing countries 3.2. Implications for real exchange rates, monetary policy and fiscal policy 3.3. Fiscal policy and real exchange rates in Turkey 4. Conclusion References List of participants

The Challenge of Debt Reduction during Fiscal

Consolidation

Luc Eyraud and Anke Weber

WP/13/67

© 2013 International Monetary Fund WP/13/

IMF Working Paper

Fiscal Affairs Department

The Challenge of Debt Reduction during Fiscal Consolidation

Prepared by Luc Eyraud and Anke Weber1

Authorized for distribution by Martine Guerguil

2013

Abstract

Studies suggest that fiscal multipliers are currently high in many advanced economies. One
important implication is that fiscal tightening could raise the debt ratio in the short term, as
fiscal gains are partly wiped out by the decline in output. Although this effect is not long-
lasting and debt eventually declines, it could be an issue if financial markets focus on the
short-term behavior of the debt ratio, or if country authorities engage in repeated rounds of
tightening in an effort to get the debt ratio to converge to the official target. We discuss
whether these problems could be addressed by setting and monitoring debt targets in
cyclically-adjusted terms.

JEL Classification Numbers: C54, E62, H60

Keywords: fiscal consolidation, fiscal multipliers, public debt

Authors’ E-Mail Addresses: leyraud@imf.org; aweber@imf.org

1 This paper received useful comments and suggestions from Swarnali Ahmed, Jochen Andritzky, Nathaniel
Arnold, Aqib Aslam, Benjamin Carton, Carlo Cottarelli, Lorenzo Forni, Dmitry Gershenson, Phillip Gerson,
Emine Hanedar, Martine Guerguil, Kotaro Ishi, Bernd Lucke, Esther Perez Ruiz, Alasdair Scott, Ikuo Saito,
Justin Tyson, and Yingbin Xiao. It also benefited from the discussions during the surveillance meeting seminar
held at the IMF on April 10, 2012, and the joint IMF-OECD Workshop on December 12, 2012 (OECD, Paris).
Raquel Gomez provided excellent research assistance.

This Working Paper should not be reported as representing the views of the IMF.
The views expressed in this Working Paper are those of the author(s) and do not necessarily
represent those of the IMF or IMF policy. Working Papers describe research in progress by
the author(s) and are published to elicit comments and to further debate.

2

I. INTRODUCTION

With large fiscal adjustments underway in many advanced economies, there is a renewed
interest in the effects of fiscal policy on economic activity. Surprisingly, the feedback effect
from growth to fiscal aggregates, and in particular to public debt, has received less attention.
This issue is becoming more pressing as debt reduction has become a key target of fiscal
policy in a number of advanced economies.

This paper examines the impact of fiscal consolidation on debt dynamics.2 We first analyze
the size of short-term fiscal multipliers, which are key parameters in the nexus between fiscal
consolidation, growth, and debt reduction. We show that in the current environment,
multipliers for advanced economies could be close to 1, significantly above the average
multiplier found in the literature before the recent financial crisis. The negative impact of
fiscal tightening on economic activity in the near term is indeed amplified by some features
of the current environment, including the proportion of credit-constrained agents, the
depressed external demand, and the limited room for monetary policy to be more
accommodative.

With multipliers close to 1, fiscal consolidation is likely to raise the debt ratio in the short-
run in many countries. Although the debt ratio eventually declines, its slow response to fiscal
adjustment could raise concerns if financial markets react to its short-term behavior. It may
also lead country authorities to engage in repeated rounds of tightening in an effort to get the
debt ratio to converge to the official target. Not explicitly taking into account multipliers or
underestimating their value may lead policymakers to set unachievable debt targets and
miscalculate the amount of adjustment necessary to bring the debt ratio down.

These findings do not imply that fiscal consolidation is undesirable or that debt is
unsustainable. The short-term effects of fiscal policy on economic activity are only one of the
many factors that need to be considered in determining the appropriate pace of fiscal
adjustment. Our results rather highlight the need for care in projecting the debt path and the
benefits of setting and monitoring debt targets in cyclically-adjusted terms.

Our paper contributes to the literature in several ways. First, it formalizes the relation
between fiscal adjustment and debt dynamics by explicitly taking into account how this
relation is influenced by fiscal multipliers. Second, it provides empirical evidence supporting
the hypothesis that fiscal consolidation can raise the debt ratio in the short-term. Third, to our
knowledge, it provides one of the first extensive analyses of the “cyclically adjusted debt
ratio” concept.3

The paper is organized as follows: Section II reviews the literature and discusses the

2 This analysis builds on and extends Eyraud and Weber (2012).

3 The European Commission has analyzed a similar concept in the context of the discussions on the debt
reduction benchmark (see EC, 2011).

3

analytical framework, including the fiscal multiplier assumptions. Sections III and IV
analyze the impact of fiscal consolidation on debt dynamics through simulations and
econometric estimations. Section V draws policy implications, and in particular, evaluates
the relevance of debt indicators in cyclically-adjusted terms. Section VI concludes.

II. FISCAL CONSOLIDATION AND DEBT REDUCTION: THE ANALYTICAL FRAMEWORK

A. Literature Review

Although debt dynamics have been the focus of many theoretical and empirical papers (in
particular those dealing with fiscal sustainability, like Escolano 2010), few have specifically
analyzed the effect of fiscal policy on the debt ratio, probably because the existence of an
accounting relationship suggests that the relation between these variables is straightforward.

From a conceptual point of view, Gros (2011) and the European Commission (2012) develop
a framework akin to ours. Gros (2011) shows that austerity could be self-defeating and
increase the debt ratio in the short-term. However, Gros does not examine the impact of
repeated episodes of tightening; neither does he explore the implications of multiplier
persistence—two key factors affecting the debt response, as we show below.

Model-based simulations are an important instrument to assess the effect of fiscal policy on
the debt ratio. Many dynamic stochastic general equilibrium (DSGE) models incorporate the
government budget constraint and report fiscal multipliers on this basis, while explicitly
tracking the behavior of the public debt-to-GDP ratio. For instance, Forni and others (2009)
analyze the response of macro variables to a range of fiscal shocks in the Euro Area using
quarterly data. They find that following a 1 percent increase in government purchases of
goods and services, the debt ratio initially declines by about 1 percent before increasing from
the fourth quarter onwards. The response is symmetric, so that in case of a negative shock to
government spending, the model would predict an initial increase in the public debt-to-GDP
ratio.

Until recently, empirical papers analyzing the impact of fiscal policy on output have omitted
the debt variable. Favero and Giavazzi (2007, 2009) have shown that this omission may
result in incorrect estimates of the dynamic effects of fiscal shocks. They argue that by not
taking into account the government intertemporal budget constraint, previous studies may
have assessed the effects of fiscal policy along unsustainable debt paths. They incorporate
debt dynamics into a structural vector autoregression (SVAR) for the United States, allowing
for the possibility that taxes, spending and interest rates respond to the debt level. They show
that fiscal multipliers differ in models with and without debt feedbacks. Building on the
Favero and Giavazzi framework, Cherif and Hasanov (2010) use generalized impulse
responses to analyze macro variable dynamics, and find that positive shocks to the primary
surplus reduce public debt. Ilzetzki (2011) extends the analysis to a sample of developing
countries but concludes that in most countries, including a debt feedback effect does not
change the size of fiscal multipliers significantly.

4

B. Some Unpleasant Fiscal Arithmetic

The debt ratio does not decrease one-for-one with fiscal tightening.4 The reason is that fiscal
tightening reduces GDP in the near-term—an effect referred to as the “fiscal multiplier.”5
Lower GDP in turn reduces the denominator of the debt ratio and also partly offsets the
consolidation effort, thus affecting the numerator (automatic stabilizers). This may mitigate
or even eliminate the direct tightening effect, at least in the short run.

The mitigating effect is stronger if fiscal multipliers, the initial debt ratio, and/or automatic
stabilizers are larger (see Appendix 1 for detailed formulas).6 For instance, a 1 percent of
GDP consolidation relative to the baseline should reduce the debt ratio in the first year by:

οቀ����
��ൗ ቁ כ ͳͲͲ ൎ െͳฏ
ୢ୧୰ୣୡ୲�ୣ୤୤ୣୡ୲

൅ ݎ݈݁݅݌݅ݐ݈ݑ݉ כ ᇣᇧᇧᇧᇧᇧᇧᇧᇤᇧᇧᇧᇧᇧᇧᇧᇥ݋݅ݐܽݎ�ݐܾ݁݀
ௗ௘௡௢௠௜௡௔௧௢௥�௘௙௙௘௖௧

൅ ݎ݈݁݅݌݅ݐ݈ݑ݉ כ ᇣᇧᇧᇧᇧᇧᇧᇧᇧᇤᇧᇧᇧᇧᇧᇧᇧᇧᇥ݋݅ݐܽݎ�݁ݑ݊݁ݒ݁ݎ
௡௨௠௘௥௔௧௢௥�௘௙௙௘௖௧�ሺ௔௨௧௢௠௔௧௜௖�௦௧௔௕௜௟௜௭௘௥௦ሻ

ᇩᇭᇭᇭᇭᇭᇭᇭᇭᇭᇭᇭᇭᇭᇭᇭᇭᇭᇭᇭᇪᇭᇭᇭᇭᇭᇭᇭᇭᇭᇭᇭᇭᇭᇭᇭᇭᇭᇭᇭᇫ
௠௜௧௜௚௔௧௜௡௚�௘௙௙௘௖௧

In a country with a debt ratio of 80 percent of GDP, a revenue ratio of 40 percent of GDP,
and a first-year multiplier of 0.6, a fiscal tightening of 1 percent of GDP lowers the debt ratio
by only 0.3 percentage points of GDP in the first year (other factors being equal, including
the interest rate). The offsetting effect of the multiplier amounts to 0.7 percent of GDP, of
which 0.5 is due to the denominator effect and 0.2 is due to automatic stabilizers in the
numerator.

In addition, this mitigating effect is persistent. After some time, multipliers are likely to
decline, and GDP eventually reverts to its baseline level,7 but the debt ratio will not fully
catch up. The effect on the debt ratio will still be less than the cumulated amount of fiscal
tightening, because unlike the denominator, the numerator of the debt ratio, the debt stock,
has increased relative to the baseline and does not fully revert. The following formula
illustrates this point. A permanent tightening of 1 percent of GDP should reduce the debt

4 In this paper, “debt” refers to public debt in gross terms, unless otherwise indicated.

5 Fiscal multipliers are defined as the ratio of a change in output to an exogenous change in the fiscal deficit
with respect to their respective baselines. In the formulas of Appendix 1 and the simulations of Section III,
multipliers are calculated as ratios of nominal variables. These “nominal” multipliers may be larger than
standard multipliers calculated in real terms. Indeed, when real GDP declines with fiscal tightening, inflation
also decelerates; thus, the decline in nominal GDP is larger than the decline in real GDP. This is one of the
reasons why our simulations should be based on higher-than-average multipliers (see Section II.C).

6 We use a simplified framework where the size of automatic stabilizers is measured by the revenue ratio.

7 Several factors explain why the negative effect of fiscal tightening on output eventually disappears (even when
the tightening is permanent). These include: (i) anticipating lower output and inflation in the future, the central
bank may lower interest rates; (ii) fiscal tightening may be perceived as credible and reduce the risk premium
on interest rates; (iii) the currency may depreciate in response to lower interest rates; and (iv) households may
anticipate a decline in their tax burden in the future and increase current consumption.

5

ratio by N percent of GDP after N periods if there is no fiscal multiplier.8 But with the fiscal
multiplier effect, both the denominator and the numerator of the debt ratio deviate from their
expected path (in a cumulative way for the numerator). If multipliers decline over time and
the N-year multiplier is zero, the denominator effect disappears but the numerator effect does
not, reflecting the impact of automatic stabilizers on past deficits. As a result, the mitigating
effect does not fully disappear.

οቀ����୒ �୒ൗ ቁ כ ͳͲͲ ൎ െ�
ฏ

ୢ୧୰ୣୡ୲�ୣ୤୤ୣୡ୲
൅ ேݎ݈݁݅݌݅ݐ݈ݑ݉ כ ேᇣᇧᇧᇧᇧᇧᇧᇧᇤᇧᇧᇧᇧᇧᇧᇧᇥ݋݅ݐܽݎ�ݐܾ݁݀

ௗ௘௡௢௠௜௡௔௧௢௥�௘௙௙௘௖௧
൅ ෍݉ݎ݈݁݅݌݅ݐ݈ݑ௜

ே

௜ୀଵ
כ ݋݅ݐܽݎ�݁ݑ݊݁ݒ݁ݎ

ᇣᇧᇧᇧᇧᇧᇧᇧᇧᇧᇤᇧᇧᇧᇧᇧᇧᇧᇧᇧᇥ
௡௨௠௘௥௔௧௢௥�௘௙௙௘௖௧�ሺ௔௨௧௢௠௔௧௜௖�௦௧௔௕௜௟௜௭௘௥௦ሻ

ᇩᇭᇭᇭᇭᇭᇭᇭᇭᇭᇭᇭᇭᇭᇭᇭᇭᇭᇭᇭᇭᇪᇭᇭᇭᇭᇭᇭᇭᇭᇭᇭᇭᇭᇭᇭᇭᇭᇭᇭᇭᇭᇫ
௠௜௧௜௚௔௧௜௡௚�௘௙௙௘௖௧

C. Fiscal Multipliers in the Current Environment

The near-term path of the debt and deficit in response to a discretionary tightening depends
on the size of fiscal multipliers. We need to make an assumption on multipliers before
conducting simulations in Section III. While the literature suggests an average first year
fiscal multiplier of around 0.6 for advanced economies,9 there are reasons to believe that in
the current environment, it could be closer to 1.10

Recent empirical research finds that fiscal multipliers are significantly larger in downturns
than in expansions (Auerbach and Gorodnichenko, 2012; Batini and others, 2012; Baum and
others, 2012). Intuitively, in recessions, the proportion of credit-constrained households and
firms, which adjust spending in response to a change in disposable income, is higher. For G7
economies, Baum and others (2012) find that first year spending and revenue multipliers in
downturns are, on average, estimated at 1.3 and 0.4, respectively.11 Assuming, in line with

8 All formulas are calculated relative to baseline. Absent fiscal multipliers, a one-off permanent tightening
would improve the fiscal balance by 1 percent of GDP in each period, and lower public debt by N percent of
GDP relative to the baseline after N periods.

9A recent literature review extends and updates earlier IMF work by Spilimbergo and others (2009) and finds
that average first-year multipliers amount to 0.8 for spending and 0.3 for revenue measures (Mineshima and
others, 2013). Since about two-thirds of recent fiscal adjustments in advanced economies rely on spending
measures, this gives an average overall multiplier of 0.6 for the first year.

10 Obviously not all countries may currently experience multipliers close to 1. Multipliers depend on country
characteristics, as reflected in the wide range of spending multipliers across OECD economies. In line with the
theory, simple correlations suggest, in particular, that fiscal multipliers tend to be smaller in more open
economies, in countries with larger automatic stabilizers and higher interest rates (IMF, 2012).

11The finding that first year spending multipliers are higher than revenue multipliers is in line with a number of
recent studies (Mineshima and others, 2013), although this result is debated. For instance, IMF (2010) finds that
spending-based adjustments are less contractionary and notes that this is partly due to the fact that central banks
lower interest rates more in case of expenditure-based consolidations (perhaps because they regard them as
more long-lasting). However, when policy rates are already low, the interest rate response becomes less likely,
which may imply that, in the current environment, the Keynesian theory prediction prevails.

6

recent fiscal adjustment packages in advanced economies, that two thirds of the adjustment
come from spending measures, a weighted average of spending and revenue multipliers in
downturns yields an overall fiscal multiplier of about 1, which we will use in the next section
for simulating the impact of fiscal consolidations.

Several other factors also explain why multipliers are currently likely to be above average in
advanced economies. In many countries there is limited room for monetary policy to become
more accommodative. Policy rates have been near the zero lower bound since the onset of the
Great Recession and, as shown by simulations in IMF (2010), under such conditions the
multiplier would be close to 1. Also, a number of advanced economies experienced a major
credit boom and financial crisis in the past decade. Post-crisis deleveraging by the financial
sector and households, both of whom are seeking to rebuild their balance sheets, can further
impair the transmission of easier monetary policy as both the supply (banks tighten lending
standards) and demand (more credit constrained households trying to cut their debts) for
credit fall.12

In addition, at present, a country cannot rely as much as before on external demand to help
cushion the impact of fiscal consolidation on growth because many of the advanced
economies are growing slowly or not at all. Results from simulations support these
conclusions. IMF (2010) finds that when the rest of the world is tightening at the same time,
the output cost of a 1 percent of GDP fiscal consolidation can double to 2 percent for a small
open economy where the interest rate is at the zero lower bound.

High multipliers do not necessarily imply that fiscal consolidation should be avoided or
postponed. As mentioned above, many other factors have an impact on the choice or the
desired pace and size of fiscal adjustment. Nonetheless, the level and persistence of
multipliers shape the near term paths of deficits and debt. The simulations in the next section
illustrate specifically how.

III. SIMULATION RESULTS

This section assesses the impact of discretionary fiscal tightening on the debt ratio, based on
the debt dynamics formulas of Appendix 1. It shows that fiscal consolidation can increase the
debt ratio in the short-term, and estimates how long it would take for the debt ratio to decline
substantially. Section A presents the results for a one-off consolidation, while section B
extends the analysis to repeated episodes of tightening with persistent multipliers.

12 For certain countries the argument that the constraint imposed by the zero lower bound restricts the ability of
monetary policy to become more accommodative needs to be qualified somewhat. In those countries where
sovereign debt spreads and private sector borrowing rates are high, despite the fact that the policy rate is near
the zero lower bound, monetary policy would effectively become more accommodative if further actions caused
spreads to fall in those countries.

7

A. Short-Term and Medium-Term Impacts of Fiscal Consolidation on Debt

When multipliers are close to 1, fiscal consolidation is likely to raise the debt ratio in the first
year in most advanced economies.13 Figure 1 shows the first-year impact of a 1 percent of
GDP discretionary tightening on the debt ratio. The impact is simulated for a range of fiscal
multipliers and initial debt ratios (assuming other factors, such as interest rates, remain
constant).

We find that, with a multiplier of 1—a reasonable level in downturns (as discussed in Section
II)—fiscal consolidation leads to a debt ratio increase in the first year in countries where the
debt ratio initially lies above 60 percent. The debt threshold varies with the multiplier, which
itself depends on the composition of the adjustment (spending vs. revenue) and other
country-specific factors.

Figure 1. Impact on the Debt Ratio of a 1 Percent of GDP
Discretionary Tightening In the First Year

(Relative to Baseline)

Source: IMF Staff calculations. Note: Multipliers are weighted
averages of spending and revenue multipliers. High and low
multipliers are based on the empirical literature (excluding outliers).

To assess the impact of fiscal tightening after the first year, we simulate the debt response
over a 5-year period. The impact on the debt ratio is calculated for three groups of countries
with different initial debt ratios, using a range of fiscal multipliers. For the simulations
presented in the following charts, we use the 2011 debt and revenue ratios of the euro area
countries, but the conclusions hold beyond this sample of countries.14 The calculations
assume that other factors remain constant, in particular interest rates.

13The first year is the year when fiscal policy is tightened; in the following charts, this is year t+1. All the
simulations are relative to the baseline (or counterfactual scenario) with zero GDP growth, a balanced budget,
and a constant debt ratio.

14The three groups are defined on the basis of the debt ratios observed in 2011 for the 17 euro area countries.
The average debt and revenue ratios for each group are then used in the simulations (the average debt ratios are,
respectively, 38, 73, and 113 percent of GDP). Simulations use different multipliers (high, low, and downturn),
derived from the literature survey by Mineshima and others (2013).

–

1.5

–

1

–

0.5

0

0.5
1
1.5

2

10 60 110 1

60

C
ha

ng
e

in
th

e
de

b

t

ra

tio
(p

er
ce

nt
o

f

G
D

P
)

Initial public debt ratio
(percent of GDP)

High multiplier = 1.3

Downturn multiplier = 1

Low multiplier = 0.1

8

If fiscal policy is tightened in the first year only, the debt ratio would generally decline from
the second year. Figure 2 shows the change in the public debt ratio with respect to the
baseline when the government tightens fiscal policy by 1 percent of GDP in the first year t+1.
The tightening is permanent (not reversed in the following years), but it is assumed that its
effect on output is temporary and fades away within 5 years—a standard assumption in the
literature. In medium- and high-debt countries, the debt ratio would decrease from the second
year while, consistent with the previous results, in low-debt countries the debt ratio already
declines in the first year.

Figure 2. Impact on the Debt Ratio of a Discretionary Fiscal Tightening
of 1 Percent of GDP in the First Year

(Relative to Baseline)

Source: IMF Staff calculations.
Note: These simulations measure the change in the debt ratio relative to a baseline determined by 2011 WEO
data and constant thereafter. Cumulative multipliers: High is 1.3 in the first year, 1.6 in the second year;
Downturn is 1 in the first year, 0.8 in the second year; Low is 0.1 in the first year, 0.1 in the second year.
Multipliers steadily decline to zero between the second and the fifth year.

B. Implications of Repeated Tightening and Persistent Multipliers

The previous results rely on two key assumptions; that fiscal tightening only takes place in
the first year, and that the negative effect of fiscal tightening on GDP fades away within
5 years. These two assumptions are questionable, particularly in the current environment.
Most countries are engaged in repeated rounds of fiscal tightening. Moreover, the negative
effect of fiscal consolidation on output may in fact be more persistent or even permanent
(i.e., GDP may stay durably below the baseline).15 We relax these two assumptions in the
following simulations. The formulas used in the simulations are provided in Appendix 1.

In high-debt countries that undertake repeated episodes of fiscal tightening, the decline in the
debt ratio may not occur before the third year. Figure 3 shows the change in the debt ratio
with respect to the baseline when the government introduces a cumulative 5-year package of

15For example, the IMF (2011) shows that fiscal consolidation has negative effects on output which persist into
the medium term. DeLong and Summers (2012) also argue that fiscal tightening may have hysteresis effects on
potential output.

-1

4

-12

–

10

-8

–

6

-4

-2

0
2
4

t t+1 t+2 t+3 t+4 t+5

Medium Debt Countries

High multipliers Downturn multipliers Low multipliers

-14

-12
-10
-8

-6

-4
-2
0
2
4
t t+1 t+2 t+3 t+4 t+5

Low Debt Countries

-14
-12
-10
-8
-6
-4
-2
0
2
4
t t+1 t+2 t+3 t+4 t+5

High Debt Countries

C
ha
ng
e

in
d

eb
t r

at
io

(

p
er

ce
nt

o
f

G
D

P
)

9

permanent measures equivalent to 1 percent of GDP per year, starting in year t+1.16 As
shown above, the debt ratio is projected to decline from the first year in low-debt countries
and from the second year in medium-debt countries. However, in contrast to the previous
simulation, a notable decline may only occur in the third year in high-debt countries,
especially if the economy is experiencing a downturn. Incorporating an interest rate response
to take into account possible credibility effects does not significantly alter these results.17

The debt ratio eventually declines as the effect of past fiscal shocks on growth fades away
and their debt-reducing impact grows accordingly. This means that debt sustainability is not
affected by the multiplier effect. However, the slow response of the debt ratio could lead to
an increase in spreads if financial markets focus on its short-term behavior.

Figure 3. Impact on the Debt Ratio of a Discretionary Fiscal Tightening
of 1 Percent of GDP per Year over 5 Years

(Relative to Baseline)

Source: IMF Staff calculations.
Note: These simulations measure the change in the debt ratio relative to a baseline determined by 2011 WEO data and
constant thereafter. Cumulative multipliers: High is 1.3 in the first year, 1.6 in the second year; Downturn is 1 in the first
year, 0.8 in the second year; Low is 0.1 in the first year, 0.1 in the second year. Multipliers steadily decline to zero between
the second and the fifth year.

A combination of high and persistent multipliers, repeated tightening, and high debt would
make debt reduction more challenging. Previous simulations assume that the negative effect
of fiscal tightening on GDP fades away within 5 years. However, this assumption is debated
and fiscal consolidation could have a more persistent or even permanent effect on output
(i.e., GDP may stay durably below the baseline). Figure 4 shows the change in the debt ratio
under the assumption that multipliers remain constant instead of declining after the second
year. Persistence significantly raises the debt ratio response. For instance, if we use the
downturn multipliers, the fiscal adjustment necessary to bring the debt ratio down to a

16This could for instance consist in cutting expenditure by 1 percent of GDP in the first year, another percent in
the second year etc.

17We found broadly similar results under an alternative scenario with long-term interest rates decreasing by
50bp for each percentage point of GDP of fiscal consolidation (elasticities are derived from Haugh and others,
2009).

-14
-12
-10
-8
-6
-4
-2
0
2
4
t t+1 t+2 t+3 t+4 t+5
Medium Debt Countries
High multipliers Downturn multipliers Low multipliers
-14
-12
-10
-8
-6
-4
-2
0
2
4
t t+1 t+2 t+3 t+4 t+5
Low Debt Countries
-14
-12
-10
-8
-6
-4
-2
0
2
4
t t+1 t+2 t+3 t+4 t+5
High Debt Countries
C
ha
ng
e
in
d
eb
t r
at
io
(p
er
ce
nt
o
f
G
D
P
)

10

specific level after 5 years would be 70 percent larger in medium-indebted euro area
countries if shocks had permanent effects. These results point to a potentially important
effect that should be taken into account in the design of fiscal consolidation. However, these
are simplified calculations with an illustrative purpose. They ignore other key mechanisms,
such as the monetary policy response, the impact on growth of interest rate movements, and
credibility effects. To obtain a more complete assessment, a persistent shock scenario should
be simulated within a macroeconomic model.

Figure 4. Impact on the Debt Ratio of a Discretionary Fiscal Tightening
of 1 Percent of GDP per Year over 5 Years (With Persistent Effects on Output)

(Relative to Baseline)

Source: IMF Staff calculations.
Note: These simulations measure the change in the debt ratio relative to a baseline determined by 2011 WEO data and
constant thereafter. First-year multipliers: High multiplier is 1.3; Downturn multiplier is 1; Low multiplier is 0.1. Multipliers are
constant thereafter.

IV. EMPIRICAL EVIDENCE

This section presents empirical evidence supporting the conclusions drawn from the
simulations, namely, that fiscal consolidation can raise the public debt ratio in the short-term.
Our empirical analysis is based on comparing actual debt ratios with those that our
simulations would predict and on some simple correlations (section A); as well as on a
structural vector autoregression incorporating a debt feedback rule (section B).

A. Descriptive Analysis

A first rough test of the validity of the previous simulations is to see how they compare to
actual debt dynamics in advanced economies. Table 1 compares actual gross public debt
ratios in 2011 with simulation results in a sample of advanced economies that have carried
out large fiscal adjustments in 2010-11. Simulations assume that the effect of fiscal shocks
on GDP fades away within 5 years. Therefore, in order to calculate the debt ratio that our

-14
-12
-10
-8
-6
-4
-2
0
2
4
t t+1 t+2 t+3 t+4 t+5
High Debt Countries
C
ha
ng
e
in
d
eb
t r
at
io
(p
er
ce
nt
o
f
G
D
P
)
-14
-12
-10
-8
-6
-4
-2
0
2
4
t t+1 t+2 t+3 t+4 t+5
Medium Debt Countries

High multipliers Downturn m ultipliers Low m ultipliers

-14
-12
-10
-8
-6
-4
-2
0
2
4
t t+1 t+2 t+3 t+4 t+5
Low Debt Countries

11

model would predict in 2011, we have to start the simulation with actual 2007 data18. We set
the first year multiplier to 1 in all countries. While we recognize that multipliers are country
specific, this ensures that results are transparent and comparable across countries.
Simulations also assume that the underlying trend growth rate over 2008-11 is equal to the
2008 potential growth taken from the WEO database. We then add to this baseline the
cumulative effects of fiscal policy measures, which are proxied by the annual changes in the
structural primary balance (in percent of potential output) over the period.

Table 1. Actual versus Simulated Gross Public Debt Ratios in Selected
Advanced Economies in 2011

2007 Actual 2011 Actual 2011 Simulated
Greece 107.4 165.4 167.3
Iceland 29.1 99.2 93.2
Ireland 25.0 106.5 68.1
Portugal 68.3 107.8 92.0
Spain 36.3 69.1 61.6

Source: IMF Staff calculations and WEO database.
Note: The simulation uses actual data for 2007 as the starting point. We use a first year multiplier of 1, which slowly declines to
zero after five years. We assume that the underlying trend growth rate is 2008 potential growth. We then add to the baseline
the cumulative effects of discretionary fiscal policy, which we proxy by the annual changes in the structural primary balance
ratio. The structural balance is extracted from the WEO database.

Table 1 shows that 2011 actual and simulated gross public debt ratios are very close in those
countries where output fluctuations have been mainly driven by fiscal measures. Of course,
non-fiscal factors also explain debt dynamics, such as banking sector recapitalization
measures for example. These are particularly important in countries like Ireland, where
simulations fail to explain the large debt increases observed since 2007.19

As a second step to test the validity of the Section III findings, we investigate whether
episodes of discretionary tightening are associated with debt increases in a broader sample of
countries. We identify annual episodes of discretionary fiscal tightening in European (EU27)
and other G20 economies since 1980, using, as a criterion, a positive increase in the
structural primary balance. In general, the debt ratio does not decline when the structural
primary balance ratio improves from one year to the next. Instead, we find a (small) positive
correlation between fiscal consolidation and annual change in the debt ratio, as indicated by
the positive slope of 0.3 in Figure 5. This result is of course only illustrative, as correlation
does not entail causality and the data dispersion is quite large. Therefore, a more advanced
econometric analysis is warranted to determine whether a robust empirical relationship exists
between fiscal consolidation and the initial increase in the debt ratio.

18While most of the countries listed in Table 1 did not start consolidating before 2010, some introduced fiscal
stimulus measures in 2008 and 2009.

19In Portugal, a statistical reclassification required by Eurostat raised the debt stock by including some public
sector enterprises in the consolidated accounts of the public sector.

12

Figure 5. Correlation between Discretionary Fiscal Policy and Change in the
Debt Ratio in European and G20 economies (1980-2011) 1/

Source: IMF Staff calculations.
1/ The figure plots the annual changes in both variables for all years between 1980 and 2011.

B. Econometric Estimation

We investigate whether a structural VAR estimation provides further support to the
simulation results of Section III, taking into account that the government has to meet its
intertemporal budget constraint. This approach allows us to investigate the effect of fiscal
shocks on various macroeconomic variables, while keeping track of the debt dynamics. The
empirical analysis, which is based on Japanese data, only has an illustrative purpose. It shows
that, if multipliers and the debt-to-GDP ratio are high, it is possible to observe an initial
increase in the debt ratio in the aftermath of a fiscal consolidation.

General Approach

The VAR specification follows the setup of Favero and Giavazzi (2007, 2009). The main
difference between their approach and a more standard VAR analysis of fiscal policy lies in
the treatment of the debt-to-GDP ratio. Most studies do not include the public debt ratio in
the VAR specification, even though it is an important factor since taxes and spending are
likely to respond to the debt level (Bohn, 1998).20 Favero and Giavazzi show that the VAR
estimates are likely to be biased when this feedback effect is omitted. The estimated residuals
of the regression would include the response of other macroeconomic variables to the level of
debt. The regressors would be correlated with the error terms and therefore would not be
exogenous.

20 Chung and Leeper (2007) also estimate a structural VAR, explicitly incorporating an inter-temporal budget
constraint.

y = 0.3x – 0.3

-6
-4
-2
0
2
4
6

0 1 2 3

ǻ
D

eb
t-

to
-G

D
P

R
at

io

ǻ Structural primary balance as a share of potential GDP

13

Simple solutions that consist of either (i) omitting the debt variable but including in the VAR
all the variables that are part of the standard debt accumulation equation, or (ii) directly
including in the VAR the debt ratio as an endogenous variable, are not fully satisfactory. This
is because the VAR has a linear structure, while the debt dynamics equation is non-linear:
taxes, government spending, output, inflation and the interest rate are known to be related to
the debt ratio through an accounting identity (see equation (2) below). Favero and Giavazzi
(2007, 2009) therefore propose to introduce the debt ratio as an exogenous variable in the
VAR, and add a separate equation describing the debt accumulation. Box 1 provides details
on the estimation technique.

We apply this method to Japan and estimate the impact of fiscal policy on the gross debt-to-
GDP ratio. The choice of Japan is motivated by two considerations. First, Japan has
experienced high public debt ratios of at least 60 percent of GDP for several decades.
Second, as shown by the OECD (2009) and Baum, Weber and Poplawski-Ribeiro (2012),
Japan has comparatively high multipliers, exceeding 1, especially for spending, since it is a
relatively closed economy and interest rates have been at the zero lower bound since the late
1990s. In light of our simulation results, this combination suggests that the debt ratio should
initially increase following a discretionary fiscal consolidation. This is the hypothesis that we
test empirically.

Specification and Data

We estimate the following VAR including the ratio of gross public debt to GDP, ݀௧, as an
exogenous variable and explicitly modeling its dynamic with a separate equation:

௧ࢅ ൌ σ ௧ି௜ࢅ௜࡯ ൅ σ ௜݀௧ି௜ࢽ ൅ ௧௞௜ୀଵ௞௜ୀଵ࢛ (1)

݀௧ ൌ

ଵା௜೟
ሺଵାο௣೟ሻሺଵାο௬೟ሻ

݀௧ିଵ ൅ (௧ (2݀݌

where ࢅ௧ denotes a vector of endogenous variables, including real GDP (ݕ௧, in logarithms),
the average cost of servicing debt (݅௧, in percent), inflation21 (ο݌௧, in percent), and the
primary deficit in percent of GDP (݀݌௧).

Our analysis uses quarterly data for Japan between 1970Q1 and 2012Q1. Inflation is defined
as the change in the logarithm of the GDP deflator and the average cost of servicing debt is
obtained by dividing net interest payments by the total gross public debt stock at time t-1.
The primary balance is computed as total revenue minus total expenditure plus net interest
payments. We focus on the primary balance-to-GDP ratio instead of including expenditure
and revenue separately. This is because we want to model the effects of fiscal consolidation
on the debt ratio. If we included expenditures and revenues separately, revenues would

21 The VAR estimation takes into account the possible deceleration of inflation caused by the real GDP decline
(which is itself due to fiscal tightening).

14

evolve endogenously following a shock to expenditures (and vice versa) and, therefore, an
improvement in the overall balance could not be guaranteed in the short-term.22 All data are
obtained from the OECD Economic Outlook (No. 91 – June 2012).

22 In this case, the estimated relationship between revenue and spending would partly reflect historical patterns.
It could be that in the past, changes have offset each other, leaving the fiscal balance broadly unchanged.

Box 1: Calculation of Impulse Response Functions

Our objective is to estimate the impact of structural fiscal shocks on the variables entering the VAR and to
investigate how the debt ratio evolves along the path induced by these shocks. In practice, the structural fiscal
shocks are not directly observed since the matrix of reduced form residuals (࢛௧) in (1) contains: (i) the
automatic response of the fiscal balance to macroeconomic variables; and (ii) truly exogenous shifts in fiscal
policy࢚ࢋ�, which are the shocks we need to identify.1

We compute the impulse response functions following a shock to the primary deficit in four steps:

1. We estimate the VAR specified in equations (1), with ࢅ௧ ൌ ሺ݀݌௧ǡݕ௧ǡο݌௧ǡ ݅௧ሻ as the vector of
endogenous variables and including�݀௧ as an exogenous variable. Based on the Schwartz Information
criterion, two lags are included in this estimation.

2. The structural fiscal shocks ࢚ࢋ�are retrieved from the estimated VAR residuals ࢛௧�by employing the

structural identification methodology proposed by Blanchard and Perotti (2002). This method consists
in identifying the elements of the transformation matrices A and B relating structural shocks and
reduced-form VAR residuals:2
A࢚࢛, =B࢚ࢋ:

11
1

21 22
2

31 32 33
3

41 42 43 44

1 0 0 0 0
1 0 0 0 0 0

1 0 0 0 0
1 0 0 0

t

pd pd
pdy pd p t t

y
t t
p

t t
i
t t

a a bu e
a bu e
a a bu e
a a a bu e

‘

‘

§ · § ·§ · § ·
¨ ¸ ¨ ¸¨ ¸ ¨ ¸
¨ ¸ ¨ ¸¨ ¸ ¨ ¸
¨ ¸ ¨ ¸¨ ¸ ¨ ¸
¨ ¸ ¨ ¸¨ ¸ ¨ ¸¨ ¸ ¨ ¸

© ¹© ¹© ¹ © ¹

(3)

where ite (i=1,2,3) are the non fiscal shocks.

Some of the unknown elements of (3) can be determined without resorting to econometric estimation:

x Economic theory allows us to impose some exclusion restrictions (translating into zero

coefficients in the matrix). For example, we assume that the impact of an unexpected change in
interest rate on the primary deficit is zero.

x We also use some external information about the value of the elasticities of fiscal variables to

fluctuations in output and inflation. (i) The semi-elasticity of the budget deficit with respect to
output, (ߙ௣ௗ௬ሻ, is provided by the OECD (André and Girouard, 2005), and is estimated at -0.33 for
Japan. (ii) To our knowledge, the elasticity of the budget deficit with respect to inflation (ߙ௣ௗο௣ሻ
is not publically available. Therefore we had to estimate it, by regressing the change in the
revenue-to-GDP ratio or the change in the primary expenditure-to-GDP ratio on a constant, the
lagged debt-to-GDP ratio, the lagged primary balance-to-GDP ratio, real GDP growth and
inflation, as suggested by Marin (1998). The results show a negative significant impact of inflation

15

Results

The empirical results support the insights from our simulations.23 In Japan, the combination
of high debt and relatively large fiscal multipliers leads to an initial increase in the debt-to-
GDP ratio for about one and a half year following an exogenous fiscal consolidation.

Following a fiscal tightening of 1 percent of GDP, the debt-to-GDP ratio initially rises by
about 1 percentage point (Figure 6).24 It takes about 6 quarters until the debt-to-GDP ratio

23 Our empirical results are subject to caveats, including issues related to the predictability of structural shocks,
since economic agents may receive news about future fiscal measures. Leeper and others (2012) show that such
“news shocks” are particularly relevant for tax measures, because the process of changing taxes is subject to
long lags. This may result in incorrect identification of structural shocks. Since this section merely serves as an
illustration of our simulation results, we abstract from these considerations.

Box 1: Calculation of Impulse Response Functions (concluded)

on the primary expenditure-to-GDP ratio of 0.2 but no significant influence on the revenue-to-
GDP ratio. We therefore set the elasticity of the primary deficit-to-GDP to inflation at -0.2. Given
the uncertainty surrounding this estimate, we conduct a sensitivity analysis with elasticities
ranging from 0 to -0.4.

The remaining coefficients of the matrixes are estimated using a Cholesky decomposition (but without
the debt ratio since this does not enter the identification problem). The estimated variance-covariance
matrix of the four equation VAR innovations contains 10 different elements and with 10 parameters to
be estimated, the model is just-identified.

3. We use the results from both the identification problem in (3) and the coefficient estimates from the

estimation of (1), (࡯௜, ࢽ௜), to compute predictions of the variables, based on a no shock scenario and a
scenario in which an exogenous fiscal tightening of 1 percent is implemented in period 1. The third
step involves solving dynamically forward the identified system under the two scenarios ( 1 1

pde �
and 1 0) :

pde

௧ࢅ ൌ σ ௧ି௜ࢅ௜࡯ ൅ σ ௧ି௜݀࢏ࢽ ൅ ௞௜ୀଵ௞௜ୀଵ࢚ࢋ࡮ଵି࡭ (4)

݀௧ ൌ
ଵା௜೟

ሺଵାο௣೟ሻሺଵାο௬೟ሻ
݀௧ିଵ ൅ (௧ (5݀݌

A and B are the solution of the structural identification problem in (3) and (࡯௜, ࢽ௜) are the coefficient
estimates of (1) with two lags.

4. We compute the difference between the two scenarios, which corresponds to the impulse response

functions.

1As in Blanchard and Perotti (2002) and given the quarterly frequency of our data, we assume that there is no discretionary response
of fiscal policy to news in macroeconomic variables, since this response typically takes longer than a quarter.
2The assumptions regarding the interaction of ݑ௧௜ with other variables in Matrix A follow Perotti (2008) and Favero and Giavazzi
(2007).

16

begins to decline. The response in the debt ratio can be explained by the evolution of the
endogenous variables following the fiscal shock. Inflation and real output immediately fall,
offsetting about half of the decrease in the primary deficit-to-GDP ratio (flow effect). In
addition, the decline in output also raises the debt ratio because of the denominator decrease
(stock effect).

Figure 6. Response of Macro Variables to 1 percent of GDP
Discretionary Fiscal Tightening

Source: IMF Staff calculations.

Given the uncertainty surrounding the inflation elasticity estimate, we perform a robustness
analysis as indicated above. Figure 7 compares the evolution of the debt-to-GDP ratio
following a 1 percent fiscal tightening under different assumptions about the elasticity of the
primary fiscal deficit to inflation (0, -0.2, and -0.4). The higher the elasticity of the primary
deficit, the longer it takes until the debt ratio eventually declines. This is intuitive since the
initial tightening leads to a fall in inflation, which increases the fiscal deficit, with this
response being stronger if the elasticity is larger. When the elasticity is -0.4, it takes 2.5 years
until the debt ratio eventually declines following the discretionary fiscal tightening. When the
elasticity is zero, the debt ratio starts declining after the third quarter.

24 This empirical estimate is consistent with previous simulation results. With a revenue ratio of 30 percent of
GDP, a public debt ratio of 100 percent of GDP (corresponding to the 1970-2011 averages), and a spending
multiplier of 1.5, the formula of Section II.B. predicts that a fiscal shock of 1 percent of GDP would initially
increase the debt ratio by about 1 percentage point in the first year.

Ͳ0.5

-0.4

-0.3

-0.2

–

0.1

0
0.1

0 2 4 6 8 10 12 14 16 18

20

p
er

ce
n

ta
ge

p
o

in
ts

Quarters

Primary Fiscal Deficit (in percent of GDP)
Real GDP
Inflation
Average Cost of Servicing Debt

-9
-8
-7
-6
-5
-4
-3
-2
-1
0
1
2

0 2 4 6 8 10 12 14 16 18 20
Quarters

Public Debt-to-GDP Ratio

17

Figure 7. Sensitivity Analysis: Response of the Debt Ratio with Alternative
Parameters

Source: IMF Staff calculations.

V. IMPLICATIONS FOR THE DEBT RATIO INDICATOR

This section examines whether there is scope for using cyclically-adjusted debt ratios for
monitoring purposes and/or as fiscal targets, particularly for high-debt countries.

A. Cyclicality of Nominal Debt Ratios

The nominal debt ratio is a highly cyclical fiscal indicator. Figure 8 illustrates the evolution
of the debt and fiscal balance ratios over the cycle (under the assumptions described below
the chart). Interestingly the cyclicality of the debt ratio is far more pronounced than that of
the overall balance. In our example, the debt ratio oscillations are about twice as large as
those of the overall balance ratio (7 compared to 3 percent of GDP). The reason is that a
decline in output raises the debt ratio not only because it deteriorates the fiscal balance, but
also because it increases the initial debt-to-GDP ratio (“scaling factor”).

A simple equation illustrates this point. The changes in the debt ratio overtime are amplified
by a factor proportional to the size of the initial debt ratio. One implication is that the cyclical
fluctuations of the debt ratio are likely to be wider in high debt countries.

οሺ����������୲ሻ ൌ ��ϐ����������୲ െ
�୲

ͳ ൅ �୲
כ ����������୲ିଵʹͷ

25 �୲ denotes nominal GDP growth . ¨ denotes a change over time; οܺ ൌ ܺ௧ െ ܺ௧ିଵ.

-16

-14
-12
-10
-8
-6
-4
-2
0
2

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20

p
er
ce
n
ta
ge
p
o
in
ts
Quarters

Inflation elasticity of 0

.2

Inflation elasticity of 0.4

Inflation elasticity of 0

18

֞ οሺ����������୲ሻ െ οሺ��ϐ����������୲ሻ ൎ ��ϐ����������୲ିଵ െ �୲ כ ����������୲ିଵᇣᇧᇧᇧᇧᇧᇧᇧᇧᇧᇧᇤᇧᇧᇧᇧᇧᇧᇧᇧᇧᇧᇥ
ୟ୫୮୪୧ϐ୧ୡୟ୲୧୭୬�୤ୟୡ୲୭୰

When public finances deteriorate, it is likely that ��ϐ����������୲ିଵ ൐ �୲ כ ����������୲ିଵ, at least
at the lowest point of the economic cycle. Thus, οቀ����
��ൗ ቁ ൐ οቀ��ϐ����
��ൗ ቁ and the debt
ratio increase exceeds that of the deficit ratio. Conversely, when fiscal performance
improves, it is likely that ��ϐ����������୲ିଵ ൏ �୲ כ ����������୲ିଵ, at least at the peak of the
economic cycle. Thus, οቀ����
��ൗ ቁ ൏ οቀ��ϐ����
��ൗ ቁ and the debt ratio decline is larger than
that of the deficit ratio.

Figure 8. Debt and Deficit Ratios over the Cycle

Source: IMF Staff Estimates.
Hypotheses: (i) Potential growth is 3 percent per year (in nominal terms); (ii) the fiscal position is
initially balanced in period 0; (iii) the output gap oscillates in the -5 /+5 percent range; (iv) no change
in the structural stance; the revenue-to-GDP ratio is constant; spending grows with potential output;
there is a no cyclical component in spending; and (v) the initial debt ratio is 60 percent of GDP.

In general, debt ratios are used as medium-term fiscal anchors; they are not meant to inform
short-term fiscal policy. Indeed, debt rules do not provide sufficient guidance for fiscal
policy when debt is well below its ceiling. In addition, the link between the fiscal stance and
public debt is not stable since debt dynamics are affected by exogenous and/or volatile
factors, such as the exchange rate, the global component of the sovereign risk premium, and
off-budget operations.

Despite the problems with using the nominal debt ratio as a short-term fiscal target, in the
post-crisis period, a number of advanced economies have begun to so. Indeed, debt reduction
is increasingly perceived as a pressing objective in order to restore market confidence and
fiscal sustainability. European countries are probably the most striking examples of this new
approach, with the 1/20th rule setting annual debt reduction objectives for countries in breach
of the 60 percent ceiling (European Commission, 2011 and 2012).

-10
0
10
20

30

40

50

60

70

0 1 2 3 4 5 6 7 8 9 10

Fiscal Balance (in percent of GDP)

Public Debt (in percent of GDP)

Output gap (in percent)

19

B. Towards Cyclically-Adjusted Debt Indicators?

In light of recent trends towards using the debt ratio as a short-term fiscal target, the previous
results have two important implications for fiscal policy:

Looking backward, the use of the nominal debt ratio for monitoring purposes may be
problematic, as this ratio is blurred by considerable cyclical “noise.” In the same way as the
structural fiscal balance strips away the part of the deficit related to the economic cycle to
isolate the fiscal stance, a cyclically-adjusted debt ratio (CADR) could be calculated to filter
out debt ratio fluctuations, and isolate underlying trends. In addition, exploring the
discrepancies between the past evolutions of the nominal debt ratio and the CADR could be
informative.

Looking forward, using the nominal debt ratio as an operational fiscal target poses three
practical difficulties:

x First, targeting a nominal debt ratio would result in a highly procyclical fiscal
stance.26 In fact, procyclicality is likely to be stronger with a nominal debt rule than
with a nominal budget balance rule. In downturns, a nominal debt rule requires
additional fiscal effort to offset the cyclical deterioration in the budget, but also to
compensate for the “scaling effect” of the initial debt.

x Second, even if the government was willing to pursue a procyclical fiscal stance to
bring the debt ratio to its target, it may not be able to do so in the short-term because
of the fiscal multiplier effect. Previous sections have shown that there is not a one-
for-one relation between discretionary fiscal measures and changes in the nominal
debt-to-GDP ratio. Fiscal tightening, for instance, has generally a small or even an
adverse immediate impact on debt. Therefore targeting the nominal debt ratio would
be either very costly in terms of discretionary measures, or even impossible in the
short-term.27

x Third, debt targeting could be self-defeating: if authorities focus on the short-term
behavior of the nominal debt ratio, they may engage in repeated rounds of tightening
in an effort to get the debt ratio to converge to the official target, undermining
confidence, and setting off a vicious circle of slow growth, deflation, and further
tightening.

26Targeting the nominal debt ratio is only one of the many reasons for which fiscal policy can be procyclical.
For instance, some governments are forced to follow such course of action, because they have lost market
access.

27 The size of the required discretionary tightening needed to bring down the debt ratio increases when fiscal
multipliers and/or the initial debt ratio are larger.

20

In light of these three issues, policymakers could consider setting debt targets in cyclically-
adjusted terms. This would mitigate the procyclical bias of nominal debt ratios by letting
automatic stabilizers operate, while still constraining the discretion of fiscal policy.

However, measuring the CADR raises several practical and conceptual difficulties (see
Appendix 2 for more details). The CADR level is sensitive to the start date, as the calculation
starts from an initial debt in nominal terms and adds the subsequent cyclically-adjusted
deficits. It can be demonstrated that direction and rate of change (i.e., the slope) of the CADR
is more informative than its level. In addition, several variants of the indicator can be
contemplated, depending on whether the numerator and/or the denominator of the ratio are
corrected for the cycle. Appendix 2 shows that the simplest ratio consisting in dividing
nominal debt by potential GDP is not the most useful indicator. Also, like a structural
balance, calculating a CADR requires measuring the output gap, which is a tricky task.
Measurement errors are likely to be larger than with a structural balance rule, because errors
cumulate in the CADR numerator (if they all go in the same direction, which may be the case
if potential output is misestimated).

Another question is whether the CADR improves upon the existing cyclically-adjusted
balance both in terms of monitoring and objective setting. If a country follows a fiscal rule
based on the latter, shouldn’t a rule based on the former also be met? There is no simple
answer to this question, although two elements warrant the use of the CADR. First, changes
in the CADR are not only due to the cyclically-adjusted fiscal stance. As explained above,
the CADR also depends on potential growth, and on a scaling factor related to the size of the
initial debt ratio. Therefore, the cyclically-adjusted balance alone does not contain enough
information to infer the path of the CADR. Second, the debt ratio has become an operational
target of fiscal policy in many advanced economies. Correcting the debt ratio may be
necessary to make the fiscal balance and debt targets consistent.28 Otherwise, the procyclical
fiscal stance induced by the nominal debt target would conflict with the automatic
stabilization pursued by many countries with cyclically-adjusted balance rules (or
expenditure rules).

VI. CONCLUSIONS

This paper examines the effect of fiscal consolidation on the debt ratio. It first assesses the
importance of the fiscal multiplier assumption in the nexus between fiscal consolidation,
growth, and debt reduction. With multipliers close to 1 in the current environment, fiscal
consolidation is likely to raise the debt ratio in the short-run in most advanced countries. We
provide empirical evidence supporting this hypothesis. The slow response of the debt ratio to
fiscal adjustment could be an issue if financial markets focus on its short-term behavior.

28 Most countries pursue both debt and fiscal balance targets (see the EU fiscal governance framework, for
instance).

21

Our analysis suggests three main operational conclusions, which are particularly relevant for
Europe today. First, underestimating fiscal multipliers may cause unpleasant surprises. As
fiscal consolidations generally take place in a depressed economic environment with
relatively high multipliers, fiscal tightening may initially raise the debt ratio. Not explicitly
taking into account multipliers or underestimating their value could lead authorities to set
unachievable debt (and deficit) targets and miscalculate the amount of adjustment necessary
to curb the debt ratio. Missing announced targets could impact the credibility of adjustment
programs and increase uncertainty about the future path of fiscal policy.

Second, using the debt ratio as an operational fiscal target presents risks. If country
authorities focus on the short-term behavior of the debt ratio, they may engage in repeated
rounds of tightening in an effort to get the debt ratio to converge to the official target,
undermining confidence, and setting off a vicious circle of slow growth, deflation, and
further tightening. A possible solution could be to monitor debt ratios and set debt targets in
cyclically-adjusted terms, though, as explained in Section V, using the CADR for this
purpose can entail certain difficulties as well.

Third, an appropriate design of consolidation packages can minimize adverse loops involving
fiscal tightening and short-term debt dynamics. As Cottarelli and Jaramillo (2012) highlight,
in many countries, the composition of fiscal adjustment can be rebalanced to make it more
“growth friendly.” Setting the right pace of consolidation is also important: if financing
allows, a more gradual approach is preferable, since adjustment measures can be taken when
the economy recovers and multipliers are lower. For example, Bagaria and others (2012)
simulate scenarios with alternative timings for the UK government’s fiscal consolidation
plans. The results indicate that delaying the consolidation effort until more normal economic
conditions prevail would substantially lessen the size and duration of the fiscal adjustment’s
impact on growth. Finally, structural reforms can also contribute to breaking adverse loops.
While their benefits usually take time to materialize, there is evidence that some structural
reforms deliver gains already in the short run, and can boost growth relatively quickly
(OECD, 2012).

Our results should be interpreted with caution. In particular, they do not imply that fiscal
consolidation is undesirable or could place public debt on an unsustainable path. Almost all
advanced economies face the challenge of fiscal adjustment in response to elevated
government debt levels and future pressures on public finances from demographic change.
The short-term effects of fiscal policy on economic activity are only one of the many factors
that need to be considered in determining the appropriate pace of fiscal consolidation. In
addition, multipliers differ across countries and time. In some cases, confidence effects may
partly offset the negative impact of fiscal tightening on growth.

22

APPENDIX 1: IMPACT OF FISCAL CONSOLIDATION ON THE DEBT RATIO

1. Impact of a one-off Permanent Fiscal Tightening on the Debt Ratio

If fiscal policy is tightened by 1 percent of GDP in the first year and the tightening is
permanent, the debt ratio increases relative to the baseline29 by:

ο൭����୒ �୒ൗ ൱ כ ͳͲͲ ൎ െ� ൅ ݐܾ݁݀ ே݋݅ݐܽݎ כ ேݐ݈ݑ݉ ൅ ݒ݁ݎ ݋݅ݐܽݎ כ ෍݉ݐ݈ݑ௜
ே

௜ୀଵ

with ݉ݐ݈ݑ௜�denoting the N-year multiplier, and ����୒ the debt stock at the end of period N.

In the case of a tightening larger than 1 percent of GDP, the formula should be multiplied by
the size of the adjustment.

The following sections demonstrate this result.

a. Impact of GDP Revisions on the Deficit and Debt Ratios

Assumptions

– �οܻ ܻൗ ൌ െͳΨ

– ο൫ܴ݁ݒ ܻൗ ൯ ൌ Ͳ; the elasticity of revenue with respect to GDP is 1.

– ο݌ݔܧ ൌ Ͳ

Effect on the deficit ratio

ο൭݂݀݁݅ܿ݅ݐ ܻൗ ൱ ൌ ο൭
݌ݔܧ

ܻൗ ൱ െ ο൫ܴ݁ݒ ܻൗ ൯ ൌ ο൭
݌ݔܧ

ܻൗ ൱

ൌ ൭݌ݔܧ ܻൗ ൱Ǥ൭
ο݌ݔܧ

ൗ݌ݔܧ െ οܻ ܻൗ ൱

֜ ο൭݂݀݁݅ܿ݅ݐ ܻൗ ൱ ൌ െ൭
݌ݔܧ

ܻൗ ൱Ǥ൫οܻ ܻൗ ൯ (1)

29 In this appendix, ο refers to a change relative to baseline (not relative to the previous period).

23

Effect on the debt ratio

οቀ���� �ൗ ቁ ൌ οቀ
ୢୣୠ୲షభାୢୣϐ୧ୡ୧୲

ଢ଼ ቁ ൌ ����ିଵǤο൫
ͳ �ൗ ൯ ൅ οቀ��ϐ���� �ൗ ቁ

= െቌ൭ܾ݀݁ିݐଵ ܻൗ ൱ ൅ ൭
݌ݔܧ

ܻൗ ൱ቍ൫οܻ ܻൗ ൯� given (1) and ο൫ͳ �ൗ ൯ ൌ െοܻ ܻଶൗ

֜οቀ���� �ൗ ቁ�=�െ൬ቀܾ݀݁ݐ ܻൗ ቁ ൅ ൫ܴ݁ݒ ܻൗ ൯൰൫οܻ ܻൗ ൯ (2)

b. Impact of a One-Off Fiscal Tightening on the Debt Ratio

Assumptions

– A 1 percent of GDP fiscal tightening takes place in year 1, in the form of a permanent
spending cut; there is no further adjustment in subsequent periods:

o ο݌ݔܧଵ ൌ ο݌ݔܧଶ ൌ ڮ ൌ ο݌ݔܧே ൌ െ ଵܻ ͳͲͲ

ൗ

– The elasticity of revenue with respect to GDP is 1:

o ܴ݁ݒଵ
ଵܻ

ൗ ൌ ଶݒܴ݁
ଶܻ

ൗ ൌ ڮ ൌ ேݒܴ݁
ேܻ

ൗ �

o ο൭ܴ݁ݒଵ
ଵܻ

ൗ ൱ ൌ ο൭ܴ݁ݒଶ
ଶܻ

ൗ ൱ ൌ ڮ ൌ ο൭ܴ݁ݒே
ேܻ

ൗ ൱ ൌ Ͳ�

– Multipliers:30

o ݉ݐ݈ݑଵ ൌ ο ଵܻ ο݌ݔܧଵൗ ֜
ο ଵܻ

ଵܻ
ൗ ൌ െ ଵଵ଴଴ Ǥ݉ݐ݈ݑଵ

and οܴ݁ݒଵ ൌ ο ଵܻ ଵܻൗ Ǥܴ݁ݒଵ= -�݉ݐ݈ݑଵǤ
ଵݒܴ݁ ͳͲͲൗ

o ݉ݐ݈ݑே ൌ ο ேܻ ο݌ݔܧଵൗ ֜
ο ேܻ

ேܻ
ൗ ൌ െ ଵଵ଴଴ Ǥ݉ݐ݈ݑேǤ

ଵܻ
ேܻ

ൗ

30 Fiscal multipliers are defined as the ratio of a change in output to an exogenous change in the fiscal deficit
with respect to their respective baselines. In our definition, multipliers are cumulative.

24

and οܴ݁ݒே ൌ ο ேܻ ேܻൗ Ǥܴ݁ݒே= -�݉ݐ݈ݑேǤ
ଵܻ

ேܻ
ൗ Ǥܴ݁ݒே ͳͲͲൗ ͵ͳ

First year

ο൭����ଵ �ଵൗ ൱ ൌ οቀ
ୢୣୠ୲బାୢୣϐ୧ୡ୧୲భ

ଢ଼భ
ቁ ൌ ����଴Ǥοቀͳ �ଵൗ ቁ ൅ ο൭

��ϐ����ଵ �ଵൗ ൱

ൌ�െ൭����଴ �ଵൗ ൱Ǥ൭
ο�ଵ �ଵൗ ൱ ൅ ο൭

ଵ݌ݔܧ �ଵൗ ൱���

= െ൭ܾ݀݁ݐଵ �ଵൗ െ
ଵ݌ݔܧ �ଵൗ ൅

ଵݒܴ݁ �ଵൗ ൱Ǥ൭
ο�ଵ �ଵൗ ൱ +�൭

ଵ݌ݔܧ �ଵൗ ൱Ǥ൭
ο݌ݔܧଵ ଵൗ݌ݔܧ െ

ο�ଵ �ଵൗ ൱

= െ൭ܾ݀݁ݐଵ �ଵൗ െ
ଵ݌ݔܧ �ଵൗ ൅

ଵݒܴ݁ �ଵൗ ൱Ǥ൭
ο�ଵ �ଵൗ ൱ െ

ͳ ͳͲͲൗ െ ൭
ଵ݌ݔܧ �ଵൗ ൱Ǥ൭

ο�ଵ �ଵൗ ൱�

֜ ο൭����ଵ �ଵൗ ൱ כ ͳͲͲ ൌ�െͳ ൅�݉ݐ݈ݑଵǤ൭
����ଵ �ଵൗ ൅

ଵݒܴ݁ �ଵൗ ൱

Intuitively, the impact of a fiscal consolidation on the debt ratio combines two effects:

– Direct effect: absent the multiplier effect, a fiscal consolidation of one percent of
GDP should decrease the debt ratio by 1 percentage point.

– Multiplier effect: the direct effect is (partly) offset by the lower GDP which increases
the debt ratio according to equation (2). The multiplier effect is channeled through
two terms: the effect of a lower GDP on the debt ratio (captured by the debt ratio) and
that of automatic stabilizers (captured by the revenue ratio).

Year N

ο൭����୒ �୒ൗ ൱ ൌ ο൬
����଴ ൅ ��ϐ����ଵ൅��ϐ����ଶ൅ڮ൅ ��ϐ����୒

�୒
൰

ൌ ����଴Ǥοቀͳ �୒ൗ ቁ ൅ ο൭
��ϐ����ଵ �୒ൗ ൱ ൅ ൅ڮ ο൭

��ϐ����୒ �୒ൗ ൱

31 ேܻ is only affected by the initial shock (with a N-year multiplier), as there are no further shocks in subsequent
years.

25

= െ൭����୒ �୒ൗ െ
��ϐ����ଵ �୒ൗ െ

��ϐ����ଶ �୒ൗ ǥെ
��ϐ����୒ �୒ൗ ൱.൭

ο�୒ �୒ൗ ൱

+���ϐ����ଵ �୒ൗ Ǥ൭
ο��ϐ����ଵ

��ϐ����ଵൗ െ
ο�୒ �୒ൗ ൱�+ǥ൅

��ϐ����୒ �୒ൗ Ǥ൭
ο��ϐ����୒

��ϐ����୒ൗ െ
ο�୒ �୒ൗ ൱

=�െ൭����୒ �୒ൗ ൱.൭
ο�୒ �୒ൗ ൱�+ቆ

ሺο݌ݔܧଵ ൅ ൅ڮ ο݌ݔܧேሻ �୒ൗ ቇ െ ቆ
ሺο���ଵ ൅ ൅ڮ ο���୒ሻ �୒ൗ ቇ

= ����୒ �୒ൗ כ
ଵܻ

ேܻ
ൗ Ǥ݉ݐ݈ݑே ͳͲͲൗ െ ଵܻ ேܻൗ Ǥ

ܰ ͳͲͲൗ +�
ଵݐ݈ݑ݉ ͳͲͲൗ Ǥ

ଵݒܴ݁ �ଵൗ Ǥ
�ଵ �୒ൗ

ଶݐ݈ݑ݉�+ ͳͲͲൗ Ǥ
ଶݒܴ݁ �ଶൗ Ǥ

�ଵ �୒ൗ +…+
ேݐ݈ݑ݉ ͳͲͲൗ Ǥ

ேݒܴ݁ �୒ൗ Ǥ
�ଵ �୒ൗ

֜ ο൭����୒ �୒ൗ ൱ כ ͳͲͲ ൌ
ଵܻ

ேܻ
ൗ Ǥ൭െ� ൅ ே݋݅ݐܽݎ�ݐܾ݁݀ כ ேݐ݈ݑ݉ ൅ ݋݅ݐܽݎ�ݒ݁ݎ כ ෍݉ݐ݈ݑ௜

ே

௜ୀଵ
൱

ൎ െ� ൅ ே݋݅ݐܽݎ�ݐܾ݁݀ כ ேݐ݈ݑ݉ ൅ ݋݅ݐܽݎ�ݒ݁ݎ כ σ ௜ே௜ୀଵݐ݈ݑ݉ if ଵܻ ேܻൗ ൎ ͳ (3)

Comment 1

By dividing both sides of equation (3) by N and as N tends to infinity,

���୒՜ஶ ο൭����୒ �୒ൗ ൱Ȁ� ൌ ͳ�ሺ���������������������������ሻǤ��This means that, after

some time, the multiplier effect should eventually vanish, and the debt ratio response solely
reflects the direct consolidation effect.

Comment 2

If fiscal consolidation permanently lowers GDP, the multiplier does not decrease over time.32
For instance, if the multipliers are constant after the first year, ݉ݐ݈ݑ௜ ൌ ݅׊ଵǡݐ݈ݑ݉ ൒ ʹ�.
Then:

ο൭����୒ �୒ൗ ൱ כ ͳͲͲ ൎ െ� ൅ ே݋݅ݐܽݎ�ݐܾ݁݀ כ ଵݐ݈ݑ݉ ൅ ܰ כ ݋݅ݐܽݎ�ݒ݁ݎ כ (ଵ (4ݐ݈ݑ݉

32 The impact of a permanent tightening on GDP could well be temporary. Equation (4) assumes that a
permanent consolidation has a permanent output effect.

26

2. Impact of Repeated Tightening on the Debt Ratio

In case of repeated tightening, the formula becomes:

ο൭����୒ �୒ൗ ൱ כ ͳͲͲ ൎ െ�ሺ� ൅ ͳሻȀʹ ൅ ே݋݅ݐܽݎ�ݐܾ݁݀ כ ෍݉ݐ݈ݑ௜
ே

௜ୀଵ
൅ ݋݅ݐܽݎ�ݒ݁ݎ כ ෍ሺܰ െ ݅ ൅ ͳሻ כ ௜ݐ݈ݑ݉

ே
௜ୀଵ

Assumptions

– Spending is permanently cut by 1 percent of GDP in year 1 relative to baseline, then
reduced by another percent of GDP in year 2 etc.

o ο݌ݔܧଵ ൌ െ ଵܻ ͳͲͲൗ ; ο݌ݔܧଶ ൌ െ ଵܻ ͳͲͲൗ െ ଶܻ ͳͲͲൗ ; etc.

– The elasticity of revenue with respect to GDP is 1.

– Multipliers:

o ݉ݐ݈ݑଵ ൌ ο ଵܻ ο݌ݔܧଵൗ ֜
ο ଵܻ
ଵܻ
ൗ ൌ െ ଵଵ଴଴ Ǥ݉ݐ݈ݑଵ

and οܴ݁ݒଵ ൌ ο ଵܻ ଵܻൗ Ǥܴ݁ݒଵ= െ�݉ݐ݈ݑଵǤ
ଵݒܴ݁ ͳͲͲൗ

o ݉ݐ݈ݑଶ ൌ ο ଶܻ
ᇱ
ο݌ݔܧଵ൘ (reflecting the impact on ଶܻ�of the first-period shock, but

excluding the effect of the second-period shock).

o ο ଶܻ ൌ ଶ݌ݔܧଵǤሺοݐ݈ݑ݉ െ ο݌ݔܧଵሻ ൅ ଵ. Indeed, ଶܻ�is affected by݌ݔܧଶǤοݐ݈ݑ݉
both the first- and the second-period shocks.

ο ଶܻ
ଶܻ

ൗ ൌ െ ଵଵ଴଴ Ǥ݉ݐ݈ݑଵ െ
ଵ
ଵ଴଴ Ǥ݉ݐ݈ݑଶǤ

ଵܻ
ଶܻ

ൗ , and οܴ݁ݒଶ ൌ ο ଶܻ ଶܻൗ Ǥܴ݁ݒଶ ൌ

െ൭�݉ݐ݈ݑଵ ൅�݉ݐ݈ݑଶǤ ଵܻ ଶܻൗ ൱Ǥ
ଶݒܴ݁ ͳͲͲൗ

27

Year 2

ο൭����ଶ �ଶൗ ൱ ൌ ο൬
����଴ ൅ ��ϐ����ଵ൅��ϐ����ଶ

�ଶ
൰

ൌ ����଴Ǥοቀͳ �ଶൗ ቁ ൅ ο൭
��ϐ����ଵ �ଶൗ ൱ ൅ ο൭

��ϐ����ଶ �ଶൗ ൱

= െ൭����ଶ �ଶൗ െ
��ϐ����ଵ �ଶൗ െ

��ϐ����ଶ �ଶൗ ൱.�൭
ο�ଶ �ଶൗ ൱ +�

��ϐ����ଵ �ଶൗ Ǥ൭
ο݂݀݁݅ܿ݅ݐଵ

ଵൗݐ݂݅ܿ݅݁݀ െ

ο�ଶ �ଶൗ ൱�+�
��ϐ����ଶ �ଶൗ Ǥ൭

ο݂݀݁݅ܿ݅ݐଶ
ଶൗݐ݂݅ܿ݅݁݀ െ

ο�ଶ �ଶൗ ൱

=�െ൭����ଶ �ଶൗ ൱.�൭
ο�ଶ �ଶൗ ൱�+ቆ

ሺο݌ݔܧଵ ൅ ο݌ݔܧଶሻ �ଶൗ ቇ െ ቆ
ሺοܴ݁ݒଵ ൅ οܴ݁ݒଶሻ �ଶൗ ቇ

֜ ο൭����ଶ �ଶൗ ൱ כ ͳͲͲ ൎ െ͵ ൅ ଶ݋݅ݐܽݎ�ݐܾ݁݀ כ ሺ݉ݐ݈ݑଵ ൅ ଶሻݐ݈ݑ݉ ൅ ݋݅ݐܽݎ�ݒ݁ݎ כ ሺʹǤ݉ݐ݈ݑଵ ൅ �ଶሻݐ݈ݑ݉

if ଵܻ
ଶܻ

ൗ ൎ ͳ

This calculation can easily be extended to subsequent years.

28

APPENDIX 2: THE CYCLICALLY-ADJUSTED DEBT RATIO: DEFINITION AND MEASUREMENT

This appendix analyzes some practical issues raised by the concept of cyclically-adjusted
debt ratio (CADR).

What is a CADR?

If it was possible to go sufficiently far back in time, it would be possible to compute a CADR
as the sum of all past cyclically-adjusted deficits divided by potential GDP. Given that it is
not feasible to do so, the concept is more modest. It starts from an initial debt level (in
nominal terms), and adds the cyclically-adjusted deficits since the start date.33 As such, the
CADR is a counterfactual series, describing the path the debt ratio would have taken, had
GDP been growing at potential between ݐ െ ܰ and ݐ.

����୲ ൌ ൫������������୲ି୒ ൅ σ �����ϐ����୧୧ୀ୲୧ୀ୲ି୒ାଵ ൯ ����������
��୲Τ �, (1)

with ݐ െ ܰ being the starting year for the calculation.

Does the choice of the start date matter? What is most informative: the level or the slope of
the CADR?

The CADR should not be understood as a measure of “structural debt” in the sense given to
the “structural balance.”34 Indeed there is not a structural debt ratio around which debt would
oscillate. There are alternative CADRs depending on the initial debt ratio chosen to compute
the series. Appendix Figure 1 plots CADRs using alternative start dates, calculated with
formula (1). The chart shows that the level of the CADR depends on initial conditions, and is
not, in itself, very informative.

What is informative is the trend of the CADR. As is the case with a price index, we could say
that the CADR is most usefully interpreted in terms of changes, not in levels. As illustrated
by Appendix Figure 1, the choice of the start date has only a marginal effect on the CADR
slope. The reason is that the CADR dynamics are mainly driven by the structural fiscal stance
and potential growth—two factors that are unrelated to the starting point of the series.35 A
sensitivity analysis confirms that the CADR is not too sensitive to the start date.36 Thus, there

33 For instance, the CADR proposed by the European Commission only corrects the numerator and the
denominator for the cyclical developments over the period t-3 to t (EC, 2011).

34 This point is also noted by European Commission (2011).

35 ����୲�is affected by the start date but the effect is unlikely to be large. Indeed, the only difference between
two alternative CADRs in period t is the sum of the cyclical deficits between their respective start dates, and
this sum is bounded, as cyclical deficits eventually zero out over a full cycle.

36 We used different assumptions on potential growth, discretionary fiscal policy, and initial fiscal conditions.

29

seems to be little room for manipulating the CADR by choosing the date associated with the
most favorable debt path.37

Appendix Figure 1. Alternative Cyclically-Adjusted Debt Ratios

Source: IMF Staff Estimates.
Note: The nominal debt ratio is the ratio of nominal debt to nominal
GDP. The CADR measures are in percent of potential GDP.
Hypotheses are similar to Figure 8.

Is it enough to calculate the nominal debt-to-potential GDP ratio?

Several variants of the indicator could be contemplated. Instead of correcting both the
numerator and the denominator for the cycle, as done in (1), a short-cut could be taken by
correcting only one of them. For instance, a simple measure of the CADR could be the ratio
of nominal debt to potential GDP.

Box 2 simulates alternative CADRs over the cycle. We find that the cyclicality of the debt
ratio mostly comes from the numerator, where the cyclical components of past fiscal
balances build up. Correcting the denominator seems to be less critical. Consequently,
calculating the ratio of nominal debt stock to potential GDP is not sufficient to strip away the

37 Nonetheless, selecting a start date when the output gap is zero would result in the CADR presenting an
interesting feature. In this case, the nominal debt ratio and the CADR would coincide at the beginning of each
cycle—a property that would bring the CADR closer to the cyclically-adjusted balance concept. Appendix
Figure 1 provides a visual illustration of this point.

40
50
60
70
0 1 2 3 4 5 6 7 8 9 10

Nominal Debt Ratio
CADR starting in period 0
CARD Starting in Period 1
CADR Starting in Period 2
CADR Starting in Period 3
CADR Starting in Period 4

30

part of the debt related to the economic cycle.

Is there a rule of thumb for the CADR?

There is no simple formula equivalent to the one existing for the cyclically-adjusted
balance.38 To quickly assess the CADR, Box 2 proposes a formula (ܴܦܣܥ௧ସ) that performs
well under the restrictive assumption that changes to the structural stance are not too large. In
this case, the path of the CADR is mostly driven by long-term growth and the initial
structural fiscal position. This simple indicator should not be used during large fiscal
consolidation episodes (in which case, the hypothesis of stable structural balance is not
valid).

38 Under simple assumptions, the cyclically-adjusted balance ratio is approximately equal to the difference
between the nominal balance ratio and the product of the expenditure ratio and the output gap.

31

Box 2: Alternative Measures of the CADR

We simulate the behavior of four CADRs over the economic cycle. Appendix Figure 2 illustrates the evolution
of these indicators over the cycle.

x The first indicator is based on our initial definition, and is used as a benchmark. It applies the cyclical
correction to both the numerator and the denominator:
௧ଵܴܦܣܥ ൌ ൫������������୲ି୒ ൅ σ �����ϐ����୧୧ୀ୲୧ୀ୲ି୒ାଵ ൯ ����������
��୲Τ

x For the second indicator, the numerator of the debt ratio is corrected for the cycle (but the denominator is
not): ܴܦܣܥ௧ଶ ൌ ൫������������୲ି୒ ൅ σ �����ϐ����୧୧ୀ୲୧ୀ୲ି୒ାଵ ൯ ��������
��୲Τ

x The third one does the opposite, and uses as a target the nominal debt stock divided by potential GDP:
௧ଷܴܦܣܥ ൌ ሺ������������୲ሻ ����������
��୲Τ

x Finally the fourth indicator is a simple “growth-based” debt indicator (the formula is calculated in
Appendix 3):

௧ସܴܦܣܥ� ൌ ൫��ϐ����������୲ି୒ାଵେ୅ ൯ כ
ͳ െ ሺͳ െ
ሻ୒

൅ ������������������୲ି୒ כ ሺͳ െ
ሻ
୒

where G is potential growth (in nominal terms). ܴܦܣܥ௧ସ is close to ܴܦܣܥ௧ଷ in the absence of major structural
changes in fiscal policy. For G, long-term growth could be used, while the average deficit ratio over a full cycle
could proxy ��ϐ����������୲ାଵି୒େ୅ .

Appendix Figure 2. Alternative CADR Indicators Over The Cycle

Source: IMF Staff Estimates.
Note: CADR1 is the benchmark.
Hypotheses: (i) Potential growth is 2 percent per year (in nominal terms); (ii) the initial deficit is 2
percent of GDP in period 0; (iii) the output gap oscillates in the [-5;+5] range; (iv) the revenue-to-
GDP ratio is constant; (v) there is a no cyclical component in spending; (v) spending is affected
by ad hoc variation, so that the structural balance oscillates around 2 percent within the -3.5/ –
0.5 range, and (vii) the initial debt ratio is 60 percent of GDP.

50

52

54

56

58

60

62

64

66

68

70
0 1 2 3 4 5 6 7 8 9 10

Nominal debt in percent of nominal GDP

CADR1: CA Debt in percent of potential GDP

CADR2: CA Debt in percent of nominal GDP

CADR3: Nominal debt in percent of potential GDP

CADR4: Growth-based debt ratio

32

APPENDIX 3: A RULE OF THUMB FOR THE CYCLICALLY-ADJUSTED DEBT RATIO

We propose a simple formula to compute the CADR under the assumption that there is no

large change in the structural fiscal position over the period considered: οቀୈୣϐ୧ୡ୧୲౪ి
ఽ

ଢ଼౪ౌ ో౐
ቁ ൌ Ͳ

With ��ϐ����������୲େ୅ ൌ
ୈୣϐ୧ୡ୧୲౪ి ఽ
ଢ଼౪ౌ ో౐

, and
୲ ൌ
ଢ଼౪ౌ ో౐
ଢ଼౪షభౌో౐

െ ͳ ؠ
, then:

௧ସܴܦܣܥ ൌ ��ϐ����������୲େ୅ ൅ ሺ
ͳ

ͳ ൅
ሻ כ ௧ିଵܴܦܣܥ
ସ �

֜ ௧ସܴܦܣܥ ൎ ��ϐ����������୲େ୅ ൅ ሺͳ െ
ሻ כ ௧ିଵସܴܦܣܥ �

֜ ௧ସܴܦܣܥ ൎ ሺ��ϐ����������୲ି୒ାଵେ୅ ሻ כ ෍ሺͳ െ
ሻ୧
୒ିଵ

୧ୀ଴
൅ ௧ିேସܴܦܣܥ כ ሺͳ െ
ሻ୒�

௧ସܴܦܣܥ֜ ൌ ሺ��ϐ����������୲ି୒ାଵେ୅ ሻ כ
ͳ െ ሺͳ െ
ሻ୒

൅ ௧ିேܴܦܣܥ
ସ כ ሺͳ െ
ሻ୒

௧ସܴܦܣܥ֜ ൌ ൫��ϐ����������୲ି୒ାଵେ୅ ൯ כ
ͳ െ ሺͳ െ
ሻ୒

൅ ������������������୲ି୒ כ ሺͳ െ
ሻ
୒

as the CADR and nominal debt ratios are equal at the start date.

For G, long-term growth could be used, while the average deficit ratio over a full cycle (for
instance, over the last 5 years) could proxy the variable: ��ϐ����������୲ି୒ାଵେ୅ .

33

References

André, C., and N. Girouard, 2005, “Measuring Cyclically Adjusted Budget Balances for
OECD Countries,” OECD Economics Department Working Papers, No. 434
(Paris: Organization for Economic Cooperation and Development).

Auerbach, A.J., and Y. Gorodnichenko, 2012b, “Fiscal Multipliers in Recession and
Expansion,” in “Fiscal Policy after the Financial Crisis,” A. Alesina and F. Giavazzi,
(eds.), University of Chicago Press.

Bagaria, N., D. Holland and J. van Reenen, 2012, “Fiscal Consolidation during a
Depression,” Center for Economic Performance, Special Paper No. 27 (London:
London School of Economics and Political Science)

Batini, N., G. Callegari, and G. Melina, 2012, “Successful Austerity in the United States,
Europe and Japan,” IMF Working Paper 12/190 (Washington: International Monetary
Fund).

Baum, A., M. Poplawski-Ribeiro, and A. Weber, 2012, “Fiscal Multipliers and the State of

the Economy,” IMF Working Paper 12/286 (Washington: International Monetary
Fund).

Blanchard, O., and R. Perotti, 2002, “An Empirical Characterization of the Dynamic Effects
of Changes in Government Spending and Taxes on Output,” Quarterly Journal of
Economics, Vol. 117, pp. 1329–1368.

Bohn, H., 1998, “The Behaviour of U.S. Public Debt and Deficits,” Quarterly Journal of

Economics, 113, pp. 949-963.

Cherif, R., and F. Hasanov, 2010, “Public Debt Dynamics and Debt Feedback,” MPRA Paper
No. 27918.

Chung, H., and E.M. Leeper, 2007, “What has Financed Government Debt,” NBER Working
Paper 13425.

Cottarelli, C., and L. Jaramillo (2012), “Walking Hand-in-Hand: Fiscal Policy and Growth in
Advanced Economies,” IMF Working Paper 12/137 (Washington: International
Monetary Fund).

DeLong, B.J., and L.H. Summers, 2012, “Fiscal Policy in a Depressed Economy,” presented

to Brookings Papers on Economic Activity, March 2012.

Escolano, J., 2010, “A Practical Guide to Public Debt Dynamics, Fiscal Sustainability, and
Cyclical Adjustment of Budgetary Aggregates,” IMF Technical Notes and Manuals
10/02 (Washington: International Monetary Fund).

European Commission, 2011, “Public Finances in EMU-2011,” European Economy. 3.

34

ISBN: 978-92-79-19297-5 (Brussels: European Commission).

European Commission, 2012, “Report on Public Finances in EMU-2012,” European
Economy. 4. ISBN: 978-92-79-22852-0 (Brussels: European Commission).

Eyraud, L., and A. Weber, 2012, “Debt Reduction during Fiscal Consolidations: The role of
Fiscal Multipliers,” unpublished paper presented at the IMF surveillance meeting
seminar of April 10, 2012.

Favero, C., and F. Giavazzi, 2007, “Debt and the Effect of Fiscal Policy,” NBER Working
Paper 12822.

Favero, C., and F. Giavazzi, 2009, “How Large Are The Effects of Tax Changes?” NBER
Working Paper 15303.

Forni, L., L. Monteforte and L. Sessa, 2009, “The General Equilibrium Effects of Fiscal
Policy: Estimates for the Euro Area,” Journal of Public Economics, vol. 93(3-4), pp.
559-585.

Gros, D., 2011, “Can Austerity Be Self-Defeating?”, Economic Policy, CEPS Commentaries.

Haugh, D., P. Ollivaud, and D. Turner, 2009, “What Drives Sovereign Risk Premiums? An
Analysis of Recent Evidence from the Euro Area,” Economics Department Working
Papers N. 718, (Paris: Organization for Economic Cooperation and Development).

International Monetary Fund, 2009, “Fiscal Rules—Anchoring Expectations for Sustainable
Public Finances” (Washington: International Monetary Fund).

———, 2010, World Economic Outlook, October (Washington: International Monetary
Fund).

———, 2011, World Economic Outlook, October (Washington: International Monetary
Fund).

———, 2012, Fiscal Monitor, April (Washington: International Monetary Fund).

Ilzetzki, E., 2011, “Fiscal Policy and Debt Dynamics in Developing Countries,” World Bank
Policy Research Working Paper 5666.

Leeper, E.M., T. B. Walker and S.S. Yang, 2012. “Fiscal Foresight and Information Flows,”
IMF Working Paper 12/153 (Washington: International Monetary Fund).

Marin, J., 1998, “Monitoring Budgetary Discipline: Some Simple Indicators“ Bank of Italy –

Indicators of Structural Budget Balances Conference, p. 285. Available at SSRN:
http://ssrn.com/abstract=2109658

Mineshima, A., M. Poplawski-Ribeiro, and A. Weber, 2013, “Fiscal Multipliers,” in Post-
Crisis Fiscal Policy, C. Cottarelli, P. Gerson, and A. Senhadji (eds.) (forthcoming;

35

Washington, International Monetary Fund).

Olivera, J. G.H., 1967, “Money, Prices and Fiscal Lags: A Note on the Dynamics of
Inflation”, Quarterly Review (Banca Nazionale del Lavoro) 20, pp.258–267.

Organization for Economic Cooperation and Development, 2009, Economic Outlook Interim
Report.

Organization for Economic Cooperation and Development, 2012, Economic Policy Reforms
2012: Going for Growth.

Perotti, R., 2008, “In Search of the Transmission Mechanism of Fiscal Policy”, NBER
Chapters, in: NBER Macroeconomics Annual 2007, Volume 22, pages 169-226
(National Bureau of Economic Research).

Spilimbergo, A., S. Symansky, and M. Schindler, 2009, “Fiscal Multipliers,” IMF Staff
Position Note, SPN/09/11, May (Washington: International Monetary Fund).

Using a data set on government debt that was previously unavailable, the article analyzes
who bears the burden of government debt. The database includes 12 countries with both
debt and GDP data on the countries stretching back over a century. The paper shows
that in addition to the level of the Debt/GDP ratio, anticipated future changes in this
ratio, as well as the interest cost of covering the debt are important variables affecting
the economy. Most nations have seen their government debt/GDP ratio exceed 100%
in the past, but not all have sparked a financial crisis. The impact of the government
debt/GDP ratio also depends upon the causes, whether the increase is short-term due
to war or economy fluctuations, or secular due to unfunded increases in government
spending. Reducing the Debt/GDP ratio is a political decision. The government must
decide to reduce it by reducing compensation to government employees, recipients of
government funding, through higher taxes, or an outright or inflationary default.

PAYING OFF
GOVERNMENT DEBT

Two Centuries of Global Experience

Dr. Bryan Taylor, Chief Economist,
Global Financial Data

G L O B A L F I N A N C I A L D A T A

2 | Paying Off Government Debt GLOBAL FINANCIAL DATA

The current economic recession has led to
unprecedented peacetime deficits and increases in
government debt in developed countries. For only
the second time in the history of the United States,
government debt will soon exceed GDP. The long-run
costs and the impact of this growing debt remains
highly uncertain and is the chief topic of political
debate in the current US elections. While the White
House says these deficits are necessary despite the
costs, others say the debts impose costs on future
generations. Tea Party supporters say government
expenditure must be cut.

Unfortunately, very little is known about the historical
levels of government debt for different countries of the
world outside of the United States and how different
countries have dealt with large levels of government
debt in the past. Global Financial Data has collected
historical government debt and GDP data for the
major world economies going back to the 1800s. This
paper is based upon the findings of this research.

The Origin of Government Deficits
The government runs a deficit because it is unable or
unwilling to collect a sufficient amount of taxes within
any given year to cover its expenditures. For most
of its history, the United States balanced its budget
except in war time. After the war, the government
ran surpluses to pay down the debt accumulated
during the war or ran deficits less than the growth in
nominal GDP. A long-term graph of US debt on page

23 shows rises in the Debt/GDP ratio during the War
of 1812, Civil War, World War I, and World War II.

The true cost to the economy of government is
the expenditures it makes, not the taxes it collects.
Government can either collect taxes today or issue
promises (currency or bonds) to pay for its purchases
in the future. When the government increases the
money supply, it can cause inflation, and if it issues
bonds, it can “crowd out” the private sector.

Running deficits implies less spending in the future
since the government must pay interest or retire
bonds. In some cases, short-run deficits can be
justified. Just as consumers or businesses may wish
to smooth out the cost of consumption over time,
so can government. If the government is building
infrastructure which has long-term benefits, it may
borrow money today to be paid off in the future.
Similarly, the government may run a cyclical deficit
during a recession which it can pay off when the
economy recovers.

Structural deficits are another matter. A structural
deficit that is used to pay for services or transfer
income, unlike capital investment, does not add to
the net wealth of society, and implies higher taxes
or lower government services in the future to offset
the accumulated structural deficits. As Robert Barro
has shown, these types of structural deficits can have
multipliers less than one because of their impact on
incentives and the economic misallocations they
create. Unfortunately, a portion of the current deficit
the US is running is structural in nature.

A structural deficit implies structural adjustments
in the future; however, it may be difficult to generate
the future surpluses needed to pay off this debt for
demographic reasons. An aging population implies
both a higher recipient to taxpayer ratio, and higher
health care costs for the elderly. Calculations of
the implied cost of the entitlement programs the

3 | Paying Off Government Debt GLOBAL FINANCIAL DATA

government has promised in the future, such as Social
Security, Medicare, Medicaid and other programs,
predict large increases in these costs in the future
without large reductions in the promised benefits.
Any attempt to run surpluses to pay back the debt
will require large increases in taxes.

Paying off the government debt
Paying off the debt is largely a political choice. Who
bears the cost of paying off the debt? Is it government
workers through lower pay and lower benefits? Is
it individuals who see a reduction in government
services or entitlements, either directly through cuts
or indirectly through slower growth in benefits?
Is it taxpayers who pay higher taxes and fees? Are
the additional taxes born by the rich or the poor or
both? Is it bondholders who get paid back in inflated
currency or don’t get paid back at all?

Government debt as a share of GDP can be reduced
or eliminated in a number of ways.

1. Run surpluses and pay off the debt as happened
in the US in the 1830s, or reduce the debt/
GDP ratio by running surpluses as occurred
around 2000 under Bill Clinton. Here the cost
of the debt is imposed directly on taxpayers
with no loss to fixed income investors.

2. Run a deficit that is less than the growth in nominal
GDP so even if you continue to run a deficit, the
debt/GDP ratio shrinks. The government may
have to run a surplus before interest payments in
order to achieve this. The lower the interest rate, the
easier this is to do. This is largely what happened
in the United States between 1945 and 1973. This
imposes a lower cost on taxpayers in the short-
run, but raises the total cost of debt over time.

3. Inflate your way out of the debt. If the inflation
rate is high enough, nominal GDP can grow faster
than the deficit reducing the debt/GDP ratio. This

imposes high costs on bondholders who get paid
back in inflated currency, but relieves taxpayers of
the burden. A hyperinflation as in Germany can
wipe out fixed income investors. This relieves
taxpayers of the interest and principal burden of the
debt, but at a high cost to fixed income investors.
This solution works well with non-recurring
debt (wars), but not with secular social debts.

4. Outright Default. This can be done through
a currency reform if most government
debt is held domestically (Germany, 1948),
devaluation if the debt is held by foreigners but
in the local currency, or a default on foreign
currency bonds. Here the entire cost is born
by bondholders to the benefit of taxpayers,
but it becomes difficult to issue new bonds.

Just as the purpose of running a deficit is to hide the
cost of government services and expenditures through
indirect taxation (inflation tax) or delaying the costs
(issuing bonds), so the goal of the government in
paying down the debt will be to make the cost as
indirect as possible, or to impose the costs on those
without political power.

The rest of this paper will look at the experience of
twelve major economies to see how they have created
and paid off deficits in the past. Each country’s
experience could be the subject of a book, so only
the barest of outlines is possible. Nevertheless, these
brief histories and their subsequent graphs will give
an idea of the choices the major developed countries
now face. We will look at both the debt/GDP ratio and
Interest Share of GDP which equals the benchmark
bond interest rate times the debt/GDP ratio.

Historically, there have been several factors which
have caused increases in the debt/GDP ratio. One is
war. Globally, the primary examples are World War
I and World War II. The two wars were “paid for”
differently.

4 | Paying Off Government Debt GLOBAL FINANCIAL DATA

Most governments used inflation to reduce the cost
of debt accumulated during World War I and in
Germany even used “inflationary default” as a way
of eliminating the debt, wiping out bondholders
completely. In the United States during World War
II, government controlled prices and interest rates
which produced a higher return of principal in real
terms, but lower interest rates to investors. The debt
was paid off by allowing economic growth to shrink
the deficits. On the other hand, inflation in Italy and
France wiped out their debt after World War II while
Germany used a Currency Reform to eliminate its
obligations.

Major recessions and depressions also increased
government debt, the Great Depression of the
1930s and the current Great Recession being prime
examples. Governments tend to grow their way out
of the deficits generated by recessions. Wars and
Recessions provide quick increases in government
debt which can be reversed over time.

The final source of government deficits is the attempt
to increase government benefits and entitlements
faster than people are willing to pay for them. These
deficits are secular in nature and generally require
a restructuring of government expenditures and
obligations to stop the accumulation of debt. Because
of their structural nature, an outright or inflationary
default is unlikely to work. Those who fear the current
deficits will lead to inflation miss this point. Because
of the politics involved in making these structural
adjustments, reversing structural deficits is the most
difficult of all.

Lenders are willing to tolerate deficits due to War and
Recessions because they are temporary events that can
be followed by surpluses after the cessation of war or
a return to growth. Deficits that occur due to secular
increases in government services that cannot be
immediately reversed will inevitably lead to a financial
crisis that ends in the repeal of these services.

The current growth in government debt is both
structural and cyclical. Many people feel that once
the government debt/GDP ratio exceeds 100%, a
financial vurred. Unfortunately, history shows that
governments have to be forced into a crisis to solve
these problems, rather than addressing them before
the crisis occurs.

Australia
Australia saw its Debt/GDP ratio rise steadily from 1850
to 1900 when it established itself as a Commonwealth.
By 1900, the Debt/GDP ratio was over 100%. It rose
above 100% again during World War I and peaked at
almost 200% of GDP during the Great Depression.
Significantly, the World War II Debt/GDP peak was
below the Great Depression peak, though still over

180% of GDP. Since World War II, Australia has
grown out of its Debt, which now represents less than
10% of GDP placing it in one of the most fiscally sound
positions of any developed country.

Australia was able to increase its Debt/GDP ratio
through 1900 because it was a growing colony. It
could borrow money in London with little problem.
As it grew, Australia moved its debt burden from
international to domestic borrowers and has now
almost completely eliminated the foreign debt
portion of its government debt. Australia borrowed
as it developed its economy, and has grown out of its
debt successfully with few periods of high inflation.
Returns to fixed income investors in Australia have
been high as a result.

5 | Paying Off Government Debt GLOBAL FINANCIAL DATA

Figure 1
Australia government debt/
GDP ratio 1850-2010

Figure 2
Australia government debt
Interest/GDP ratio 1850-2010

6 | Paying Off Government Debt GLOBAL FINANCIAL DATA

Canada
Unlike Australia, Canada kept its government debt
relatively low until World War I, at which point
it rose to 70% of GDP. The debt rose to over 80%
of GDP during the 1930s and peaked at over 150%
during World War II. The debt declined steadily
until the 1970s. Canada reached a debt crisis in the
1990s when secular increases in government services
and entitlements pushed debt to over 70% of GDP
and the interest cost to over 6% of GDP. Even during

World War II when debt exceeded 150% of GDP, the
“interest cost” of the debt was only 4% of GDP.

The increase in government debt was clearly unsus-
tainable. The Canadian government was forced to cut
back on its spending to eliminate its deficits. Conse-
quently, because Canada put its government finances
in order in the 1990s, it has suffered less during the
current Recession than other developed countries.

Figure 3
Canada government debt/
GDP ratio 1870-2010

Figure 4
Canada government debt
interest/GDP ratio 1870-2010

7 | Paying Off Government Debt GLOBAL FINANCIAL DATA

France
France saw rising deficits during the 19th century
until it reached 100% of GDP by 1900. Most of
this increase occurred after 1870 when Germany
imposed a costly indemnity on France as a result of
the Franco-Prussian War. Consequently, when World
War I began, France’s Debt/GDP ratio exceeded 80%
(vs. 3% in the US). Despite inflation during World
War I, France’s Debt/GDP ratio rose to over 200% by
the early 1920s. Because of this debt, it is no wonder
France wanted to impose a large indemnity on
Germany when Germany lost World War I.

By the beginning of World War II, France’s Debt/
GDP ratio was down to 100% but shot over 200%
during World War II. With no prospect of an
indemnity from Germany after World War II, France
inflated its way out of its debt imposing heavy losses
on bondholders, but to the benefit of taxpayers. This
laid the foundations for France’s rapid growth after
World War II. Despite the fact that France’s Debt/
GDP ratio has grown since the 1970s, it has not
reached crisis levels due to high tax rates.

What is important to see about France is that after
almost three decades (1915-1945) of having taxpayers
bear a high debt cost, the government finally punished
bondholders through an “inflationary default”.

Today, this would be more difficult to do because
debt is issued in Euros rather than Francs or another
local currency, and, as we have seen with Greece, the
Euro countries will help countries that could default
on their debt because of the costs of the default
contagion effect. Nevertheless, there is no guarantee
that a country such as Greece couldn’t remove the
Euro strait jacket, abandon the Euro, convert its debt
into Drachmas and inflate its way out. This possibility
is what keeps Greek debt at its current high yields.

Territory of the French Republic

8 | Paying Off Government Debt GLOBAL FINANCIAL DATA

Figure 5
France government debt/
GDP ratio 1850-2010

Figure 6
France government debt
Interest/GDP ratio 1850-2010

9 | Paying Off Government Debt GLOBAL FINANCIAL DATA

Germany
The graphs for Germany are deceiving because there
are key periods, during World War I and World War
II, when data on Germany’s debts are unavailable.
Germany inflated its way out of its debts from
World War I through hyperinflation, wiping out
bondholders. In 1948, Germany used a currency
conversion from Military Marks to Deutschemarks
to effectively reduce its debt obligations by 90%.

In part, because Germany twice destroyed the assets
of bondholders, it has been more fiscally responsible
than other countries since World War II. Although
its Debt/GDP ratio has been rising since the 1970s,
it remains lower than most other countries. Holders
of Confederate and German bonds know that if you
lend to the losing side of a war, bondholders can be
wiped out.

Figure 7
Germany government debt/
GDP ratio 1870-2010

Figure 8
Germany government debt
interest/GDP ratio 1870-2010

10 | Paying Off Government Debt GLOBAL FINANCIAL DATA

Italy
Although Italy has had a long history of running
deficits, its Debt/GDP ratio has rarely exceeded
100% of GDP by a large margin. Like France, Italy
inflated its way out of its debts after World War II,
imposing large losses on bondholders. Unlike France,
it consistently ran budget deficits after World War II
and used inflation as a way of minimizing the true
cost. In the 1990s it reformed its finances to stop the
Debt/GDP ratio from growing more and upon joining
the Euro benefitted from lower interest rates cutting
the Interest Coverage Cost of its debt to more realistic
levels. Significantly, despite its high Debt/GDP ratio,

the yield on its government bonds has not risen as
steeply as those of Ireland, Portugal, Greece and
Spain.

The lesson for Italy is that persistently high deficits
impose persistently high costs on investors even if
the debt never hits ruinous levels. Because Italy has
persistently refused to balance its budgets, it has
provided the worst returns of any G-7 country to both
equity and fixed income investors. This is because Italy
has had consistently high inflation, yielding low or
negative interest rates at the expense of bondholders.

Milan Stock Exchange

11 | Paying Off Government Debt GLOBAL FINANCIAL DATA

Figure 9
Italy government debt/GDP
ratio 1860-2010

Figure 10
Italy government debt interest/
GDP ratio 1860-2010

12 | Paying Off Government Debt GLOBAL FINANCIAL DATA

Japan
Japan’s Debt/GDP history differs from other
developed countries in several ways. The cost of
the Russo-Japanese War in 1905 shows up, and as
a result of the cost of World War II, the Debt/GDP
ratio peaks at 200%. Since the borrowing for the war
was almost exclusively domestic, Japan was able to
inflate its way out of its debt in the late 1940s, laying
the foundations for its economic growth after World
War II. Since the 1970s, Japan’s debt has increased
steadily, reaching almost 200% of GDP.

Although Japan’s Debt/GDP ratio is approaching
200%, if you look at the interest cost of the
Government debt, it was actually higher in the 1980s
than it is today. Japan has seen no increase in nominal
GDP for almost 20 years, so the Debt/GDP ratio has
steadily risen. Japan has been unable to inflate its way
out of its debt, but has paid miniscule interest rates
to bondholders and absorbed much of the savings
within the country.

The government’s debts have crowded out the private
sector and virtually eliminated nominal returns to
investors. It is no surprise then that Japan has suffered
two lost decades of no growth. Japan has boxed itself
into a corner in which bondholders either get low
returns due to disinflation. If inflation were to return,
bond prices would decline and inflation would reduce
real returns. With an aging population, no population
growth, and low interest rates with savings absorbed
by the government, it is difficult to see how Japan can
ever return to any level of economic growth. Every
country caught in the current financial crisis would
be wise not to follow in Japan’s footsteps.

Tokyo Stock Exchange

13 | Paying Off Government Debt GLOBAL FINANCIAL DATA

Figure 11
Japan government debt/
GDP ratio 1885-2010

Figure 12
Japan government debt Interest/
GDP ratio 1885-2010

14 | Paying Off Government Debt GLOBAL FINANCIAL DATA

Netherlands
The Netherlands had high debt in the mid-1800s
due to the cost of the war with Belgium (1830-1831).
The Debt/GDP ratio declined consistently until the
1930s, in part because the Netherlands was neutral in
World War I. By the end of World War II, its debt had
increased to over 100% of GDP, but through growth
and tight fiscal policy, the Netherlands was able

to get its debt down to almost 20% of GDP by the
1970s. The Netherlands has had no period in which
fixed income investors were significantly punished
by the government. Probably, this is because of the
Netherlands history as a financial center, and it status
as a small, open economy.

Figure 13
Netherlands government debt/
GDP ratio 1850-2010

Figure 14
Netherlands government debt
Interest/GDP ratio 1850-2010

15 | Paying Off Government Debt GLOBAL FINANCIAL DATA

Spain
Relative to other countries, Spain had persistently
high deficits during the 1800s and defaulted or
rescheduled its debt in 1809, 1820, 1831, 1834, 1851,
1867, 1872, 1882 and during 1936-1939. Spain’s debt
rose from 60% of GDP in 1860 to over 160% of GDP
in 1875 leading to a default in 1882. By staying out
of both World War I and World War II, Spain was
able to avoid the large debts created by these wars for
other European countries. By the 1970s, its Debt/
GDP ratio had fallen to 10% of GDP, but has risen
since then as its social policies have changed.

Spain’s debt history is different from other European
countries. It had persistent deficits in the 1800s when
many European countries were reducing their debts
or keeping them small, then avoided the costs of both
World Wars. Despite reducing its debt after World
War II, adjusted for inflation, bondholders lost
money between 1942 and 1985 due to rising inflation
and rising interest rates. Inflation is fixed income
investors’ worst enemy.

Figure 15
Spain government debt/GDP
ratio 1850-2010

Figure 16
Spain government debt Interest/
GDP ratio 1850-2010

16 | Paying Off Government Debt GLOBAL FINANCIAL DATA

Sweden
Sweden, along with Switzerland, is one of the few
countries that has never seen its Debt/GDP ratio rise
above 100%. Sweden was neutral in World War I and
saw its Debt/GDP ratio rise to only 50% during World
War II. The most significant increase in debt came in
the 1980s and 1990s when large increases in secular
social spending increased the Debt/GDP ratio to 75%
and the Interest Coverage to 8% of GDP. This sparked
a financial crisis in Sweden in the 1990s which led to
a reform of its fiscal finances similar to the Canadian
reforms. These financial reforms included school

vouchers, privatization of pensions and other reforms
that the market-loving United States has refused
to make. As a result of this, its interest coverage has
fallen back to 2% of GDP and the Debt/GDP ratio
back to the 40% level.

Sweden and Canada both show that entitlements can
be reformed when a financial crisis forces a country to
do so. However, making these reforms in a relatively
small, homogeneous country like Sweden is easier than
in a diverse, large country such as the United States.

Figure 17
Sweden government debt/GDP
ratio 1880-2010

Figure 18
Sweden government debt
Interest/GDP ratio 1880-2010

17 | Paying Off Government Debt GLOBAL FINANCIAL DATA

Switzerland
Switzerland also has not seen its Debt/GDP ratio exceed 100%. The only time the ratio exceeded 50%
was during World War II, but this ratio has consistently remained below 30% since the 1950s. Switzerland
provides a model for the rest of the world for fiscal control and the benefits of a decentralized governmental
structure.

Figure 19
Switzerland government debt/
GDP ratio 1925-2010

Figure 20
Switzerland government debt
Interest/GDP ratio 1925-2010

18 | Paying Off Government Debt GLOBAL FINANCIAL DATA

United Kingdom
The United Kingdom has the longest history of
government data available for any country. GFD’s
data on government revenues for the UK goes back
to 1168 and government debt back to 1689. Both
the United States and France inflated their way out
of their debts of the 1700s, but the United Kingdom
did not. Consequently, its debt exceeded 200% of
GDP when the Napoleonic wars ended in 1815 and
its interest coverage exceeded 10% of GDP. Under
normal circumstances, this would have sparked a
financial crisis, but in the midst of the Napoleonic
wars, there were few investment alternatives and
the Pound Sterling was the reserve currency of the
19th Century. After the wars were over with, the UK
signaled that it would meet its debt obligations, which
it did, and the Debt/GDP ratio steadily declined until
1915.

The UK could not have carried such a heavy debt load
had it not been the world’s reserve currency and the
financial center of the world. That the UK was awash
with capital is witnessed by the various bubbles that
occurred in the 1810s with canals, 1820s with mining
and Latin American stocks, the 1840s with railroads,
and for the rest of the 1800s with the steady rise in
domestic and foreign securities listed on the London
Stock Exchange. The United States reaped similar
benefits in the 20th Century and today.

Despite declining to almost 20% by 1914, the UK’s
Debt/GDP ratio rose to over 150% after World War
I and over 200% after World War II. Unlike France,
Germany, Italy and other continental countries, the
UK refused to pursue an inflationary default after
either World War I or World War II. London remained
the financial center of Europe after World War I, and
recognized the cost of an inflationary default.

No doubt, the debt burden this imposed, equal to
around 5% of GDP from the 1920s to 1970s when
the UK finally went through double-digit inflation
which lowered its Debt/GDP ratio. Someone has to
pay the cost of the debt and politics determines this.
Continental countries punished bondholders after
World War II, but benefited in the long run. The
UK respected its debt obligations, but suffered lower
growth as a result.

The UK now faces a large deficit which Prime Minister
Cameron is addressing with large cuts in spending.
It appears to be following the path of Canada and
Sweden in the 1990s, though how successful it will
be remains to be seen.

The UK shows both the benefits and the costs of being
a financial center and having a reserve currency. On
the one hand, this enables the country to borrow
more as a share of GDP without sparking a financial
crisis than would otherwise be possible. On the other
hand, it makes it more difficult to default on debt,
either through inflation or an outright default. This
constrains growth in the long-run since the country
is unable to reduce its debt load through inflationary
or outright default. These are important lessons for
the United States.

The Bank of England; the central
bank of the United Kingdom

19 | Paying Off Government Debt GLOBAL FINANCIAL DATA

Figure 21
United Kingdom government debt/
GDP ratio 1800-2010

Figure 22
United Kingdom government debt
Interest/GDP ratio 1800-2010

20 | Paying Off Government Debt GLOBAL FINANCIAL DATA

United States
The debt history of the US is probably better known
than any other country. The US largely paid for
the Revolutionary war through inflation with 1000
Continental Dollars converted into 1 Silver Dollar by
the end of the war. The Federal Government paid off
its debt in the 1830s and despite borrowing 40% of
GDP to pay for the Civil War, by 1915, the US had
virtually paid off its debt once again. The Debt/GDP
ratio rose back to 40% after World War I, back to
40% during the 1930s and to over 120% of GDP
during World War II, fell to 35% of GDP by the mid-
1970s and is now reapproaching 100% of GDP. Most
European countries saw their Debt/GDP ratios fall
between 1995 and 2008, but the US ratio rose due
to tax cuts, wars and expansions in government
spending leaving the country with a higher Debt/
GDP ratio than most OECD countries when the
Great Recession began.

The interest cost of covering the government’s debt
has consistently risen. Despite the fact that the Debt/
GDP ratio exceeded 120% after World War II, low
interest rates meant that the interest coverage cost
was around 2% of GDP. The cost remained around
2% until the 1970s as declining Debt/GDP was offset
by rising interest rates. By the early 1980s, this rose
to the 5% level and stayed above 4% until the late
1990s. It was during this period of time the bond
market began to impose fiscal stringency on the
Federal Government.

The US now faces a situation similar to Japan in
which the Debt/GDP ratio is rising, but the Interest
Coverage cost is declining due to falling interest
rates. If interest rates were to rise, the high Debt/
GDP ratio would put the US in the same situation
as the UK after World War II in which high interest
cost coverage constrained growth. Unless there is a
fundamental change to the fiscal policy of the United
States, its government debt load could place it in a
position of facing Japanese slow growth for a decade
or two.

This problem is compounded by the fact that the debt
increase is caused by secular growth in government
services, not temporary military expenditures or
cyclical economic fluctuations. Even if the Bush tax
cuts were allowed to completely expire, as neither
party wants, the US would remain in a precarious
fiscal position. Being a reserve currency with a
financial center, as the UK was until World War II,
the US cannot easily default by inflation. As has been
seen with Japan, if interest rates were to remain in the
2% range, the government’s Debt/GDP ratio could
rise to 200% without sparking a financial crisis.
However, as the Debt/GDP ratio rises, any sharp rise
in interest rates could push the US into a financial
crisis which would force the US to make choices on
limiting entitlements and social spending similar to
what Canada and Sweden did in the 1990s.

Who bears the burden of government expenditures
is a political choice. Politicians can decide to impose
costs on bondholders directly through an inflationary
default, currency reform, or direct default. With US
government bonds held
so widely throughout
the world, this would be
difficult. However, until
a crisis hits Washington,
there is no incentive for
fiscal reform. As long
as interest rates remain
low, the US can pile up
debt for another decade,
as Japan has done.
Whether bondholders
want to watch a slow-
motion train wreck or
will jump the tracks
remains to be seen.

The Stock Exchange at 10-12 Broad
street, in 1882

21 | Paying Off Government Debt GLOBAL FINANCIAL DATA

Figure 23
United States government debt/
GDP ratio 1790-2010

Figure 24
United States government debt
Interest/GDP ratio 1792-2010

22 | Paying Off Government Debt GLOBAL FINANCIAL DATA

Conclusions
Our review of debt histories from twelve large
countries has been revealing and can lead us to a
number of conclusions.

1. There is no ceiling on Debt/GDP ratios. Most of
the countries we looked at had Debt/GDP ratios
that exceeded 100% without sparking a financial
crisis and in some cases, the ratio hit 200%.
Whether the increase in Debt is perceived as
temporary and will decline when the cause (war,
recession) is over, or whether the cause is secular
growth in unpaid for government services is more
important than the level of debt.

2. The Interest Coverage Cost is more important
in sparking a financial crisis than the Debt/GDP
ratio. Japan’s low interest rates have been able to
sustain rising debt. If the Interest Coverage rises
above 5%, this can spark a financial crisis if it
appears the level will remain above that level and
will continue to rise.

3. Even if high debt does not spark a financial crisis,
high Debt/GDP ratios can constrain economic
growth making it more difficult to break the
burden of the debt. The UK and Japan offer good
examples of this.

4. Countries with reserve currencies or countries
that are financial centers can find it easier to raise
money to cover government debts, but will also
find it more difficult to default.

5. Countries that default through inflation, a
currency reform or an outright default benefit by
relieving themselves of the burden of the debt. The
cost is that the country must reestablish itself as
fiscally conservative before it can borrow again.

6. Debt/GDP and returns to fixed income investors
are inversely related with a lag. Quickly or slowly,
government tries to inflate their way out of debt
reducing returns to fixed income investors. Fixed
income investors got very poor returns globally
between 1945 and 1980 after the sharp rise in debt
during World War II, but superior returns between
1980 and 2010 as inflation subsided. Rising Debt/
GDP ratios imply low returns to investors in the
future either through low (Japanese) interest rates
or rising inflation, interest rates and falling bond
prices.

7. Higher interest rates can spark a financial crisis
that forces the government to reform, especially
if the debts were created because of rising secular
social costs. War debts are more likely to face
inflationary default than rising social costs because
the war debts are a one-time non-recurring cost.
Since social expenditures largely redistribute
income, governments cannot inflate their way out
of these costs, but must eventually reform. For this
reason, eventually the entitlement problem, which
is the basis of the current rising deficits, must be
fixed. The only question is whether this problem is
taken care of now, or the government waits until a
financial crisis forces the government to reform is
fiscal ineptitude.

8. The burden of government debt is born by
government employees, taxpayers and bondholders.
Above all, politics determines who bears the costs.
With low interest rates or economic growth,
politicians can continue to run deficits. Above
all, politics determines who bears the costs of the
government’s debt. The interest groups with the
least political influence are the ones who will pay
the price.

The Canadian Experience in Reducing
Budget Deficits and Debt

Paul Martin

My objective in these remarks is to describe what we in Canada
are doing to improve the fiscal health of the federal government. I
will focus particularly on the political dimensions of the process,
since obviously there is little I can teach this audience about the
economic dimensions. That said, we should begin with some back-
ground as to the origins and magnitude of Canada’s fiscal problem.

Our general pattern of deficits and debt has much in common with
that of many other Organization for Economic Cooperation and
Development (OECD) countries, including the United States—that
is, a very high ratio of debt to GDP coming out of the Second World
War, followed by a rapid decline of the debt ratio as the military was
demobilized, as economic growth took off, and as the effective
interest rate on government debt remained low.

In Canada’s case, the ratio of total government debt—federal and
provincial combined—fell from slightly more than 100 percent of
GDP in 1945, to bottom out at about 20 percent in 1974. Since then,
the debt ratio has risen in virtually every year and is now once again
back to approximately 100 percent of GDP.1 But unfortunately, the
contemporary circumstances do not compare with the situation fifty
years ago. Today, there is no potential for massive spending cuts
from military demobilization, and interest rates exceed the rate of
economic growth—precisely the reverse of the postwar conditions.

203

The growth of debt since the mid-1970s has coincided not only
with the maturation of the modern welfare state, but also with the
great productivity slowdown in all of the advanced economies. In
this regard, Canada’s situation has much in common with most G-

7

countries.

In what follows I will restrict attention to the situation of the
federal government.2 There, as we headed into the 1981-82 reces-
sion, debt was a manageable 30 percent of GDP. Five years later, the
ratio had climbed over 50 percent and warning lights had begun to
flash. With the last recession, it jumped again, from about 55 percent
of GDP in 1989-90 to approximately 73 percent today.

This extremely large stock of debt, when combined with interest
rates that exceed Canada’s rate of economic growth by 3 to 4 percentage
points, means that simply to stabilize the federal government’s debt

12

0

Percent of GDP

100

80

60 199

4

1

40

0
40

France Germany U.K.Japan Canada Italy

Source: Statistics Canada for Canada. OECD estimates for other countries.

U.S.

Total Government Net Fiscal Debt – G-7 Countries
(National Accounts Basis)

Chart

1

–

20

20

1974

204 P aul Martin

ratio, our revenue intake must now exceed program spending by
about 3.5 percent of GDP or close to C$30 billion.

Coming into office in the fall of 1993, we fully expected to find a
nasty fiscal situation—the combination of misplaced priorities,
which is the stuff of politics, and compound interest, the stuff of the
inexorable laws of arithmetic. Indeed, the federal deficit in 1992-9

3

reached almost 6 percent of GDP. But the much more fundamental
problem was the apparent intractability of the sheer arithmetic of
debt when the interest rate is higher than the economic growth rate.
It became clear that a very long period of restraint would have to be
endured to turn the debt momentum around. As part of that, we also
needed to fundamentally re-think the role of the national govern-
ment and the structure of its spending. But there was no ducking the
issue.

The Canadian Experience in Reducing Budget Deficits and Debt

70

Percent of GDP

60

50

40
80

10

30

1983-84

Federal Net Debt
(Public Accounts Basis)

Chart

2

0
20

1986-87 1989-90 1992-93 1995-9

6

Forecast

1980-81

205

The economic warning flags were everywhere. Despite Canada’s
having one of the world’s best inflation records since 1989, the
currency was under constant pressure and real interest rates were
increasing, putting a drag on growth and obviously exacerbating the
fiscal problem.3

Meanwhile, the country was not generating sufficient domestic
savings to finance both the investment needs of the private sector as
well as huge and chronic public sector deficits. Canada was there-
fore borrowing increasingly abroad. The result has been an accumu-
lated net foreign debt—owed by the public and private sectors
combined—that now exceeds 45 percent of GDP.

Despite these warning signs, the debt and deficit were slow to
become top-of-mind issues with the Canadian public. Then, rather
suddenly, all that changed. People came to understand that the

40
Percent of GDP
30
20
10
50

1977

Canada’s Net Foreign Debt
Chart 3

0
1980 1983 19891974 19921986 1994

206 P aul Martin

problem had reached a genuinely critical juncture; that endless
deficits really did have something to do with Canada’s high real
interest rates; and that higher government spending really did trans-
late into higher taxes, the tolerance for which had reached its limit.

All of this amounted to a crucial turning point in national psychol-
ogy. Without some such shift in the public mindset, it would seem
to me that democratic societies cannot come fully to grips with what
needs to be done to solve a debt problem.

In Canada’s case, the philosophical chasm that had always divided
the deficit hawks and doves began to close. The fact was that no one
could deny the stark arithmetic of compound interest. For example,
interest charges, which had consumed only 11 cents of every federal
revenue dollar in the mid-1970s, now consume almost 34 cents or
about 6 percent of GDP.

And while some economists may argue about the extent to which
deficits crowd out private investment, those of us who believe in a
socially progressive and pro-active role for government understand
clearly how debt charges crowd out spending on valued public
programs such as health care and old age security. Indeed, from the
perspective of the political philosophy of the Liberal Party of Can-
ada, the debt burden has become an even more serious threat to the
social conscience of government than to our bond ratings.

The governments of most of Canada’s provinces—regardless of
their political ideology—have come to essentially the same conclu-
sion. Within the last couple of years, most have moved to put their
fiscal houses in order. Undoubtedly, this helped to create a wide-
spread public expectation that the federal government would do
likewise—that it would finally stop talking about the deficit problem
and really buckle down to do something about it.

Let me turn next to what we have done.

We entered the 1993 election campaign with a commitment to
ultimately balance the budget and a very specific interim target to

The Canadian Experience in Reducing Budget Deficits and Debt 207

reduce the deficit to 3 percent of GDP by 1996-97.4 While some
have considered this to be an insufficiently ambitious target, the
reality is that Canada’s federal deficit has exceeded 4.5 percent of
GDP virtually every year since 1976 and the resulting accumulation
of debt made hitting the 3 percent target in 1996-97—down from
about 6 percent in 1993-94—a really significant challenge.

I would emphasize that 3 percent is an interim target on the way
to a balanced budget. A zero deficit is not only of fiscal signifi-
cance—it is of great symbolic significance, a benchmark of fiscal
responsibility that has been adopted as well by provincial govern-
ments and embraced by the Canadian public. Deficit elimination is
thus a goal that will continue to discipline our budget choices. But
it can be argued that an even more fundamental objective in strictly
economic terms is to put the debt-to-GDP ratio on a steady down-
trend. This is the key to fiscal stability and assured sustainability

8

Percent of GDP
7
6
4

9

Federal Government Deficit
Chart 4

0
1977-78

3
2
1

5

1980-81 1983-84 1986-87 1992-931974-75 1995-961989-90

208 P aul Martin

and thus to a stronger economy. A much lower debt ratio must be a
legacy of this government.

We have already made a great deal of fiscal progress—more, in
fact, than is generally recognized. We began to turn the corner with
our first budget in February, 1994, which secured significant savings,
especially in the Department of Defense and in the Unemployment
Insurance program. That was followed a year later by what many regard
as the most significant federal budget of the postwar era. As a result:

Fiscal actions—that is, spending cuts and some very limited
revenue measures—will total C$29 billion over the three
years through fiscal 1997-98. To put that figure in a U.S.
context, it would be equivalent to roughly US$210 billion
of action over the same period.

The fiscal savings will come overwhelmingly from spend-
ing cuts—they will outweigh tax increases by a ratio of 7
to 1. While we took some action to tighten up the corporate
tax system and generally to improve tax fairness, we held
the line—as we also had in the 1994 budget—on sales taxes
and on personal income tax rates.

Program spending by 1996-97 will be 10 percent lower
than in 1993-94. In fact, Canada is the only member of
the G-7 to budget an absolute decline in program spending.
For a comparative perspective, consider that federal pro-
gram spending in both Canada and the United States was an
identical 17.5 percent of GDP in 1992-93. Looking to 1996-
97, the U.S. budget forecasts a reduction to 16.3 percent of
GDP while Canada’s ratio will have fallen to just over 13
percent, the lowest level since the early 1950s.

By 1996-97, the 3 percent deficit target will be met and our
market borrowing requirement—equivalent to the Unified
Budget Basis deficit in the United States—will be down to
1.7 percent of GDP, projected to be the lowest among
central governments of G-7 countries.

The Canadian Experience in Reducing Budget Deficits and Debt 209

The operating surplus is forecast to be 3.6 percent of GDP
(C$30 billion) by 1996-97. Most significantly, this will be
sufficient to finally begin cutting the debt ratio. Increasing
operating surpluses in the future will ensure an accelerating
reduction of that ratio.

These facts and figures tell only part of the story—in fact, the
lesser part in our view. More fundamentally, we have sought to
change permanently the structure of federal government spending.
Since our fiscal problem is structural, our remedies must be struc-
tural as well.

1996-97

20
Percent of GDP

15

10

Federal Government Program Expenditure
Chart

5

0
1988-89

5

1990-91 1992-93 1994-95
Forecast

Notes: 1. Total expenditure minus gross debt service charges for Canada and the U.S.
2. Fiscal years ending March 31 and September 30 of the same year for Canada and the

U.S., respectively.

Sources: Canada: Department of Finance Canada: United States: Department of Commerce and
Budget of the United States Government, Fiscal Year 1996.

Canada
United States

210 P aul Martin

Our overarching objective is to promote jobs and growth, a par-
ticularly resonant theme in Canada given the nation’s unusually
weak recovery from the 1990-91 recession. For us, deficit cutting is
not an end in itself, but rather an urgent and necessary means to
achieve our fundamental jobs and growth objective.

We have developed a comprehensive game plan to address the
underlying issues, focusing primarily on promoting productivity
growth while working to correct aspects of the labor market that
have caused Canada’s core unemployment rate to roughly double
over the past 20 years.5 That game plan guides our budget choices
which, beyond their purely fiscal purpose, are designed to help
achieve the structural changes that lie at the heart of our jobs and
growth strategy.

For example, it was clear to us that many long-standing business
subsidies were, in fact, hurting our productivity and competitive-
ness. So in last February’s budget we cut total subsidies to business
by 60 percent over the next three years. This included ending a more
than one-half-billion-dollar-a-year subsidy for grain transportation
in the west that had been in place since the last century.

We realized that many features of Canada’s unemployment insur-
ance system were impeding rather than promoting the efficient
function of our labor market. So we are taking significant steps to
transform unemployment insurance—with emphasis on getting the
incentives right and on active measures to help the long-term job-
less.

We also took a hard look at federal transfer payments to the
provinces, which this year will account for almost 23 percent of
program spending. Clearly, the fiscal problem could never be tamed
without some reduction of these transfers. We were nevertheless
determined not to cut back our support to the provinces by any
greater percentage than we were hitting programs in our own back
yard. Furthermore, it would not have made sense simply to off-load
the problem onto another level of government.

The Canadian Experience in Reducing Budget Deficits and Debt 211

So to give the provinces greater flexibility, we will convert the
present cost sharing of social assistance payments into part of a
larger block grant. This will also increase the incentive to develop
more innovative and cost-efficient ways of delivering social assis-
tance. That said, the new block-funded transfer will still require
provinces to respect certain nationwide principles, particularly in
respect to health care delivery but also in the social assistance
domain.

There can be no questioning the commitment of the government
of Canada, and of Canadians themselves, to our publicly-funded
national Medicare program. We view this as a joint responsibility of
the federal and provincial governments. As such, one of the impera-
tives of getting our fiscal house in order is to be able to have
continuing, stable federal funding for health care.

The 1995 budget also announced significant reductions in spend-
ing by federal departments and agencies. As a result, departmental
outlays by 1997-98 will be close to 20 percent lower (in absolute
dollar terms) than last fiscal year; the public service will be reduced
by about 15 percent or roughly 45,000 positions; and several activi-
ties, notably in the transport sector, will be privatized or commer-
cialized.

Some departments—for example, our Departments of Industry
and of Transport—will cut their spending in half. In fact, only one
department of the f ederal gover nment—Indian and Northern
Affairs—will be spending more in three years’ time than it is today.
And that one exceptional case reflects the extraordinary circum-
stances of Canada’s native people. Every other branch of govern-
ment will be required to get by with less.

We believe, nevertheless, that because of the nature of our spend-
ing decisions—which reflect a commitment to get government right
and to promote the structural changes that lead to higher productivity
and more jobs—we can restore fiscal health while greatly improving
the micro-foundations of Canada’s economy. Let me emphasize that
the measures I have been describing are not merely a budget wish

212 P aul Martin

list. In this case, one of the advantages of Canada’s parliamentary
system—where the executive and legislative branches are one in the
same—is that the budget that is announced is also the budget that is
enacted (provided the government holds a majority of seats in the
House of Commons). In fact, our February budget was completely
passed into law by June of this year.

The public reaction to the 1995 budget has been favorable on the
whole, especially given that few recipients of federal spending were
left untouched. In particular, the reaction of affected interest groups
was muted, perhaps reflecting the fact that the measures were
carefully balanced and apparently considered to be equitable by the
great majority of Canadians. As for the financial markets, their
verdict was generally positive. Indeed, the 1995 budget probably
met or exceeded most expectations of what we would actually
deliver.

Indian & Northern Affairs

Citiz enship & Imm igrati on

Industry – S&T

Parliam ent & Gove rnor Gene ral

Canada Mortga ge & Housing

Vetera ns Affairs
Defense
Genera l Governmen t Services
Foreign Affairs & Int ’l Trade
Inte rnatio nal Assista nce

Natura l Resources

Fisheries & Oce ans
Environment
Human Re sourc es Deve lopmen t

Agricul ture

Transport
Industry – Departm ent
Regiona l Agenci es

Herit age & Cul tural Program s

The Canadian Experience in Reducing Budget Deficits and Debt

Changes in Federal Department Spending
1997-98 Relative to 1994-95

Chart 6

-60-80 -40 -20 0 20

Average

Percent change

Hea lth

Justic e
Sol icit or General

213

Given the history of federal governments, some in the financial
community remain skeptical that the government will stay the course
for as long as it takes. I can assure you, their skepticism is misplaced.
In fact, we believe that, overall, markets have been reassured by the
scope and nature of our actions. And while one must be wary of
attributing too much cause and effect, it may be indicative that the
Canada-U.S. Treasury bill spread has narrowed from the vicinity of
200 basis points last February to about 100 basis points now (mid-
August 1995) while the exchange rate has remained quite stable.

Let me turn now to how we managed to do all this—our strategy,
the politics, and some of the lessons that might be more broadly
applicable.

Our theme in the 1993 election was “Jobs and Growth,” and
reflecting our conviction that sound finances and a sound economy
go hand in hand, a key element of that platform was the 3 percent

Exchange rate

T-bill spread percentage points

65

60

675

1

Feb

Canada-U.S. Comparison
Chart 7

0
5

18

Exchange rate U.S. cents

2
3

470

Spread

1 15 1 15 29 12 26 10 24 7 54 19 2 1621
Jan Mar Apr May Jun AugJul

1995

214 P aul Martin

interim deficit target for 1996-97.6 This proved to be an essential
political anchor for our budget strategy. Without some fixed and
quantified target, any finance minister risks ending up on a slippery
slope. But with the unequivocal support of the prime minister, and
that target as a foothold, we were able to combine an economic
forecast with our fiscal model so as to quantify, in relatively unas-
sailable terms, the total amount of fiscal action required.

That made the politics manageable. Without a target to which we
were all irrevocably committed, the natural reluctance of minis-
ters—myself included—to accept cuts in their own domains would
likely have caused things to unravel. But once the fiscal target was
set, tradeoffs became inevitable and the only question was precisely
what those tradeoffs would be.

The budget strategy

Our broad budget strategy rested on three principal elements—
that is, setting our targets and assumptions; allocating the spending
cuts; and consulting with the public.

Short-term targets; prudent assumptions

We began by confirming the 3 percent deficit target for 1996-97,
sticking with our campaign commitment despite a fiscal mess that
turned out to be even worse than we had anticipated. For us, the
importance of maintaining that political anchor was uppermost. We
were also not content simply to state the 3 percent target for 1996-97.
So we committed to interim annual targets that would lead us there.
In fact, we have bettered the target for last fiscal year.

Foremost in our thinking was the need to restore the severely
damaged credibility of the government, which for years had been
over-promising and under-achieving its fiscal targets. Without credi-
bility, any positive market effects of a budget will be delayed as
skeptics adopt a wait-and-see attitude. So when planning the 1995
budget, we used forecasts of growth and interest rates that were
considerably more cautious than the private sector average.

The Canadian Experience in Reducing Budget Deficits and Debt 215

We also include in our deficit projections a “contingency reserve”
equal to about 1 percent of combined revenue and spending to buffer
nasty surprises arising in the economy. And significantly, if the
reserve is not required to hit our target, it will not be spent. It will
contribute instead to an even lower deficit.

We have also decided to adopt a two-year budget horizon—rolling
the second year’s target forward one year at a time. This is central
to our overall strategy. We have rejected the traditional approach
where typically a balanced budget would be projected five or more
years down the line. Frankly, that is political never-never land for
the simple reason that elections intervene before the magic date
arrives. Political accountability is lost and the bureaucracy can
safely put off the day when they really have to buckle down and find
the savings. The result, as we saw in Canada during at least the last
ten years, is a progression of missed targets, looming fiscal crisis,
and growing public cynicism.

With our two-year rolling targets—and our promise to hit them
“come hell or high water,” the situation is very different. Since the
targets are always staring us straight in the face, our feet are held to
the political fire. This keeps the goal uppermost in the cabinet’s mind
and puts maximum pressure on the program managers in the public
service to deliver promised savings.

The result is that we have been able to meet all of our targets to
date—something of a novelty for federal fiscal managers in recent
times—and we are totally committed to meeting them in the future.

We nevertheless still face pressure to abandon the strategy of
two-year rolling deficit targets and announce a firm date when the
budget will be balanced. We will resist that pressure and stick to our
game plan. Next February’s budget will include a deficit target for
1997-98 (that is, one year beyond the announced 3 percent target for
1996-97) together with the measures needed to achieve it. The
balanced budget target will be announced once we are confidently
within two years of its achievement.

216 P aul Martin

A frequently advocated alternative to the approach we have
adopted is to use “balanced budget” legislation—or even a consti-
tutional provision—to guarantee the fiscal responsibility of legisla-
tures. In our view, that is not the way to go. Apart from limiting the
choices of duly elected governments, this legalistic approach simply
encourages ingenious politicians and bureaucrats to spend time
looking for ways to get around the rules through accounting hocus-
pocus and subterfuges of various kinds. It seems to us that a simple
ironclad commitment to credible short-term targets is more demo-
cratic, and given every politician’s desire to avoid public oppro-
brium, more effective.

Program review—allocating the cuts

The second principal element of our fiscal strategy is a practical
procedure to address the seemingly intractable problem of allocating
spending cuts among departments and agencies. Even with a strong
collective commitment to a fiscal target, it is inherently difficult for
a large group of ministers to accept spending cuts that differ signifi-
cantly from ministry to ministry. The reason is that such variations
put the spotlight on differences in government priorities and put
individual ministers at risk of looking like losers to their constitu-
encies, their perceived losses being the stuff of headlines.

That is why all governments are tempted to resort to uniform cuts
across all programs. And while this can sometimes be justified in
the early stages of fiscal consolidation, it eventually becomes a
cop-out. Moreover, it is fraught with moral hazard since a policy of
uniform cuts destroys the incentive for individual departments to
become as lean and efficient as possible—that is, in the next round
of cuts, the keener would risk hitting bone while their lax counter-
parts would still have fat to slice.

We therefore rejected a uniform, across-the-board approach to
meeting our deficit targets. Instead, we launched a comprehensive
review of virtually all programs to identify those where a continuing
federal role was still justified and to find ways to deliver our services
more efficiently. The prime minister appointed a special committee

The Canadian Experience in Reducing Budget Deficits and Debt 217

of ministers to consider proposed departmental spending cuts. This
involved my colleagues directly in the tough job of examining
spending, line by line, and thus fostered an even stronger buy-in to
our budget goals. The process was disciplined by a firm requirement
that the individual spending cuts had to add up to a predetermined
level of savings needed to meet our budget targets. Once the depart-
mental amounts were ratified, it was left to individual ministers to
establish priorities within their own areas of responsibility so as to
achieve their sub-targets.

The program review exercise was completed in about six months
and produced agreement on departmental spending reductions total-
ing almost 20 percent from 1994-95 levels, spread over three years.
This represents an unprecedented, and in many ways revolutionary,
change in the way the Government of Canada operates. It is forcing
the government to focus sharply on those things—and only on those
things—it is in the best position to do.

The incentive to choose carefully is particularly powerful since
under our Expenditure Management System there are no longer any
“policy reserves” set aside to finance new initiatives. A new proposal
must therefore find funds from re-allocation of existing spending.

All of this is part of a process that we call “getting government
right” and, like the analogous process in the private sector, it is a job
that is really never finished. What is needed therefore is to inculcate
in government a culture of continuous improvement and continuous
assessment of priorities.

Public consultation—the open budget process

The third major element of our budget strategy has been to engage
the public—the experts, the interest groups, and the average citi-
zen—in dialogue as to the adequacy of our targets, the prudence of
our assumptions, and the potential fiscal measures themselves. In
Canada, by contrast with the practice in the United States, budget
secrecy has been very much the tradition. My predecessors began to
change that, and we have built on their efforts.

218 P aul Martin

Although the details of last February’s budget were kept confiden-
tial right up to budget day, we initiated the consultation process more
than four months in advance. It was kicked off with the release of
major background papers that launched an extensive round of public
hearings by the Finance Committee of the House of Commons.7

This proved to be a remarkably effective public education process,
both for the public and for us. Among other things, it stimulated an
out-pouring of detailed mock budgets by various interest groups,
media columnists, and individual citizens. Although there was no
clear consensus—except perhaps that our economic assumptions
should be prudent and that personal tax rates should not be increased—
the open budget process unearthed virtually every feasible option.

Overall, we believe that the consultation contributed importantly
to creating reasonable expectations as to the magnitude and the
general nature of the budget actions that were needed. This is surely
of great importance in building public understanding and support
for any ambitious program of fiscal consolidation.

We also took care to ensure that the budget was understood abroad.
Senior economic ministers traveled to the financial capitals—New
York, London, Tokyo—and were available on budget day to deliver
briefings and to answer questions directly on the economic and
political significance of what we were announcing simultaneously
in Ottawa.

To summarize—the principal elements of our budget strategy
were:

First, to set two-year rolling deficit targets backed up by an
ironclad political commitment to hit the targets and to base
fiscal forecasts on prudent economic assumptions further
supported by substantial contingency reserves;

Second, to establish an internal process with the authority
to allocate spending reductions among departments, reflect-
ing overall government priorities; and

The Canadian Experience in Reducing Budget Deficits and Debt 219

Third, to engage in wide-ranging pre-budget public con-
sultations.

Frankly, this neat ordering suggests a degree of logic and strategic
coherence that is more apparent in retrospect than it was as events
were unfolding in all of their uncertainty. For while it is true that we
tried to guide ourselves pretty much as I have just described, external
factors also played a key, and in some respects determinative role.

The most important of these factors arose from the macro-
economic climate during the pre-budget period from about Septem-
ber 1994 through budget day on February 27th of this year.
Successive bouts of currency volatility—particularly marked during
the Mexican peso crisis—put unexpected upward pressure on Cana-
dian interest rates. We had to deal with this just weeks before the
budget to ensure we had a credible plan to hit our upcoming target.
It also underlined our fiscal vulnerability and the loss of control that
comes from too much debt.

Mexico’s difficulties last winter were something of a wake-up
call. There we saw a concrete demonstration of a nation’s vulner-
ability to global financial markets. It was the kind of object lesson
that politicians find more compelling than hypothetical scenarios
from business economists, media pundits, and rating agencies.

The Mexican episode did influence the budget because it directly
affected something that had the potential to throw us off target—like
an unanticipated hike in Canada’s interest rates or a downbeat
change in the growth forecast. So, while the Mexican crisis clearly
fell into the category of real perturbations in the economy, Moody’s
pre-budget signal of a potential downgrade of our debt did not.

Once we set our fiscal target, the thing that matters above all else
is our absolute political commitment to hit that target. And the fact
that the targets are near-term means that we have had to react
immediately to events like Mexico. Had the deficit target instead
been several years off, we could easily have rationalized doing
nothing to correct our course. Over time, however, the consequences

220 P aul Martin

of such complacency tend to accumulate—one more reason why
targets end up being missed. Our approach avoids that risk.

What lessons might be drawn from all this? Looking back in
summary on last February’s budget, I believe it succeeded despite
the tough medicine, for basically the following reasons:

First and most fundamentally, the majority of Canadians
were already on-side with our general objective—in fact,
on the fiscal issue, the public was out in front of most
governments, a message many in our caucus brought home
to us loud and clear.

Second was the fact that the budget immediately won the
approval of opinion leaders thanks to the prudence of its
overall set of assumptions and to the structural quality of the
measures themselves.

Third, the actions in the budget were broadly seen to be
balanced and fair and to be generally responsive to the
public’s overall sense of priorities. Most of the credit for
this has to go to my cabinet colleagues who not only had to
make the sacrifices in their own ministries, but then had to
sell the overall justification to their constituencies.

Fourth, we achieved substantial fiscal savings while keep-
ing new taxes to a minimum and especially by ruling out
any personal income tax rate increases. The fact that we cut
back so heavily on our own activities, rather than putting
the burden of deficit reduction on the backs of taxpayers,
was a key plus. The biggest hits, as they would say in
Washington, were “inside the Beltway.”

Finally, we have had reasonable success in communicating
why action to deal with deficits and debt had to be the
government’s immediate priority and why this was not
inconsistent with our jobs and growth agenda—in fact,
quite the contrary.

The Canadian Experience in Reducing Budget Deficits and Debt 221

The 1995 budget is, of course, not the end of the story. The goals
of a significantly reduced debt ratio and a balanced budget are that
much closer, but still not achieved. There will be new interim deficit
targets to be set and further structural reforms to be implemented.
For example, we are already committed to reform the public pension
system to make it fairer and more sustainable. We will shortly be
introducing further structural reform of unemployment insurance.
Our program review continues.

The bottom line message here is clear. It is that our commitment
to stay the course of fiscal recovery is unequivocal, and the founda-
tions for that recovery are already solidly in place.

Reflecting, in conclusion, on the broader significance of what we
are going through, the 1994 and 1995 budgets were obviously much
more than cost-cutting exercises to get the markets off our back.
What we have really launched is a fundamental reappraisal of the
appropriate role of the national government.

The context for such a reappraisal is an increasingly interdependent
global economy where no nation, however powerful, can really
control even so basic a parameter as the exchange value of its currency.
The truth is that the limits on the ability of governments everywhere
to decree social and economic outcomes have become starkly apparent.

This has created a dissonance between what we have conditioned
our citizens to expect from their governments, and what govern-
ments are actually able to deliver. Contradictions abound.

For example, the public is increasingly skeptical that government
can directly create net new jobs, yet when the unemployment rate
rises, government is blamed.

There is a similar public skepticism about the ability of govern-
ment by itself to cure many of the social ills afflicting individuals
and communities. Yet it is in our compassionate nature—as that
nature has been conditioned over the past several decades—to still
expect government to set things right.

222 P aul Martin

We could go on citing examples like these. But the point is that
those of us who are committed to a pro-active role for government,
in both the social and economic domains, also have a responsibility
to begin distinguishing clearly those things governments can do
from those they can not. It’s time to come clean and stop creating
unrealistic expectations.

In Canada’s case—a highly sophisticated, yet relatively small and
open economy, heavily indebted to boot—these issues have particu-
lar salience and urgency. For us, globalization—whether of financial
markets or of economic competition—is simply a fact of life. The
real challenge we face is to turn globalization to our favor and to
maximize our freedom of action.

The only way to do this, it seems to us, is to put our fiscal house
in order, and to do all we can to boost productivity. This last point
is crucial because productivity is the foundation of competitiveness,
and international competitiveness is the only dependable route to
economic independence, growth, and jobs.

Seen in this light, our fiscal strategy is also a strategy to safeguard
Canada’s independence. But it is also true that the restraint associ-
ated with the strategy is leading to a government that is smaller, at
least by the measures of head count and spending volume. For some,
smaller government is an objective in itself. But for us, it is simply
a means to an end. We do believe that government should do only
what it can do best—and leave the rest for those who can do
better—whether business, labor, or the voluntary sectors.

What we must still achieve at the end of the day is a government
that is fully capable of assisting the disadvantaged; a government
that is unequivocally committed to our publicly-funded national
system of health care; a government that is more adept at providing
those things the private marketplace cannot—things such as strategic
support for aspects of science and technology; and a government
that is focused on getting the incentives right—whether to foster
environmental protection, to attract footloose investment, or to
spring people from the welfare trade and onto the job ladder.

The Canadian Experience in Reducing Budget Deficits and Debt 223

In broadest outline, the redefined role of government is becoming
clearer. In metaphorical terms, it is to be more like the tiller of a
sleek, modern sailboat than the paddle wheel of a nineteenth-century
steamer.

Yet achieving the transformation still poses a very large challenge.
This is because the habits and incentives of bureaucrats and politi-
cians, and the institutions they have created over the past fifty years,
have all been adapted to the fiscal growth of government. We have
not yet learned how to act as creatively as we must in the new
environment of static or shrinking financial resources. That con-
straint is also forcing us, as never before, to concentrate on the
setting of priorities, and on discovering how to do what is genuinely
needed without spending a lot of taxpayers’ money.

What is called for here is not only a change in attitude; it is a sea
change in the nature of politics as it has been practiced in the affluent
democracies over the past five decades. The job of getting govern-
ment right, or re-inventing government, or whatever the phrase is
much more than a slogan. Creating a public sector where it can truly
be said that “less is more” is the greatest challenge we face.

224 P aul Martin

Endnotes
1The debt figures in this paper are stated on a Public Accounts basis to accord with most

presentations of fiscal data in Canada. On a National Accounts basis—which corresponds to
OECD comparative presentations, and to U.S. budget data—net government debt in Canada
is about 64 percent of GDP.

2The federal government is currently responsible for approximately 47 percent of the
program spending of the federal and provincial governments combined, while the net federal
debt is about 73 percent of the combined total.

3Upon taking office in 1993, the government agreed formally with the Bank of Canada to
a target band for CPI inflation of 1 percent to 3 percent through 1998.

4Creating Opportunity: The Liberal Plan for Canada, 1993, p. 20.

5See A New Framework for Economic Growth, Government of Canada, October 1994.

6We have also set a deficit target of C$32.7 billion for 1995-96, but this has received much
less public attention than the 1996-97 target of 3 percent of GDP (estimated to be C$24.3
billion).

7A New Framework for Economic Growth and Creating a Healthy Fiscal Climate: The
Economic and Fiscal Update. The latter publication presented the fiscal implications of a range
of economic scenarios and derived from these the amount of fiscal action in each case needed
to achieve the designated deficit targets for 1995-96 and 1996-97. The document also contained
quite detailed information on departmental spending, tax expenditures, and revenue sources.

The Canadian Experience in Reducing Budget Deficits and Debt 225

CRSReport for Congress
Prepared for Members and Committees of Congress

Sovereign Debt in Advanced Economies:
Overview and

Issues for Congress

Rebecca M. Nelson
Analyst in International Trade and Finance

January 31, 2013

Congressional Research Service

7-5700
www.crs.gov

R41838

Sovereign Debt in Advanced Economies: Overview and Issues for Congress

Congressional Research Service

Summary
Sovereign debt, also called public debt or government debt, refers to debt incurred by
governments. Since the global financial crisis of 2008-2009, public debt in advanced economies
has increased substantially. A number of factors related to the financial crisis have fueled the
increase, including fiscal stimulus packages, the nationalization of private-sector debt, and lower
tax revenue. Even if economic growth reverses some of these trends, such as by boosting tax
receipts and reducing spending on government programs, aging populations in advanced
economies are expected to strain government debt levels in coming years.

High levels of debt in advanced economies are a relatively new global concern, after decades of
attention on debt levels in developing and emerging markets. Three Eurozone countries, Greece,
Ireland, and Portugal, have turned to the International Monetary Fund (IMF) and other European
governments for financial assistance in order to avoid defaulting on their loans. There are also
concerns about the sustainability of public debt in Japan and the United States.

To date, many advanced-economy governments have embarked on fiscal austerity programs (such
as cutting spending and/or increasing taxes) to address historically high levels of debt. This policy
response has been criticized by some economists as possibly undermining a weak recovery from
the global financial crisis. Others argue that the austerity plans do not go far enough, and that
more reforms are necessary to bring debt levels down, especially considering the aging
populations in many countries.

Issues for Congress

• Is the United States headed for a Eurozone-style debt crisis? Some economists and
Members of Congress fear that, given historically high levels of U.S. public debt, the
United States is headed towards a debt crisis similar to those experienced by some
Eurozone countries. Others argue that important differences between the United States
and Eurozone economies, such as growth rates, borrowing rates, and type of exchange
rate (floating or fixed), put the United States in a stronger position. The United States has
a long historical record of debt repayment, and bond spreads indicate that investors
currently view the United States as far less risky than Greece, Ireland, or Portugal.

• Impact on U.S. economy. The focus of most advanced economies on austerity programs
to lower debt levels could slow growth in advanced economies and depress demand for
U.S. exports. Financial instability stemming from high debt levels could also impact U.S.
markets and financial institutions.

• Policy options for Congress. Congress is debating proposals to reduce federal
debt levels in the United States. Congress could urge the Administration to
coordinate fiscal policies multilaterally to avoid simultaneous austerity measures
that undermine the economic recovery.

Sovereign Debt in Advanced Economies: Overview and Issues for Congress

Congressional Research Service

Contents
Introduction …………………………………………………………………………………………………………………….. 1
Background Definitions and Concepts ……………………………………………………………………………….. 3

Why and How Governments Borrow ……………………………………………………………………………. 3
How Sovereign Debt Differs from Private Debt …………………………………………………………….. 3
Why Governments Repay Debt, and Why They Default …………………………………………………. 4
Measuring Sovereign Debt ………………………………………………………………………………………….. 5

Trends in Sovereign Debt …………………………………………………………………………………………………. 7
Pre-Crisis: Vulnerabilities in Emerging Markets …………………………………………………………….. 7
Post-Crisis: Rising Debt Levels in Advanced Economies ………………………………………………… 8
Variation Among Advanced Economies ………………………………………………………………………… 9
Longer-Term Pressures in Advanced Economies ………………………………………………………….. 10
Challenges Posed by High Levels of Debt …………………………………………………………………… 11

Addressing High Debt Levels ………………………………………………………………………………………….. 12
Policy Options …………………………………………………………………………………………………………. 12

Fiscal Consolidation ……………………………………………………………………………………………. 12
Debt Restructuring ……………………………………………………………………………………………… 13
Inflation …………………………………………………………………………………………………………….. 14
Growth ………………………………………………………………………………………………………………. 15
Financial Repression …………………………………………………………………………………………… 16

Current Strategies …………………………………………………………………………………………………….. 16
Issues for Congress ………………………………………………………………………………………………………… 17

Is the United States Headed for a Eurozone-Style Debt Crisis? ………………………………………. 17
Implications for the U.S. Economy …………………………………………………………………………….. 19
Policy Options for Congress ………………………………………………………………………………………. 21

Figures
Figure 1. G-7 Public Debt as a Percentage of GDP, 1991 – present …………………………………………. 1
Figure 2. Public Debt Levels in Advanced Economies, Compared to Emerging and

Developing Economies, 2000-2017 …………………………………………………………………………………. 8
Figure 3. Gross General Government Debt in Advanced Economies, 2012 ……………………………… 9
Figure 4. Net General Government Debt in Advanced Economies, 2012 ………………………………. 10
Figure 5. Bond Spreads for Selected Advanced Economies …………………………………………………. 18
Figure 6. Direct Exposure of U.S. Banks to Advanced Economies ……………………………………….. 20

Tables
Table 1. Market Estimates of the Likelihood of Sovereign Defaults ……………………………………… 19
Table A-1. Gross General Government Debt in Advanced Economies, Actual and

Forecasts ……………………………………………………………………………………………………………………. 23
Table A-2. Net General Government Debt in Advanced Economies, Actual and Forecasts ………. 25

Sovereign Debt in Advanced Economies: Overview and Issues for Congress

Congressional Research Service

Table A-3. Direct Exposure of U.S. Banks to Advanced Economies …………………………………….. 27

Appendixes
Appendix. Data on General Government Debt and U.S. Bank Exposure Overseas …………………. 23

Contacts
Author Contact Information…………………………………………………………………………………………….. 27
Acknowledgments …………………………………………………………………………………………………………. 28

Sovereign Debt in Advanced Economies: Overview and Issues for Congress

Congressional Research Service 1

Introduction
In many advanced economies, the global financial crisis of 2008-2009 and ensuing recession
resulted in large fiscal stimulus packages, the nationalization of private-sector debt, lower tax
revenue, and higher government spending.1 These factors led to large budget deficits and
increased borrowing by governments from capital markets in order to fund these deficits. Figure
1 shows the rapid increase in public debt as a percentage of GDP in the major G-7 economies
(Canada, France, Germany, Italy, Japan, the United Kingdom, and the United States) following
the financial crisis. For the G-7 economies, sovereign debt rose from 84% of GDP in 2006 to a
forecasted 125% of GDP in 2012.

Figure 1. G-7 Public Debt as a Percentage of GDP, 1991 – present

Source: IMF World Economic Outlook, October 2012.

Concerns about high levels of public debt have been most focused on the Eurozone, where a debt
crisis has been ongoing since late 2009. 2 Three Eurozone countries—Greece, Ireland, and
Portugal—have had to borrow money from other European countries and the International

1 This report uses the IMF’s definition of advanced economies: Australia, Austria, Belgium, Canada, Cyprus, the Czech
Republic, Denmark, Estonia, Finland, France, Germany, Greece, Hong Kong, Iceland, Ireland, Israel, Italy, Japan,
Korea, Luxembourg, Malta, Netherlands, New Zealand, Norway, Portugal, Singapore, Slovak Republic, Slovenia,
Spain, Sweden, Switzerland, Taiwan, the United Kingdom, and the United States. According to the IMF, their
classification “is not based on strict criteria, economic or otherwise, and it has evolved over time.” The IMF uses three
major criteria to classify economies as advanced: (1) per capita income level; (2) export diversification; and (3) degree
of integration into the global financial system. For more information, see the Statistical Appendix of the IMF World
Economic Outlook (http://www.imf.org/external/pubs/ft/weo/2011/01/pdf/statapp ) and the IMF World Economic
Outlook Data Forum (http://forums.imf.org/showthread.php?t=154).
2 The Eurozone refers to the group of 17 European Union (EU) countries that uses the euro (€) as its national currency.
For more on the Eurozone crisis, see CRS Report R42377, The Eurozone Crisis: Overview and Issues for Congress,
coordinated by Rebecca M. Nelson.

Sovereign Debt in Advanced Economies: Overview and Issues for Congress

Congressional Research Service 2

Monetary Fund (IMF) in order to meet obligations and avoid defaulting on their debt. There are
also concerns about the sustainability of their public finances in Spain and Italy, much larger
economies in the Eurozone. However, concerns over rising debt levels are not limited to
Eurozone countries. Debates about the public debt are central features of political discourse in the
United States, Japan, and the United Kingdom, among others.

High levels of sovereign debt in advanced economies are of interest to Congress for a number of
reasons. First, the IMF has identified advanced economy debt as a possible threat to the global
economic recovery,3 as countries struggle to find a balance between growth and debt management
in an uncertain global economic recovery. Second, Congress has and continues to debate a
number of budget and debt issues, particularly in the context of the agreement reached on the debt
ceiling and federal budget. In many of these fiscal debates, parallels are drawn between the
United States and other advanced economies, such as Greece, Ireland, and the United Kingdom.
Analyzing debt levels and factors that shape debt sustainability can help inform these
comparisons. Third, how other countries reduce their debt impacts the U.S. economy. Most
advanced economies are implementing fiscal austerity programs to lower their debt levels.
Simultaneous austerity programs in the advanced countries, the United States’ major trading
partners, could depress demand for U.S. exports abroad, as well as deter investment in and from
advanced economies.

This report proceeds as follows:

• The first section provides background information on sovereign debt, including why
governments borrow, how sovereign debt differs from private debt, why governments
repay their debt (or not), and how sovereign debt is measured.

• The second section examines the shift of concerns over sovereign debt sustainability from
emerging markets in the 1990s and 2000s to advanced economies following the global
financial crisis of 2008-2009, and the challenges posed by high debt levels.

• The third section analyzes the different policy options governments have for lowering
debt levels. It also discusses the current strategy being used by most advanced
economies—fiscal austerity—and concerns that have been raised about its global impact.

• Finally, the fourth section analyzes issues of particular interest to Congress, including
comparisons between U.S. and European debt levels, how efforts to reduce debt levels
could impact the U.S. economy, and policy options available to Congress for engaging on
this issue.

3 Dominique Strauss-Kahn (then-Managing Director of the IMF), “Financial Crisis and Sovereign Risk: Implications
for Financial Stability,” remarks for IMF High-Level Roundtable, March 18, 2011,
http://www.imf.org/external/np/speeches/2011/031811.htm.

Sovereign Debt in Advanced Economies: Overview and Issues for Congress

Congressional Research Service 3

Background Definitions and Concepts

Why and How Governments Borrow
Sovereign debt, also called public debt or government debt, refers to debt incurred by
governments.4 Governments borrow money for a number of different reasons, such as health,
education, defense, infrastructure, and research. Some borrowing is for consumption, while other
borrowing is for investment. Governments may also borrow in order to run expansionary fiscal
policies (such as increasing spending or cutting taxes), with the goal of increasing economic
activity, spurring economic growth, and decreasing unemployment.

Most economists believe that public debt can play a productive role in the economy under certain
circumstances. They argue, for example, that borrowing by the government can help stimulate the
economy during a recession or fund long-term investment projects that increase economic output
in the future. However, they also caution that governments do not always use debt prudently.
Many economists argue that governments may be reluctant to increase taxes or cut spending
during economic booms in order to pay off debt incurred during economic downturns, leading to
growing debt levels over time. They also caution that governments may borrow for consumption
purposes, and that this type of borrowing can create difficulties when the debt obligation falls
due, because it does not yield future economic benefits.

Today, the governments of most advanced economies, as well as some emerging markets, borrow
money by issuing government bonds and selling them to private investors. They may sell bonds
to private investors overseas or to domestic investors. Some countries, such as Japan and Italy,
sell a sizeable portion of their bonds to investors at home. The United States does both, with
approximately half of its federal debt held by foreigners.5 Some emerging and developing
economies also borrow from other governments and international organizations, such as the
World Bank and the IMF.

How Sovereign Debt Differs from Private Debt
Sovereign debt differs from private-sector debt, or debt incurred by households and corporations,
for two reasons. First, there is no international bankruptcy court that can enforce debt contracts
between private investors and sovereign governments. In the domestic context, private borrowers
cannot simply refuse to repay debts to creditors. Domestic laws and courts can force debtors to
turn over existing assets to creditors or put the debtor through bankruptcy proceedings, during
which the borrower liquidates its assets and turns them over to the creditor. In the international
context, by contrast, there are no internationally accepted laws or bankruptcy courts to provide

4 Government debt, public debt, and sovereign debt are used interchangeably in this report. Sovereign debt is related to,
but different from, government deficits. A government deficit occurs when government spending exceeds government
revenue in a particular year. If spending is less than revenue, the government runs a surplus for that year. If the
government runs a deficit, it borrows to finance the deficit spending, and the deficit adds to the government’s overall
debt level. The deficit is a “flow” of borrowing that increases the “stock” of debt.
5 For data on foreign holdings of U.S. public debt, see U.S. Department of the Treasury, “Major Foreign Holdings of
Treasury Securities,” http://www.treasury.gov/resource-center/data-chart-center/tic/Documents/mfh.txt. For total U.S.
public debt, see U.S. Department of the Treasury, “Monthly Statement of the Public Debt of the United States,”
http://www.treasurydirect.gov/govt/reports/pd/mspd/mspd.htm.

Sovereign Debt in Advanced Economies: Overview and Issues for Congress

Congressional Research Service 4

creditors recourse against governments that refuse to repay their debts. Debt contracts between
governments and private creditors often include provisions that stipulate what jurisdiction’s law is
to be applied in the event of a dispute about the contract. However, there is no way to force a
government that has defaulted on its debt to abide by another country’s court ruling that it must
repay the loan.6 Proposals for creating internationally accepted bankruptcy proceedings and
regulations, possibly to be overseen by the IMF, have not been fruitful.7

A second reason why public debt differs from some private debt contracts is that sovereign debt is
“unsecured,” or not backed by collateral. Governments cannot credibly commit to turn over assets
if they are unable to repay their debts, because, again, there is no international authority to
compel them do to so. This contrasts with the private sector, where debt contracts are frequently
backed by collateral. For example, property serves as collateral for mortgages in most countries.
Some private-sector debt is not backed by collateral. Credit card debt, for example, is unsecured.

This is not to say that public debt is inherently more risky than private debt. In fact, some credit
rating agencies use the credit rating of the sovereign as an upper limit for the ratings that
domestic borrowers in that country can receive. However, the strict use of a sovereign credit
rating ceiling for domestic borrowers has waned in recent years.8 Sovereign debt may be less
risky than private-sector debt because governments have the power of taxation to raise money in
order to service debt, unlike private borrowers.

Why Governments Repay Debt, and Why They Default
If creditors have limited recourse against governments that default, why do they lend to
governments in the first place? It is generally argued that governments, even in the absence of
international bankruptcy court and secured debt contracts, will want to repay their debts in order
to build a good reputation in capital markets. Having a reputation for creditworthiness means that
the government can continue to borrow from investors at low interest rates, because investors
view the loan as having a low level of risk. If the government does not have a good reputation
with creditors, creditors will require high interest rates to compensate for the risk entailed in the
investment, or they will refuse to lend the government money at all.9 Some empirical evidence

6 However, parties can voluntarily submit to recourse, such as through the International Centre for Settlement of
Investment Disputes (ICSID), a branch of the World Bank Group. Investors can also use the threat of legal
“attachments” to prevent defaulted governments from re-entering capital markets until defaulted debt has been
resolved. Attachments refer to a legal process by which a court designates specific property owned by the debtor in
default to be transferred to the creditor. For more information, see Marcus Miller and Dania Thomas, “Sovereign Debt
Restructuring: The Judge, the Vultures and Creditor Rights,” The World Economy, vol. 30, no. 10 (2007), pp. 1491-
1509; and CRS Report R41029, Argentina’s Defaulted Sovereign Debt: Dealing with the “Holdouts”, by J. F.
Hornbeck.
7 Perhaps the most prominent proposal was in 2002, when then-IMF Deputy Managing Director Anne O. Krueger
proposed creating a “sovereign debt restructuring mechanism” to make the process of sovereign default and
restructuring of sovereign debts more predictable, smoother, and quicker. Anne O. Krueger, A New Approach to
Sovereign Debt Restructuring, International Monetary Fund, April 2002,
http://www.imf.org/external/pubs/ft/exrp/sdrm/eng/sdrm .
8 See, e.g., Eduardo Borensztein and Patricio Valenzuela, “The Credit Rating Agencies and the Sovereign Ceiling,”
Roubini, October 4, 2007, http://www.roubini.com/latam-
monitor/337/the_credit_rating_agencies_and_the_sovereign_ceiling.
9 It has also been argued that creditors are willing to lend to foreign governments because they believe their own
government will use military force to ensure repayment. However, the use of “gunboat diplomacy” to enforce debt
contracts is generally believed to have fallen out of practice in the early 20th century. See Martha Finnemore, The
(continued…)

Sovereign Debt in Advanced Economies: Overview and Issues for Congress

Congressional Research Service 5

also suggests that default can have adverse effects on international trade and economic growth,
providing other incentives for governments to repay their debts.10

Despite these incentives to repay debt, there is a long history of governments suspending debt
payments or falling behind on their debt payments, referred to as “defaulting” on their debt.11 A
“debt crisis” typically refers to a situation where a country is either unable or unwilling to pay its
debt. A debt crisis may not result in an actual default if, for example, the IMF lends the
government the money it needs to stay current in its debt obligations. However, many
governments that do default find an orderly way to restructure their debt that is acceptable to
markets. Debt restructuring refers to some reorganization of the debt, such as a reduction in
principal or lowering of interest rate, that makes debt payment easier for the borrower but still
entails some payments to creditors. A creditor may get less in a debt restructuring than was
originally agreed, but this may be preferable to getting nothing.

Defaults and debt crises can be triggered by a number of different economic and political factors,
including, but not limited to, economic recessions, fluctuations in the price of imports and
exports, currency depreciation (if debt is not payable in domestic currency), wars, and changes in
political leadership. Debates over why governments default are typically framed in terms of a
government’s “ability” to repay versus a government’s “willingness” to repay. For example, a
government may be unable to repay debt denominated in foreign currency if it does not have
sufficient access to foreign exchange. By contrast, a government may be unwilling to repay debt
incurred under a previous regime, even if it has the resources to do so.

Measuring Sovereign Debt
A nation’s debt burden is usually reported as a percentage of the country’s gross domestic product
(GDP), which indicates the size of the country’s economy. Scaling debt to the size of the
economy provides some indication of the government’s relative debt burden, since it is expected
that countries with bigger economies can sustain higher levels of debt in absolute terms than
smaller economies.

Data on sovereign debt are reported in a number of different ways. They can be reported for the
central government only, or for all levels of government (central/federal, state/province, and local
governments, often called “general government debt”). For countries with high levels of spending
by regional governments, such as Spain, there can be large differences between central
government debt and general government debt. By convention, the headline number cited in news
reports as the “U.S. debt” typically refers to the federal government debt only. Because of

(…continued)
Purpose of Intervention: Changing Beliefs about the Use of Force (Ithaca, NY: Cornell University Press, 2003). Also,
some argue that creditors are willing to lend to governments because they can seize the government’s assets overseas if
the government fails to repay. In practice, however, it is argued that governments have few assets in foreign
jurisdictions that can be seized by creditors, raising questions about the usefulness of that explanation. See Ugo
Panizza, Federico Sturzenegger, and Jeromin Zettelmeyer, “The Economics and Law of Sovereign Debt and Default,”
Journal of Economic Literature, vol. 47, no. 3 (2009), pp. 1-47.
10 E.g., see Andrew K. Rose, “One Reason Countries Pay Their Debts: Renegotiation and International Trade,” Journal
of Development Economics, vol. 77, no. 1 (2005), pp. 189-206; Eduardo Borensztein and Ugo Panizza, The Costs of
Sovereign Default, IMF Working Paper, WP/08/238, http://www.imf.org/external/pubs/ft/wp/2008/wp08238 .
11 See Carmen Reinhart and Kenneth Rogoff, This Time Is Different: Eight Centuries of Financial Folly (Princeton, NJ:
Princeton University Press, 2009).

Sovereign Debt in Advanced Economies: Overview and Issues for Congress

Congressional Research Service 6

different accounting practices in the European Union (EU), general government debt is typically
reported in news reports about EU debt levels. Some international organizations, such as the IMF
and the Organization for Economic Cooperation and Development (OECD), report debt data for
advanced economies that are standardized across countries.

Public debt can also be reported on a gross basis, referring to the government’s total liabilities, or
on a net basis, referring to the government’s total financial liabilities minus the government’s
financial assets.12 For governments with large financial assets, this can make a big difference.
Japan’s gross general government debt in 2012 is estimated to be 237% of GDP, but its net
general government debt was almost half that (135% of GDP).13 In contrast, forecasted Greece
gross general government debt equaled its forecasted net general government debt in 2012; both
were 171% of GDP.14

Finally, there is often interest in who holds the government’s debt: foreign or domestic investors.
As discussed above, some advanced economies sell most of their bonds to their citizens while
others sell to foreigners. Italy and Japan, for example, sell large portions of their bonds to
domestic investors. To address this issue, “external” public debt, or government debt owed to
foreign creditors, is sometimes distinguished from “domestic” public debt, or government debt
owed to domestic creditors.

Sovereign Debt Statistics: Key Terms
Level of Government

• General government debt: Debt for all levels of government (central/federal, state/province, and local
governments)

• Central government debt: Debt of the central government

Inclusion of Government Assets

• Gross government debt: The government’s total financial liabilities

• Net government debt: The government’s total financial liabilities minus the government’s total financial assets

Type of Creditor

• External public debt: Government debt owed to foreign creditors

• Domestic public debt: Government debt owed to domestic creditors

Many analysts warn that data on government debt should be used cautiously. They argue that
governments do not account properly for all their financial obligations, and that if these hidden
debts were included, estimates of government debts could be substantially higher.15 Data on
public debt levels are generally self-reported, and although there are international standards for
data reporting, governments have some discretion about what is included on their balance sheet.
For example, analysts caution that governments may not include their loan guarantees,

12 Financial assets of the government refer to non-physical assets, such as securities, certificates, or bank deposits that
belong to the government. Financial assets do not include all assets capable of being sold or activities capable of being
taxed.
13 IMF World Economic Outlook, October 2012.
14 Ibid.
15 See, e.g., Carmen Reinhart and Kenneth Rogoff, From Financial Crash to Debt Crisis, NBER Working Paper, No.
15795, March 2010, http://www.nber.org/papers/w15795 ; William Buiter, “The Debt of Nations,” Citi Investment
Research & Analysis, January 7, 2011.

Sovereign Debt in Advanced Economies: Overview and Issues for Congress

Congressional Research Service 7

obligations of state-owned enterprises, obligations of the central bank, or implicit guarantees in
their data reports. Some governments may also underreport data. In the Greek debt crisis, for
example, revelations of underreported budget deficits contributed to investor anxiety surrounding
the sustainability of Greece’s debt. Some economists also argue that some governments do not
fully account for spending on government programs for aging citizens, such as pensions and
health care, in their budget projections, leading to substantial underestimates of future debt
levels.16

Trends in Sovereign Debt

Pre-Crisis: Vulnerabilities in Emerging Markets
In the decades leading up to the global financial crisis of 2008-2009, concerns over sovereign
debt had been concentrated on middle-income, emerging-market countries. For example, the
1980s Latin American debt crisis, Russia’s financial crisis in 1998, and Argentina’s default in
2001 were major debt crises that received high levels of international attention and financial
support. Several emerging markets also restructured their debt in the late 1990s and 2000s,
including Russia, Ukraine, Pakistan, Ecuador, Argentina, Moldova, and Uruguay, among others.17

Emerging markets tended to be more susceptible to debt crises than advanced economies for a
number of reasons. High debt levels in some emerging markets, access to fewer resources to
repay debt, volatility in commodity prices, and weak political institutions are often cited as
factors. However, the structure of emerging-market debt contracts also made them more
vulnerable. Emerging-market debt tends to be denominated in foreign currencies, such as U.S.
dollars and euros, and tends to have shorter maturities.18 This made emerging markets vulnerable
to changes in the exchange rate, since a depreciation of the local currency could substantially
increase the amount of the debt in terms of local currency. Short debt maturities also impacted
their ability to “roll over” debt, or renew the loan upon maturity. Since the debt contracts were
short term, the debt had to be rolled over more frequently, which could be difficult if investors
lost confidence in the government. Advanced economies, by contrast, are able to borrow in
domestic currency (for example, the U.S. government borrows in U.S. dollars) and their debt
tends to have longer maturities. Because advanced economies do not bear exchange-rate risk and
can roll over their debt less frequently, sovereign debt in advanced economies had generally been
more stable than in emerging markets.

16 See, e.g., Stephen G. Cecchetti, M. S. Mohanty, and Fabrizio Zampolli, The Future of Public Debt: Prospects and
Implications, Bank for International Settlements (BIS), Working Paper No. 300, pp. 8-9,
http://www.bis.org/publ/work300 .
17 Federico Sturzenegger and Jeromin Zettelmeyer, Debt Defaults and Lessons from a Decade of Crises (Cambridge,
MA: MIT Press, 2007).
18 Barry Eichengreen, Ricardo Hausmann, and Ugo Panizza, Currency Mismatches, Debt Intolerance and Original Sin:
Why They Are Not the Same and Why It Matters, NBER Working Paper, No. 10036, October 2003,
http://www.nber.org/papers/w10036.

Sovereign Debt in Advanced Economies: Overview and Issues for Congress

Congressional Research Service 8

Post-Crisis: Rising Debt Levels in Advanced Economies
Since the financial crisis, concerns over sovereign debt sustainability have shifted from emerging
markets to advanced industrialized economies. In many advanced economies, the financial crisis
accelerated rising levels of sovereign debt (see Figure 1). Governments extended financial
support to troubled banks to stabilize the financial system, and enacted large stimulus packages to
boost demand, output, and employment. The ensuing recession resulted in lower tax receipts and
more government spending on programs such as unemployment insurance. All these factors
combined to create a substantial increase in government debt among advanced industrialized
countries. Some argue that if growth returned to the economy, debt levels would fall due to rising
tax receipts and lower spending on programs such as unemployment insurance. Long-term trends,
however, suggest that aging populations could strain public finances in advanced economies in
coming years, and that public debt levels could continue to be a problem.

Figure 2. Public Debt Levels in Advanced Economies, Compared to Emerging and
Developing Economies, 2000-2017

Source: IMF World Economic Outlook, October 2012.

Note: Gross general government debt. See Table A-1 for more detailed data.

Specifically, Figure 2 shows that gross general government debt in advanced economies
increased slightly in the years before the global financial crisis, from 72% of GDP in 2000 to 77%
of GDP in 2006. During the financial crisis, however, sovereign debt levels rose more rapidly, to
105% of GDP in 2011. The IMF forecasts that they will continue to increase through 2014, to
113% of GDP, before falling subsequently.

In contrast, debt levels in emerging markets and developing countries were lower than those in
advanced economies in 2000. Emerging-market and developing-country gross general
government debt was only 49%, compared to 72% of GDP in advanced economies. Moreover,
emerging-market and developing-country debt has fallen fairly steadily over the past decade,
from 52% of GDP in 2002 to 36% of GDP in 2011. By 2017, the IMF predicts that debt in
emerging and developing countries will fall even further to 28% of GDP.

Sovereign Debt in Advanced Economies: Overview and Issues for Congress

Congressional Research Service 9

Variation Among Advanced Economies
Although public debt has generally been rising in advanced economies, there is wide variation
among debt levels in advanced economies. Figure 3 shows forecasted gross general government
debt across advanced economies as a percentage of GDP in 2012. In that year, Japan is estimated
to have the highest ratio of gross general government debt relative to GDP, at 237% of GDP. The
second highest was Greece, at 171% of GDP. Estonia had the lowest level, at only 9% of GDP.
The United States ranked sixth among advanced economies, just after Ireland and before
Singapore, with an estimated gross general government debt of 107% of GDP. It is also worth
noting that the three Eurozone countries experiencing the most severe market pressure—Greece,
Ireland, and Portugal—have among the highest debt-to-GDP ratios among advanced economies,
but other advanced economies, such as Japan, are facing much less market pressure. This fact
highlights that markets consider a variety of indicators, not just debt levels, when evaluating a
government’s debt sustainability.

Figure 3. Gross General Government Debt in Advanced Economies, 2012

Source: IMF World Economic Outlook, October 2012.

Note: Forecasts. See Table A-1 for more data on gross general government debt in advanced economies.

Net general government debt presents a slightly different ranking of debt levels among countries,
as shown in Figure 4. By this measure, Greece is forecasted to be the most indebted economy in
2012, with a net general government debt of 171% of GDP, followed by Japan with 135% of
GDP. Some countries, such as Norway and Finland, have negative net general government debt
levels, because their financial assets are larger than their financial liabilities. U.S. net general
government debt is estimated to be 84% of GDP in 2012. Of the countries where data on net debt
are available, it ranked sixth among advanced economies, with net general government debt
higher than France’s but lower than Ireland’s.

Sovereign Debt in Advanced Economies: Overview and Issues for Congress

Congressional Research Service 10

Figure 4. Net General Government Debt in Advanced Economies, 2012

Source: IMF World Economic Outlook, October 2012.

Note: Forecasts. A negative net general government debt indicates that the government’s total financial assets
exceed the government’s total financial liabilities. See Table A-2 for more data on net general government debt
in advanced economies.

Longer-Term Pressures in Advanced Economies
Some economists caution that the rise in sovereign debt among advanced economies during the
financial crisis is only the start of more serious debt problems to come.19 Specifically, there is
concern that aging populations in many advanced economies will cause public debt to skyrocket,
as a shrinking workforce will result in lower tax revenue while more retirees will require an
increase in government spending on pensions and healthcare. Among OECD countries, for
example, there were about 27 retirees for every 100 workers in 2000.20 By 2050, the OECD
forecasts about 62 retirees for every 100 workers.21

Economists at the Bank for International Settlements (BIS) suggest that the unfunded contingent
liabilities associated with aging populations in advanced economies have not been definitively or
comprehensively accounted for in government balance sheets or in budget and debt projections.22
These economists argue that properly accounting for increases in age-related spending would
result in significantly higher forecasts of debt levels. According to their calculations, debt-to-GDP

19 See, e.g., Brian Keeley, “Debt—You Ain’t Seen Nothing Yet,” OECD Insights, April 1, 2010,
http://oecdinsights.org/2010/04/01/debt-%E2%80%93-you-ain%E2%80%99t-seen-nothing-yet/.
20 “Ratio of the Inactive Elderly Population Aged 65 and Over to the Labour Force,” OECD Factbook,
http://puck.sourceoecd.org/vl=1955252/cl=13/nw=1/rpsv/factbook2009/01/02/01/01-02-01-g1.htm. Retirees are
defined as inactive elderly population over 65 years old.
21 Ibid.
22 Stephen G. Cecchetti, M. S. Mohanty, and Fabrizio Zampolli, The Future of Public Debt: Prospects and
Implications, Bank for International Settlements (BIS), Working Paper No. 300, pp. 8-9,
http://www.bis.org/publ/work300 .

Sovereign Debt in Advanced Economies: Overview and Issues for Congress

Congressional Research Service 11

ratios could rise by 2020 to 300% of GDP in Japan, 200% of GDP in the United Kingdom, and
150% of GDP in Belgium, France, Ireland, Greece, Italy, and the United States. These forecasts
are based on a number of assumptions about government policies, interest rates, and growth, and
changes in these assumptions could result in very different (higher or lower) estimates.

Challenges Posed by High Levels of Debt
Historically high public debt levels in advanced economies, combined with the threat of
additional debt increases due to age-related spending, have become a source of serious concern
for a number of reasons. First, they make governments vulnerable to unexpected and quick
changes in investor behavior. If investors begin to fear that the government may default on its
existing debt obligations, they may start demanding higher interest rates or refuse to lend to the
government at all.23 Loss of market access or interest rates that are no longer affordable could
cause the government to make quicker and more drastic policy adjustments, including tax
increases and/or spending cuts, than would have been necessary otherwise. To date, three
Eurozone countries (Greece, Ireland, and Portugal) have come under this type of market pressure.
However, all three of these countries borrowed money from other European countries and the
IMF in order to continue to make debt repayments, while providing time for the governments to
implement reforms that improve each country’s fiscal position and regain investor confidence.

Second, government competition for loans can increase interest rates when the economy is at full
employment, causing private investment to fall. Because private investment is important for long-
term economic growth, government budget deficits tend to reduce the economic growth rate. This
phenomenon is often referred to as public debt “crowding out” private investment. However, to
the extent that government deficits finance public investment, there may not be any necessary
detriment to the rate of economic growth. Borrowing from foreign investors can help maintain
domestic investment and mitigate the problems associated with crowding out, although this
creates obligations for profits to flow overseas in the future.24

Third, high debt levels restrict the ability of the government to respond to unexpected crises, such
as natural disasters. As debt levels rise and investors become more concerned about the
sustainability of the debt in the country, the government may find that it cannot access the
financing it needs to address a crisis, and that it has to rely on other policy tools or on
international support. For example, governments with high debt may be more constrained in their
ability to employ policy tools to blunt the impact of economic downturns. Japan may be facing
such a situation, as its borrowing needs have increased for financing reconstruction following the
earthquake, tsunami, and ensuing nuclear crisis in March 2011.25 At least one credit rating agency,
S&P, has expressed concern about Japan’s plan to increase deficit spending.

23 If investors think that the government will use inflation to reduce the value of the debt in terms of goods and
services, they may demand: higher interest rates, inflation-indexed bonds (bonds where the principal is indexed to
inflation), or denomination of the bond in a different currency.
24 CBO, Federal Debt and the Risk of a Fiscal Crisis, July 27, 2010, http://www.cbo.gov/ftpdocs/116xx/doc11659/07-
27_Debt_FiscalCrisis_Brief .
25 For more on Japan’s earthquake and tsunami, see CRS Report R41702, Japan’s 2011 Earthquake and Tsunami:
Economic Effects and Implications for the United States, coordinated by Dick K. Nanto.

Sovereign Debt in Advanced Economies: Overview and Issues for Congress

Congressional Research Service 12

Addressing High Debt Levels
Given the problems that persistent high levels of public debt can create, there has been focus on
how governments can lower debt levels. Governments have five major policy tools at their
disposal for addressing high debt levels: fiscal consolidation, debt restructuring, inflation, growth,
and financial repression.26 The pros and cons of each strategy are analyzed below, as well as the
current strategies being pursued by the governments of advanced economies to address their debt
levels.

Addressing High Debt Levels: Policy Options
• Fiscal consolidation: Using tax increases and/or government spending cuts to reduce government deficits and

lower government borrowing. In this report, “fiscal consolidation” is used interchangeably with “fiscal austerity.”

• Debt restructuring: Renegotiating the debt contract to lower payments for the borrower. This can take a
number of forms, such as lowering the interest rate, extending the repayment period (maturity of the loan), and
lowering the outstanding balance (principal) of the loan.

• Inflation: Using inflation to reduce the “real” value of the debt, meaning the value of the loan in terms of goods
and services. If there is inflation, the nominal or face value of the loan purchases fewer goods and services than
at the time the debt contract was agreed upon.

• Growth: Pursing reforms, such as increasing the flexibility of labor markets, in order to spur growth. Increasing
growth lowers debt relative to GDP.

• Financial repression: Government policies that induce or force domestic investors to buy government bonds
at artificially low interest rates. All else being equal, when real interest rates (the interest rate adjusted for
inflation) are negative, debt-to-GDP falls.

Policy Options

Fiscal Consolidation

A government may lower high levels of sovereign debt through austerity or fiscal consolidation,
which generally refers to policies that reduce the government budget deficit. These include tax
increases, spending cuts, or some combination of the two.

Some argue that austerity programs are effective at reducing the debt by directly targeting the
cause of high debt levels: government spending that is too high or tax revenue that is too low.
Proponents of fiscal consolidation also argue that it can increase economic growth. They argue,
for example, that credible commitments to austerity measures can increase investor confidence in
the government and lower the interest rate charged by investors on government bonds. If lower
borrowing costs for the government also reduce interest rates for consumers and firms, consumer

26 For example, see, “Locking Up Your Money,” The Economist, May 4, 2011. Also note that some analysts have
suggested that Greece should respond to its debt crisis by defaulting on its debt and depreciating its currency, similar to
Argentina in the early 2000s. They argue that it would stimulate exports and spur economic growth. However, Greece’s
debt is denominated in euros, and leaving the Eurozone in favor of a depreciated national currency would significantly
raise the value of its debt in terms of national currency. Since it is not a strategy for reducing debt levels, this policy
option is not discussed in this report. However, for more on the consequences of Greece exiting the Eurozone, see CRS
Report R41411, The Future of the Eurozone and U.S. Interests, coordinated by Raymond J. Ahearn; and CRS Report
R41167, Greece’s Debt Crisis: Overview, Policy Responses, and Implications, coordinated by Rebecca M. Nelson.

Sovereign Debt in Advanced Economies: Overview and Issues for Congress

Congressional Research Service 13

spending and investment may increase, expanding economic output.27 It is also argued that fiscal
consolidation can be expansionary (or increase economic growth) if it lowers expectations of
future taxes, encouraging private spending.28

Many economists agree, however, that these programs are costly to implement. They argue that
austerity policies reduce aggregate demand in the short term, causing the economy to contract and
unemployment to increase.29 This is why austerity programs are often also called contractionary
policies. Additionally, they argue that if economic output falls at a faster rate than the debt does,
the ratio of debt to GDP can actually rise, failing to address effectively the government’s debt
burden. Finally, austerity programs can be politically difficult to implement, as anti-austerity
protests in a number of countries, including Belgium, Greece, Ireland, and Spain, among others,
demonstrate.30

Some contend that financial assistance from other governments or the IMF can help ease the
contractionary effects of fiscal consolidation, by allowing austerity reforms to occur over a longer
time frame than they would otherwise have to occur. However, because the financial assistance
provided by the IMF, as well as from the European countries in the case of Greece, Ireland, and
Portugal, takes the form of loans, others argue that this financial assistance only exacerbates the
country’s problems and leads to ever-increasing debt levels.

Debt Restructuring

Debt restructuring refers to reorganizing a debt that has become too large and burdensome for the
borrower to manage.31 It can refer to providing more lenient terms about how a debt will be
repaid, such as extending the time period over which the debt will be repaid (the maturity of the
loan) or lowering the interest rate. It can also refer to a reduction in (or forgiveness of some of)
the outstanding balance or principal. In either case, it means that the current owners of the debt
get less than they were originally promised.32 Debt restructurings are unusual but not
unprecedented. Several emerging markets restructured their debt in the late 1990s and 2000s,
including Russia and Argentina, among others.33

Proponents of debt restructuring argue that it is a way for governments to reduce their debt
burden while limiting the austerity measures imposed on their citizens. Instead, it pushes the cost

27 For example, see Alberto Alesina, Fiscal Adjustments: Lessons from Recent History, Working Paper, April 2010,
http://www.economics.harvard.edu/faculty/alesina/files/Fiscal%2BAdjustments_lessons .
28 For example, see Francesco Giavazzi and Marco Pagano, Can Severe Fiscal Contractions be Expansionary? Tales of
Two Small European Countries, NBER Working Paper, No. 3372, May 1990, http://www.nber.org/papers/w3372.
29 For example, see discussion of Keynesian economics in International Monetary Fund, World Economic Outlook,
October 2010, Chapter 3: “Will It Hurt? Macroeconomic Effects of Fiscal Consolidation,”
http://www.imf.org/external/pubs/ft/weo/2010/02/pdf/c3 .
30 For example, see “European Cities Hit by Anti-Austerity Protests,” BBC News Europe, September 25, 2010,
http://www.bbc.co.uk/news/world-europe-11432579.
31 Lex Rieffel, Restructuring Sovereign Debt: The Case for Ad Hoc Machinery (Washington, DC: Brookings Institution
Press, 2003).
32 Martin Feldstein, “Why Greece Will Default,” Business Insider, April 10, 2010,
http://www.businessinsider.com/why-greece-will-default-2010-4.
33 Federico Sturzenegger and Jeromin Zettelmeyer, Debt Defaults and Lessons from a Decade of Crises (Cambridge,
MA: MIT Press, 2007).

Sovereign Debt in Advanced Economies: Overview and Issues for Congress

Congressional Research Service 14

of debt reduction onto private creditors, who, some argue, should bear the consequences of taking
on higher risk in exchange for greater potential reward.34

Critics argue that restructuring is not a desirable option for lowering debt burdens. Economists at
the IMF, for example, have said that debt restructuring among advanced economies is
“unnecessary, undesirable, and unlikely.”35 For some advanced economies, a large share of
government debt is held domestically. This suggests that imposing losses on private creditors
instead of implementing austerity measures may not shield the government from domestic
backlash. Also, governments may have trouble borrowing from capital markets after restructuring
their debt, meaning that they would need to bring their government budgets into balance or
surplus more quickly than if they had not restructured.36 Or they may face higher interest rates,
making borrowing more costly.

The logistics of debt restructuring can also be difficult. Organizing and negotiating with
potentially thousands of individual bondholders can be cumbersome and time-consuming,
although recent changes in the legal processes related to sovereign bonds have helped streamline
restructuring.37 Finally, debt restructuring may be undesirable because it can increase investor
anxiety and cause the crisis to spread to other countries. For example, with the Eurozone crisis,
the European countries and the IMF are working to keep the crisis from spreading from the
relatively small economies of Greece, Ireland, and Portugal to larger economies in the region,
including Spain, Italy, or Belgium.

Inflation

If sovereign debt is denominated in the domestic currency, the government can use inflation to
reduce the real value of the debt. This is frequently referred to as a government “running the
printing presses” in order to create the money it needs to repay creditors, although there are other
ways the government can create inflation in the economy. Many economists view this policy as
an effective default on the debt, because even if creditors are repaid, the value of goods and
services they can purchase is significantly lower than what they expected when they extended the
loan to the government. Inflation has not featured prominently in recent major emerging-market
debt crises because most emerging-market debt tends to be denominated in foreign currencies.38

34 Arvind Subramanian, “Greek Deal Lets Banks Off the Hook,” Financial Times, May 6, 2010.
35 Carlo Cottarelli, Lorenzo Forni, and Jan Gottschalk, et al., Default in Today’s Advanced Economies: Unnecessary,
Undesirable, and Unlikely, IMF Staff Position Note, September 1, 2010,
http://www.imf.org/external/pubs/ft/spn/2010/spn1012 .
36 However, some argue that governments that have defaulted can regain access to capital markets by successfully
concluding a debt restructuring. See Federico Sturzenegger and Jeromin Zettelmeyer, Debt Defaults and Lessons from
a Decade of Crises (Cambridge, MA: MIT Press, 2007), pp. 50.
37 Specifically, the inclusion of “collective action clauses” (CACs) in sovereign bonds, which became popular in the
2000s, has helped expedite the restructuring process. CACs allow a supermajority of bondholders (usually 75%) to
agree to a debt restructuring that is legally binding on all bondholders. Without CACs, some bondholders may have
incentives to try to hold out for better terms, slowing down the negotiations. For more information on CACs, see, for
example, Federal Reserve Bank of San Francisco, “Resolving Sovereign Debt Crises with Collective Action Clauses,”
Economic Letter No. 2004-06 (February 20, 2004), http://www.frbsf.org/publications/economics/letter/2004/el2004-
06 .
38Inflation has reduced the real value of domestic public debt in some emerging economies in recent decades, including
Argentina, Brazil, and Turkey in the late 1980s and 1990s. Domestic public debt crises, however, tend to garner less
international attention than external public debt crises. There is a long history of countries using inflation to address
(continued…)

Sovereign Debt in Advanced Economies: Overview and Issues for Congress

Congressional Research Service 15

Inflation allows a government to repay its debt without having to implement austerity measures,
and can be less complicated than a debt restructuring. Using inflation as part of a debt
management strategy, however, can be problematic. The inflation has to be unexpected to
investors, or else investors will price in the risk of inflation through higher interest rates. Even if
the government is able to introduce surprise inflation, it will raise the government’s borrowing
costs in the future. Inflation can also have a number of adverse consequences, including wiping
out the value of savings, creating shortages of goods, and reducing future investment by creating
uncertainty in the economy. Governments may also have trouble limiting the amount of inflation
introduced into the economy: one round of inflation may raise expectations about future inflation,
which in turn could lead to more inflation. Additionally, using inflation to lower the real value of
the debt assumes the cooperation of the central bank, but in most advanced economies, the central
bank sets policies independently of the government. Finally, using inflation to address a debt
problem is not available to countries whose debt is denominated in a currency held jointly.
Individual Eurozone countries issue debt denominated in euros, but they do not have control over
monetary policy in the Eurozone and cannot use inflation to reduce the real value of their debt.

Growth

Economic growth also allows governments to lower the size of their debt relative to the size of
their economy (typically measured as gross domestic product [GDP]). It can also lead to lower
levels of government spending and increase tax revenues, lowering the dollar value of sovereign
debt as well. In the short run, economic stabilization is a necessary condition for sustained
economic growth. Growth can be stimulated by pursuing expansionary fiscal and monetary
policies or by pursuing structural reforms at the microeconomic level. Expansionary fiscal
policies, however, lead to more debt, and “easy” monetary policies, such as lowering interest
rates, may not be effective if firms and households are unwilling to borrow to increase investment
and consumption. At the microeconomic level, growth can be supported by a number of structural
reforms that can increase the competitiveness of industries in the economy. Examples include
removing barriers to labor mobility, privatizing state-owned companies, and liberalizing trade
policy. The IMF’s program for Greece, for example, includes structural reforms aimed at
encouraging growth.

The benefit of growing out of debt is that it allows countries to address their debt problems
without possibly painful fiscal cuts or alienating creditors. However, the results of these reforms
tend to manifest themselves over the long term, and a country already in a debt crisis may have
difficulty just “growing out of it” in the short term. Moreover, empirical evidence suggests that
countries with high levels of debt have trouble growing.39 The uncertainty around growth as a
strategy for short-term debt reduction is one reason why Greece’s IMF program does not just
include structural reforms; fiscal cuts are also a central component.

(…continued)
debt levels, with numerous examples from medieval Europe and as far back as Greece in fourth century B.C. See
Carmen Reinhart and Kenneth Rogoff, This Time Is Different: Eight Centuries of Financial Folly (Princeton, NJ:
Princeton University Press, 2009), chapter 8, “Domestic Debt: The Missing Link Explaining External Default and High
Inflation,” and chapter 11, “Default Through Debasement: An ‘Old World’ Favorite.”
39 Carmen Reinhart and Kenneth Rogoff, “Growth in a Time of Debt,” American Economic Review, vol. 100, no. 2
(May 2010).

Sovereign Debt in Advanced Economies: Overview and Issues for Congress

Congressional Research Service 16

Financial Repression

Some economists argue that governments can also use financial repression to lower debt levels.40
“Financial repression” generally refers to the use of government policies to induce or force
domestic investors to buy government bonds at artificially low interest rates. Specifically, they
sell bonds at interest rates below the rate of inflation, meaning that the real interest rate (the
interest rate adjusted for inflation) is negative. All else being equal, extending loans with negative
real interest rates results in falling debt-to-GDP ratios over time. In order to get investors to buy
these bonds, governments use a host of policies, such as restrictions on the outflow of capital, to
create a captive domestic audience for these bonds. For example, governments may require
pension funds to hold government bonds.

Empirical evidence indicates that financial repression was used by several advanced economies to
lower public debt levels following World War II.41 It is estimated that real interest rates in
advanced countries were negative roughly half the time between 1945 and 1980. Some
economists estimate that, in the United States and United Kingdom, financial repression helped
reduce debt levels by 3%-4% of GDP a year, or 30% to 40% each decade between the end of
World War II and the 1970s.

Financial repression may be attractive because it avoids many of the pitfalls of the other policy
options for lowering debt levels: it avoids politically painful austerity measures, is arguably less
disruptive than debt restructuring, does not require introducing surprise inflation into the
economy, and is a more certain policy option than growth. Thirty years of financial liberalization,
however, have made it technically difficult for governments to return to the capital controls
necessary to embark on a policy agenda of financial repression.42 Policy-makers may also have
trouble imposing the controls before capital flight takes place, and the controls could damage a
country’s ability to attract foreign investment. Financial repression may also be politically
difficult, as investors would likely oppose policies that restrict their investment opportunities or
require them to buy government bonds at artificially low interest rates.

Current Strategies
The primary policy response across advanced economies to historically high debt levels has been
fiscal austerity. Several advanced economies have announced austerity measures, some, such as
Greece, Ireland, and Portugal, in response to market pressures, and others, such as the United
Kingdom, ahead of changes in investor sentiment. At the G-20 summit in June 2010 in Toronto,
governments of advanced economies pledged to halve deficits by 2013 and stabilize or reduce
government debt-to-GDP ratios by 2016.43 However, G-20 commitments are not binding, and
there has been little discussion of commitments for fiscal consolidation in subsequent G-20
summits.

40 For example, see Carmen Reinhart and M. Belen Sbrancia, The Liquidation of Government Debt, NBER Working
Paper, No. 16893, March 2011, http://www.imf.org/external/np/seminars/eng/2011/res2/pdf/crbs . Also see Gillian
Tett, “Policymakers Learn a New and Alarming Catchphrase,” Financial Times, May 9, 2011.
41 Ibid.
42 “Locking Up Your Money,” The Economist, May 4, 2011.
43 For more on the G-20, see CRS Report R40977, The G-20 and International Economic Cooperation: Background
and Implications for Congress, by Rebecca M. Nelson. For the Toronto summit declaration, see
http://www.g20.org/Documents/g20_declaration_en .

Sovereign Debt in Advanced Economies: Overview and Issues for Congress

Congressional Research Service 17

One notable exception is Japan, whose plans to consolidate finances was derailed in March 2011
by an earthquake, tsunami, and ensuing nuclear crisis. The physical damage is estimated to be
between $195 billion and $305 billion, and the crisis created the need for immediate spending to
cover reconstruction costs. As the government borrows to finance reconstruction, Japan’s public
debt, which is already one of the highest among advanced economies, is expected to rise.

In addition to austerity measures in most advanced economies, in March 2012, the Greek
government implemented what is being called the largest debt restructuring in history. About 97%
of privately held Greek bonds (about €197 billion, or about $256 billion) took a 53.5% cut to the
face value (principal) of the bond, and the net present value of the bonds was reduced by
approximately 75%. However, even with the restructuring, a sharp economic contraction in
Greece has exacerbated public finances, and its debt to GDP ratio is forecasted by the IMF to be
171% of GDP in 2012, which many economists fear is still too high. Further debt relief,
particularly from “official” creditors (other European governments), could be necessary.

Some economists are concerned that the emphasis on austerity measures could undermine a
fragile global economic recovery, particularly while unemployment remains high in many
advanced economies. Others argue that the announced fiscal consolidation plans do not go far
enough, and are concerned that there will not be sufficient political will to undertake the reforms
necessary to stabilize and reduce debt levels over the long term.

Issues for Congress

Is the United States Headed for a Eurozone-Style Debt Crisis?
Some analysts,44 as well as some Members of Congress, have expressed concern that the United
States is headed towards a debt crisis similar to those experienced by some Eurozone countries,
including Greece, Ireland, and Portugal. They are concerned about loss of investor confidence
and the loss of the United States’ ability to borrow at reasonable interest rates. Like these
Eurozone countries, it is argued, the United States has been reliant on foreign investors to fund a
large budget deficit, resulting in rising debt levels and increasing vulnerability to a sudden
reversal in investor confidence. S&P’s downgrade of long-term U.S. debt in August 2011
reinforced concerns about the U.S. commitment and ability to repay its debt.

Other economists argue that the U.S. debt position is much stronger than that of the Eurozone
economies in crisis.45 Unlike Greece, Portugal, and Ireland, the United States has a floating
exchange rate and its currency is an international reserve currency, which can alleviate many of
the pressures associated with rising debt levels.46 Additionally, they argue that the stronger levels
of economic growth and the lower borrowing costs of the United States put U.S. debt levels on a
more sustainable path over time. Even with the S&P downgrade, the U.S. credit rating is still
higher than the crisis countries. The United States also has a strong historical record of debt

44 For example, see Niall Ferguson, “A Greek Crisis is Coming to America,” Financial Times, February 10, 2010.
45 For example, see Paul Krugman, “We’re Not Greece,” New York Times, May 10, 2010.
46 For example, a depreciation in the dollar relative to other countries can bolster exports and spur growth, offsetting
the effects of austerity. Likewise, the dollar’s status as an international reserve makes it a safe haven for investments
during times of distress or crisis, bolstering demand for government bonds even as debt levels are rising.

Sovereign Debt in Advanced Economies: Overview and Issues for Congress

Congressional Research Service 18

repayment that helps bolster its reputation in capital markets. Greece, by contrast, has been in a
state of default about 50% of the time since independence in the 1830s.47

Bond market data indicate that investors do not view the United States in a similar light to
Greece, Ireland, or Portugal. Figure 5 compares the spreads on Greek, Irish, Portuguese, U.S.,
and UK 10-year bonds (over 10-year German bonds) since 2008. Higher bond spreads indicate
higher levels of risk. U.S. bond spreads have remained substantially lower than Greek, Irish, and
Portuguese bond spreads throughout the Eurozone crisis. U.S. bond spreads have been much
closer in value to UK bond spreads, even during the financial crisis that originated in the U.S.
housing market.

Figure 5. Bond Spreads for Selected Advanced Economies
Spreads on 10-year bonds over German 10-year bonds

Source: Global Financial Data.

Additionally, one market research firm (Credit Market Analysis, CMA) estimates the likelihood
of default over the next five years for a number of governments, and publishes the top 10 most
and least risky sovereigns on a quarterly basis. For the fourth quarter of 2012, it estimated the
likelihood of the United States defaulting on its debt over the next five years to be 3.30%, and
ranks the United States as the fifth least-likely country to default. By contrast, Cyprus, Portugal,
and Spain are all ranked in the top 10 countries most likely to default. (Data was not availalbe for
calculating Greece’s probability of default; see note in Table 1.) CMA estimates that Cyprus is
more likely to default than not over the next 5 years, with an estimated probability of default of
60.5%.

47 See Carmen Reinhart and Kenneth Rogoff, This Time Is Different: Eight Centuries of Financial Folly (Princeton, NJ:
Princeton University Press, 2009), Table 6.6.

Sovereign Debt in Advanced Economies: Overview and Issues for Congress

Congressional Research Service 19

Table 1. Market Estimates of the Likelihood of Sovereign Defaults
2012 Q4

Highest Probability of Default Lowest Probability of Default

1. Argentina 61.4% 1. Sweden 1.64

2. Cyprus 60.5 2. Norway 1.72

3. Pakistan 42.8 3. Finland 2.64

4. Venezuela 37.4 4. Denmark 2.94

5. Ukraine 36.3 5. United States 3.30

6. Portugal 32.3 6. UK 3.66

7. Egypt 30.4 6. Germany 3.66

8. Iraq 28.1 8. Switzerland 3.77

9. Lebanon 27.5 9. Austria 3.96

10. Spain 23.5 10. Netherlands 4.04

Source: Credit Market Analysis (CMA), “CMA Global Sovereign Debt Credit Risk Report,” Q4 2012,
http://www.cmavision.com/images/uploads/docs/CMA_Global_Sovereign_Debt_Credit_Risk_Report_Q4_2012.
pdf.

Note: Likelihood of default over the next five years. Calculations based on credit default swap (CDS) values, and
due to thin trading on Greek CDS, its probability of default was not available this quarter.

Markets may perceive the United States favorably not because they believe the deficits are
currently at sustainable levels but because they believe that the government will implement
policies that reduce the deficit. However, it is important to note that market perceptions can
change quickly, and it can be difficult to predict when markets can lose confidence.

Implications for the U.S. Economy
How other advanced economies address their debt levels has implications for the U.S. economy.
Most advanced economies are addressing high debt levels through fiscal austerity. If large
austerity packages in advanced economies slow growth in those countries, demand for U.S.
exports could fall. Because advanced economies are major trading partners of the United States,
this could impact U.S. exports. Slower growth rates in advanced economies could make
investment there less attractive, and could lead to U.S. investors shifting their investment
portfolios away from advanced economies and toward emerging markets. Investors in those
countries also could shift their portfolios away from U.S. debt.

If any advanced economies do default, restructure their public debt, or use inflation to reduce the
real value of their debt, U.S. investors could face losses on their investments. Figure 6 shows
where U.S. banks have credit committed directly to borrowers overseas in general, not just to
sovereign borrowers—also referred to as how heavily U.S. banks are “exposed” overseas. Direct
U.S. bank exposure in general is more heavily concentrated among advanced economies than
emerging and developing countries. As of September 2012, 72% ($2,353 billion of $3,277
billion) of U.S. bank exposure overseas was concentrated in advanced economies.48 Among

48 Data on bank exposure from Bank for International Settlements (BIS) for September 2010. See Figure 6 source and
notes for more details.

Sovereign Debt in Advanced Economies: Overview and Issues for Congress

Congressional Research Service 20

advanced economies, U.S. banks were most exposed to the United Kingdom ($633 billion), Japan
($380 billion), Germany ($221 billion), France ($216 billion), and Canada ($128 billion) in
September 2012.

Figure 6. Direct Exposure of U.S. Banks to Advanced Economies
September 2012

Source: Bank for International Settlements (BIS), “Consolidated International Claims of BIS Reporting Banks,”
January 2013, preliminary data, Table 9D, “Consolidated Foreign Claims of Reporting Banks—Ultimate Risk
Basis,” http://www.bis.org./statistics/consstats.htm.

Note: Direct bank lending only, and exposure to the economy overall (government and private sector). Data do
not include exposure of U.S. financial institutions through the issuance of credit default swaps based on sovereign
debt, which could lower or raise U.S. bank exposure. They also do not consider secondary exposures (i.e., for
U.S. banks exposed to the United Kingdom, who, in turn, the United Kingdom is exposed to, such as Ireland).
Countries listed as advanced economies are identified as such by the IMF in the World Economic Outlook,
September 2011. See Table A-3 for exposure in dollar terms.

Direct exposure of U.S. banks to the Eurozone countries that have come under the most intense
market pressure to date—Greece, Ireland, and Portugal—is relatively small. According to the
Bank for International Settlements (BIS), U.S. bank exposure to these three countries, including
the sovereign and private sector, totaled $54 billion in September 2012, or 1.6% of direct U.S.
bank exposure overseas. Direct U.S. bank exposure to Italy and Spain (sovereign and private
sector) totaled an additional $89 billion. Some of these losses could already have been absorbed
in bank balance sheets if they are marking investments to market.

Sovereign Debt in Advanced Economies: Overview and Issues for Congress

Congressional Research Service 21

In addition to data on direct U.S. bank exposures, the BIS has recently started publishing data on
banks’ “other potential exposures” overseas, which include derivative contracts, guarantees
extended, and credit commitments. U.S. “other potential exposures” to Greece, Ireland, Italy,
Portugal, and Spain (sovereign and private sector) totaled $625 billion in September 2012.
However, the BIS data does not capture any collateral or hedges that U.S. banks may have in
place to lower their exposure to a particular borrower, and some argue that this BIS data
overstates the risks of U.S. banks. Others argue that it is helpful measure, because the amount and
quality of collateral is not publicly known, and that it captures the risk facing U.S. banks in a
systemic financial crisis where U.S. banks’ counterparties on hedging agreements could fail.

BIS data capture the exposure of banking institutions, and do not include the exposures of other
financial institutions, such as money market, insurance, and pension funds. They also do not
capture secondary exposures, such as U.S. banks that are exposed to UK banks, which are in turn
exposed to Ireland. Overall, there is a high level of uncertainty surrounding the full exposure of
the U.S. financial system to Eurozone countries under market pressure, and uncertainty
surrounding the full implications of a default or restructuring, particularly if it triggers contagion,
for the U.S. financial system.

Policy Options for Congress
Most advanced economies, including the United States, have focused on addressing high debt
levels through fiscal austerity. The task facing these countries is how to pursue fiscal
consolidation without derailing economic recovery. How to do this is contentious and has sparked
debates within Congress and more generally at the multilateral level.

Some argue that the most helpful course of action in the U.S. Congress is to fix its own fiscal
problems, although there is disagreement on the appropriate pace of and measures for achieving
fiscal consolidation. Additionally, some argue that Congress can urge the Administration to
address the issues related to historically high levels of sovereign debt issues in multilateral
discussions, particularly in the context of the G-20 and the international financial institutions
(IFIs).49 Generally, there may be multilateral interest in coordinating fiscal policies in order to
prevent large simultaneous fiscal contractions among all the advanced economies, which could
lower demand in the advanced economies and undermine a fragile economic recovery.
Coordinating fiscal policies, such as encouraging advanced countries that do not have debt
problems to pursue more expansionary fiscal policies, could soften the impact of austerity in
countries with unsustainable debt levels on the global economy. The G-20 and its process of
assessing the compatibility of policies across countries (the “mutual assessment process” [MAP])
could be one forum for these discussions.50 There may also be interest in revisiting the G-20
commitments for fiscal consolidation pledged by the advanced economies at the Toronto summit
in June 2010, and momentum may be building in the G-20 to discuss the Eurozone debt crisis.

Congress may also urge the Administration to engage with the IFIs, such as the IMF, on the
challenges posed by high debt levels in advanced economies. The IMF already analyzes fiscal
policies and debt levels in its semiannual fiscal monitor reports. However, IFI engagement on this

49 For more on the G-20 and the mutual assessment process, see CRS Report R40977, The G-20 and International
Economic Cooperation: Background and Implications for Congress, by Rebecca M. Nelson.
50 The G-20 Toronto declaration is available at http://www.g20.org/Documents/g20_declaration_en .

Sovereign Debt in Advanced Economies: Overview and Issues for Congress

Congressional Research Service 22

issue could increase. Following sovereign debt crises in the emerging markets, the IMF and the
World Bank launched an initiative to systematically collect public debt data on a quarterly basis.51
The purpose of the initiative is to increase transparency about public-sector debt by facilitating
timely dissemination of standardized public debt data. Thirty-eight emerging markets provide
data to the project. Extending participation to advanced economies could help further increase
transparency about public debt levels in these major economies as well.

51 Specifically, they launched the Public Sector Debt Statistics (PSD) database. See
http://go.worldbank.org/9PIAZORON0.

Sovereign Debt in Advanced Economies: Overview and Issues for Congress

CRS-23

Appendix. Data on General Government Debt and U.S. Bank Exposure
Overseas

Table A-1. Gross General Government Debt in Advanced Economies, Actual and Forecasts
% of GDP

Country 1980 1990 2000 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016

Australia n/a 16.4 19.5 10.0 9.7 11.8 16.9 20.5 24.2 27.1 27.2 26.4 24.7 22.9

Austria n/a 56.2 66.2 62.3 60.2 63.8 69.2 71.8 72.3 74.3 74.9 74.4 73.0 71.6

Belgium 74.3 125.6 107.8 88.0 84.0 89.3 95.7 95.6 97.8 99.0 99.4 98.6 96.5 93.9

Canada 45.6 75.2 82.1 70.3 66.5 71.3 83.3 85.1 85.4 87.5 87.8 84.6 82.3 80.3

Cyprus n/a n/a 59.6 64.7 58.8 48.9 58.5 61.5 71.6 87.3 92.6 97.6 101.6 104.1

Czech Republic n/a n/a 17.8 28.3 28.0 28.7 34.3 37.6 40.5 43.1 45.0 45.6 45.7 45.7

Denmark n/a n/a 60.4 41.0 34.1 41.9 40.6 42.9 44.1 47.1 47.6 47.8 47.9 47.3

Estonia n/a n/a 5.1 4.4 3.7 4.5 7.2 6.7 6.0 8.2 9.7 9.3 8.7 8.2

Finland 10.8 13.8 43.8 39.6 35.2 33.9 43.5 48.6 49.1 52.6 53.9 54.1 53.6 52.7

France 20.7 35.2 57.4 64.1 64.2 68.2 79.2 82.3 86.0 90.0 92.1 92.9 92.3 90.1

Germany n/a n/a 60.2 67.9 65.4 66.9 74.7 82.4 80.6 83.0 81.5 79.6 77.6 75.8

Greece 22.6 73.3 103.4 107.3 107.4 112.6 129.0 144.6 165.4 170.7 181.8 180.2 174.0 164.1

Hong Kong n/a n/a n/a 33.0 32.8 30.6 33.2 34.6 33.8 33.1 31.0 30.4 29.7 29.1

Iceland 25.5 36.2 41.0 30.1 29.1 70.3 88.2 92.8 99.2 94.2 90.5 87.4 84.0 78.6

Ireland 65.2 93.5 37.5 24.8 25.0 44.5 64.9 92.2 106.5 117.7 119.3 118.4 115.0 111.5

Israel n/a n/a 84.3 84.7 78.1 77.0 79.4 76.0 74.1 73.3 72.9 71.8 70.5 68.9

Italy n/a 94.3 108.5 106.1 103.1 105.7 116.0 118.6 120.1 126.3 127.8 127.3 125.6 123.3

Japan 50.6 67.0 140.1 186.0 183.0 191.8 210.2 215.3 229.6 236.6 245.0 246.2 247.6 248.8

Korea n/a 13.8 18.0 31.1 30.7 30.1 33.8 33.4 34.2 33.5 31.6 29.4 27.2 25.2

Luxembourg n/a n/a 6.2 6.7 6.7 13.7 14.8 19.1 18.2 21.7 24.6 27.3 30.9 33.9

Malta n/a n/a 57.9 64.2 62.2 62.2 67.8 69.1 71.6 71.8 71.1 69.7 67.8 65.5

Sovereign Debt in Advanced Economies: Overview and Issues for Congress

CRS-24

Country 1980 1990 2000 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016

Netherlands n/a n/a 53.8 47.4 45.3 58.5 60.8 62.9 65.2 68.2 70.2 71.9 72.7 73.8

New Zealand n/a 58.9 31.8 19.4 17.3 20.2 26.2 32.5 38.2 38.6 38.1 37.9 36.1 35.7

Norway 47.3 28.9 32.7 59.0 56.8 54.3 48.9 49.6 49.6 49.6 49.6 49.6 49.6 49.6

Portugal n/a 57.2 48.4 63.7 68.3 71.6 83.1 93.3 107.8 119.1 123.7 123.6 120.8 117.6

Singapore n/a 71.1 81.2 86.4 85.8 96.9 103.4 101.2 107.6 106.2 103.4 100.8 97.8 95.1

Slovak Republic n/a n/a 50.3 30.5 29.6 27.9 35.6 41.1 43.3 46.3 47.2 47.6 48.1 48.4

Slovenia n/a n/a 29.5 26.4 23.1 22.0 35.0 38.6 46.9 53.2 57.4 58.7 59.2 59.1

Spain 16.6 42.5 59.4 39.7 36.3 40.2 53.9 61.3 69.1 90.7 96.9 100.0 101.1 101.4

Sweden n/a n/a 53.3 44.8 39.7 38.4 42.0 38.8 37.9 37.1 35.9 34.1 31.0 27.7

Switzerland n/a 37.3 59.9 62.4 55.6 50.5 51.8 48.0 46.8 46.7 45.6 43.6 42.6 42.3

Taiwan n/a n/a 26.6 34.2 33.3 34.7 38.0 38.1 40.5 41.7 40.9 39.8 37.5 35.2

United Kingdom 46.1 32.4 40.9 43.0 43.7 52.2 68.0 75.0 81.8 88.7 93.3 96.0 96.6 95.8

United States 42.3 63.9 54.8 66.6 67.2 76.1 89.7 98.6 102.9 107.2 111.7 113.8 114.2 114.2

Advanced
economies

n/a n/a 72.5 76.6 74.0 81.0 94.5 100.6 104.7 109.9 112.7 113.2 112.7 111.8

G-7 advanced
economies

n/a n/a 77.3 85.5 83.5 91.8 107.0 114.7 119.9 125.1 128.8 129.7 129.4 128.7

Emerging and
developing
economies

n/a n/a 48.6 36.5 34.7 32.6 35.7 39.5 36.3 34.4 32.7 31.5 30.5 29.4

Source: IMF World Economic Outlook, October 2012.

Note: Forecasted data starts in 2009 or later, depending on the country. n/a = not available.

Sovereign Debt in Advanced Economies: Overview and Issues for Congress

CRS-25

Table A-2. Net General Government Debt in Advanced Economies, Actual and Forecasts
% of GDP

Country 1980 1990 2000 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016

Australia n/a 9.6 7.1 -6.3 -7.3 -5.3 -0.6 4.0 8.2 11.6 12.4 12.3 11.3 10.2

Austria n/a 36.7 43.2 43.1 40.9 42.0 49.2 52.5 52.1 54.1 54.7 54.1 52.8 51.4

Belgium 65.5 112.3 97.4 77.0 73.1 73.4 79.6 79.8 81.4 82.9 83.6 83.1 81.5 79.3

Canada 14.5 43.7 46.2 26.3 22.9 22.4 28.3 30.4 33.1 35.8 37.5 38.1 37.8 37.1

Cyprus

n/a n/a n/a n/a n/a n/a n/a n/a n/a n/a n/a n/a n/a n/a

Czech Republic n/a n/a n/a n/a n/a n/a n/a n/a n/a n/a n/a n/a n/a n/a

Denmark n/a n/a 22.5 1.9 -3.8 -6.1 -4.5 -1.7 0.2 4.1 6.0 7.6 9.2 10.0

Estonia n/a n/a 3.3 -4.9 -5.7 -3.5 -1.2 -1.8 -0.2 4.3 5.1 5.3 4.8 3.9

Finland -177.1 -208.3 -31.1 -69.4 -72.5 -52.3 -62.8 -65.5 -54.1 -51.1 -48.1 -45.7 -43.8 -42.4

France n/a 25.4 51.4 59.6 59.6 62.3 72.0 76.1 78.8 83.7 85.9 86.7 86.1 83.9

Germany n/a n/a 41.1 53.0 50.5 50.2 57.0 56.2 55.3 58.4 57.5 56.2 56.2 56.2

Greece 20.6 64.2 77.4 107.3 107.4 112.6 129.0 144.6 165.4 170.7 181.8 180.2 174.0 164.1

Hong Kong n/a n/a n/a n/a n/a n/a n/a n/a n/a n/a n/a n/a n/a n/a

Iceland 3.3 19.0 24.3 7.8 10.8 41.8 55.8 62.8 65.9 65.7 64.4 62.4 59.3 55.8

Ireland 65.2 93.5 36.4 12.1 11.1 24.6 42.0 74.7 94.9 103.0 107.6 108.7 107.2 104.0

Israel n/a n/a 70.7 74.0 67.3 63.6 68.6 68.3 67.5 67.0 67.0 66.3 65.4 64.0

Italy n/a 89.2 93.1 89.3 86.9 88.8 97.2 99.1 99.6 103.1 103.9 103.7 102.4 100.8

Japan 16.8 13.2 59.6 81.0 80.5 95.3 106.2 112.8 126.4 135.4 144.7 148.7 152.4 155.6

Korea n/a n/a n/a 29.4 28.7 28.8 32.3 32.1 32.9 32.0 30.3 28.1 26.1 24.2

Luxembourg n/a n/a n/a n/a n/a n/a n/a n/a n/a n/a n/a n/a n/a n/a

Malta n/a n/a n/a n/a n/a n/a n/a n/a n/a n/a n/a n/a n/a n/a

Netherlands n/a n/a 24.9 24.5 21.6 20.6 23.2 27.6 31.7 35.1 37.6 40.2 42.1 44.1

New Zealand n/a 46.5 18.2 0.2 -5.7 -4.8 -0.8 3.5 8.3 12.1 13.9 14.5 14.3 14.2

Norway 0.4 -31.8 -67.2 -133.7 -138.9 -123.5 -156.7 -165.3 -168.2 -169.3 -173.0 -178.3 -182.1 -184.5

Portugal n/a n/a 41.9 58.6 63.7 67.4 79.0 88.9 97.3 113.2 119.5 119.4 116.7 113.7

Sovereign Debt in Advanced Economies: Overview and Issues for Congress

CRS-26

Country 1980 1990 2000 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016

Singapore n/a n/a n/a n/a n/a n/a n/a n/a n/a n/a n/a n/a n/a n/a

Slovak Republic n/a n/a n/a n/a n/a n/a n/a n/a n/a n/a n/a n/a n/a n/a

Slovenia n/a n/a n/a n/a n/a n/a n/a n/a n/a n/a n/a n/a n/a n/a

Spain n/a 30.3 50.4 30.7 26.7 30.8 42.5 49.8 57.5 78.6 84.4 87.3 88.3 88.5

Sweden n/a n/a 2.2 -13.8 -17.3 -12.4 -19.4 -20.6 -18.2 -17.5 -16.5 -16.0 -16.9 -18.2

Switzerland n/a 14.2 37.0 39.7 32.0 28.0 28.7 25.7 25.9 25.8 25.2 24.1 23.6 23.4

Taiwan n/a n/a n/a n/a n/a n/a n/a n/a n/a n/a n/a n/a n/a n/a

United Kingdom 40.5 26.5 33.6 37.8 38.0 45.8 60.6 71.0 76.6 83.7 88.2 90.9 91.5 90.7

United States 25.8 45.9 35.6 48.6 48.2 53.8 65.8 73.2 80.3 83.8 87.7 89.3 89.5 89.6

Advanced
economies

n/a n/a 43.6 48.3 46.3 51.6 62.0 66.6 71.6 76.5 79.7 80.9 81.2 81.2

G-7 advanced
economies

n/a n/a 45.7 55.6 54.6 60.8 72.2 77.9 84.1 89.0 92.8 94.3 94.9 95.0

Emerging and
developing
economies
n/a n/a n/a n/a n/a n/a n/a n/a n/a n/a n/a n/a n/a n/a
Source: IMF World Economic Outlook, October 2012.

Note: Forecasted data starts in 2009 or later, depending on the country. n/a = not available.

Sovereign Debt in Advanced Economies: Overview and Issues for Congress

Congressional Research Service 27

Table A-3. Direct Exposure of U.S. Banks to Advanced Economies
September 2012, million US$

Country Amount Country Amount Country Amount

Australia 113,654 Israel 4,027 Switzerland 89,688
Austria 12,410 Italy 39,726 Taiwan 40,059
Belgium 20,069 Japan 379,550 United Kingdom 632,994
Canada 127,674 Luxembourg 30,083
Cyprus 1,502 Malta 181
Czech Republic 19,393 Netherlands 108,443
Denmark 40 New Zealand 10,750
Estonia 10,140 Norway 5,102
Finland 215,567 Portugal 594
France 221,217 Singapore 1,000
Germany 3,170 Slovakia 93,119
Greece 52,091 Slovenia 49,268
Hong Kong 699 South Korea 25,887
Iceland 45,368 Spain 4,027
Ireland 113,654 Sweden 39,726

Advanced
economies

2,353,465

All countries 3,277,159

Source: Bank for International Settlements (BIS), “Consolidated International Claims of BIS Reporting Banks,”
January 2013, preliminary data, Table 9D, “Consolidated Foreign Claims of Reporting Banks—Ultimate Risk
Basis,” http://www.bis.org./statistics/consstats.htm.

Note: Direct bank lending only, and exposure to the economy overall (government and private sector). Data do
not include exposure of U.S. financial institutions through the issuance of credit default swaps based on sovereign
debt, which could lower or raise U.S. bank exposure. They also do not consider secondary exposures (i.e., for
U.S. banks exposed to the United Kingdom, who, in turn, the United Kingdom is exposed to, such as Ireland).
Countries listed as advanced economies are identified as such by the IMF in the World Economic Outlook
(excluding San Marino) .

Author Contact Information

Rebecca M. Nelson
Analyst in International Trade and Finance
rnelson@crs.loc.gov, 7-6819

Sovereign Debt in Advanced Economies: Overview and Issues for Congress

Congressional Research Service 28

Acknowledgments

Amber Wilhelm, Graphics Specialist, assisted in preparation of the figures.

Briefing Note Template

( 2 to 3 pages max)

Issue

· Define the problem

Background/Situation Analysis

· Overview of the current situation

Economic Analysis

· What does the economic analysis show

· What does economic theory tell us

· Benefit-Cost Analysis

· Cost Effectiveness

Risk Analysis

· What are the key risks

· What are the consequences

Option Analysis

· What are the available options

· Alternative policy instruments (rational)

Assessment of the Proposal

· Strengths and Weaknesses of the paper from your perspective

· Why

Personal Recommendations

· What is your recommended option /solution to the problem

· Why ( evidence based – economic theory, lessons learned , best practices)

Note: marks will be deducted if the briefing note is more than 3 pages.

57

3

American Economic Review: Papers & Proceedings 100 (May 2010): 573–578
http://www.aeaweb.org/articles.php?doi=10.1257/aer.100.2.573

In this paper, we exploit a new multi-country

historical dataset on public (government) debt to
search for a systemic relationship between high
public debt levels, growth and in!ation.1 Our
main result is that whereas the link between
growth and debt seems relatively weak at “nor-
mal” debt levels, median growth rates for coun-
tries with public debt over roughly 90 percent
of GDP are about one percent lower than other-
wise; average (mean) growth rates are several
percent lower. Surprisingly, the relationship
between public debt and growth is remarkably
similar across emerging markets and advanced
economies. This is not the case for in!ation. We
“nd no systematic relationship between high
debt levels and in!ation for advanced econo-
mies as a group (albeit with individual country
exceptions including the United States). By con-
trast, in emerging market countries, high public
debt levels coincide with higher in!ation.

Our topic would seem to be a timely one.
Public debt has been soaring in the wake of the
recent global “nancial maelstrom, especially in
the epicenter countries. This should not be sur-
prising, given the experience of earlier severe
“nancial crises.2 Outsized de”cits and epic bank
bailouts may be useful in “ghting a downturn,
but what is the long-run macroeconomic impact,

1 In this paper “public debt” refers to gross central
government debt.  “Domestic public debt” is government
debt issued under domestic legal jurisdiction. Public debt
does not include debts carrying a government guarantee.
Total gross external debt includes the external debts of all
branches of government as well as private debt that is issued
by domestic private entities under a foreign jurisdiction.

2 Reinhart and Rogoff (2009a, b) demonstrate that the
aftermath of a deep “nancial crisis typically involves a
protracted period of macroeconomic adjustment, particu-
larly in employment and housing prices. On average, public
debt rose by more than 80 percent within three years after
a crisis.

Growth in a Time of Debt

By Carmen M. Reinhart and Kenneth S. Rogoff*

especially against the backdrop of graying pop-
ulations and rising social insurance costs? Are
sharply elevated public debts ultimately a man-
ageable policy challenge?

Our approach here is decidedly empirical,
taking advantage of a broad new historical
dataset on public debt (in particular, central
government debt) “rst presented in Carmen M.
Reinhart and Kenneth S. Rogoff (2008, 2009b).
Prior to this dataset, it was exceedingly dif”cult
to get more than two or three decades of pub-
lic debt data even for many rich countries, and
virtually impossible for most emerging markets.
Our results incorporate data on 44 countries
spanning about 200 years. Taken together, the
data incorporate over 3,700 annual observations
covering a wide range of political systems, insti-
tutions, exchange rate and monetary arrange-
ments, and historic circumstances.

We also employ more recent data on external
debt, including debt owed both by governments
and by private entities. For emerging markets,
we “nd that there exists a signi”cantly more
severe threshold for total gross external debt
(public and private)—which is almost exclu-
sively denominated in a foreign currency—than
for total public debt (the domestically issued
component of which is largely denominated
in home currency). When gross external debt
reaches 60 percent of GDP, annual growth
declines by about two percent; for levels of
external debt in excess of 90 percent of GDP,
growth rates are roughly cut in half. We are not
in a position to calculate separate total exter-
nal debt thresholds (as opposed to public debt
thresholds) for advanced countries. The avail-
able time-series is too recent, beginning only in
2000. We do note, however, that external debt
levels in advanced countries now average nearly
200 percent of GDP, with external debt levels
being particularly high across Europe.

The focus of this paper is on the longer term
macroeconomic implications of much higher
public and external debt. The “nal section, how-
ever, summarizes the historical experience of
the United States in dealing with private sector

* Reinhart: Department of Economics, 4115 Tydings
Hall, University of Maryland, College Park, MD 20742
(e-mail: creinhar@umd.edu); Rogoff: Economics Depart-
ment, 216 Littauer Center, Harvard University, Cambridge
MA 02138–3001 (e-mail: krogoff@harvard.edu). The
authors would like to thank Olivier Jeanne and Vincent R.
Reinhart for helpful comments.

MAY 2010574 AEA PAPERS AND PROCEEDINGS

deleveraging of debts, normal after a “nancial
crisis. Not surprisingly, such episodes are asso-
ciated with very slow growth and de!ation.

I. The 2007–2009 Global Buildup in Public Debt

Figure 1 illustrates the increase in (in!ation-
adjusted) public debt that has occurred since
2007. For the “ve countries with systemic “nan-
cial crises (Iceland, Ireland, Spain, the United
Kingdom, and the United States), average debt
levels are up by about 75 percent, well on track to
reach or surpass the three year 86 percent bench-
mark that Reinhart and Rogoff (2009a,b), “nd
for earlier deep postwar “nancial crises. Even in
countries that did not experience a major “nan-
cial crisis, debt rose by an average of about 20
percent in real terms between 2007 and 2009.3

3 Our focus on gross central government debt owes to
the fact that time series of broader measures of government

This general rise in public indebtedness stands in
stark contrast to the 2003–2006 period of pub-
lic deleveraging in many countries and owes to
direct bailout costs in some countries, the adop-
tion of stimulus packages to deal with the global
recession in many countries, and marked declines
in government revenues that have hit advanced
and emerging market economies alike.

II. Debt, Growth, and In!atio

n

The nonlinear effect of debt on growth is
reminiscent of “debt intolerance” (Reinhart,
Rogoff, and Miguel A. Savastano 2003) and
presumably is related to a nonlinear response
of market interest rates as countries reach debt
tolerance limits. Sharply rising interest rates,
in turn, force painful “scal adjustment in the
form of tax hikes and spending cuts, or, in
some cases, outright default. As for in!ation,
an obvious connection stems from the fact that
unanticipated high in!ation can reduce the
real cost of servicing the debt. Of course, the
ef”cacy of the in!ation channel is quite sen-
sitive to the maturity structure of the debt. In
principle, the manner in which debt builds up
can be important. For example, war debts are
arguably less problematic for future growth
and in!ation than large debts that are accu-
mulated in peacetime. Postwar growth tends
to be high as wartime allocation of manpower
and resources funnels to the civilian economy.
Moreover, high wartime government spending,
typically the cause of the debt buildup, comes
to a natural close as peace returns. In contrast,
a peacetime debt explosion often re!ects unsta-
ble political economy dynamics that can persist
for very long periods.

Here we will not attempt to determine the gen-
esis of debt buildups but instead simply look at
their connection to average and median growth
and in!ation outcomes. This may lead us, if any-
thing, to understate the adverse growth implica-
tions of debt burdens arising out of the current
crisis, which was clearly a peacetime event.

debt are not available for many countries. Of course, the
true run-up in debt is signi”cantly larger than stated here,
at least on a present value actuarial basis, due to the exten-
sive government guarantees that have been conferred on the
“nancial sector in the crisis countries and elsewhere, where
for example deposit guarantees were raised in 2008.

6

9

4

4

7

2

42

84

62

22

9

21

29

2

5

47

119

47
62
4

44

182

4

6

32

41

49

Debt/GDP
2009

100 150 200 250

Iceland

Ireland

UK

Spain

US

Crisis country average

Norway

Australia

China

Thailand

Mexico

Malaysia

Greece

Canada

Austria

Chile

Germany

Japan

Brazil

Korea

India

Average for others

2007 = 100

175.1
(increase of 75%)

120 (increase of 20%)

Figure 1. Cumulative Increase in Real Public Debt
Since 2007, Selected Countries

Note: Unless otherwise noted these “gures are for central
government debt de!ated by consumer prices.

Sources: Prices and nominal GDP from International
Monetary Fund, World Economic Outlook. For a complete
listing of sources for government debt, see Reinhart and
Rogoff (2009b).

VOL. 100 NO. 2 575GROWTH IN A TIME OF DEBT

A. Evidence from Advanced Countries

Figure 2 presents a summary of in!ation and
GDP growth across varying levels of debt for 20
advanced countries over the period 1946–2009.
This group includes Australia, Austria, Belgium,
Canada, Denmark, Finland, France, Germany,
Greece, Ireland, Italy, Japan, Netherlands, New
Zealand, Norway, Portugal, Spain, Sweden, the
United Kingdom, and the United States. The
annual observations are grouped into four cat-
egories, according to the ratio of debt to GDP
during that particular year as follows: years when
debt to GDP levels were below 30 percent (low
debt); years where debt/GDP was 30 to 60 per-
cent (medium debt); 60 to 90 percent (high); and

above 90 percent (very high). The bars in Figure
2 show average and median GDP growth for
each of the four debt categories. Note that of the
1,186 annual observations, there are a signi”cant
number in each category, including 96 above 90
percent. (Recent observations in that top bracket
come from Belgium, Greece, Italy, and Japan.)
From the “gure, it is evident that there is no
obvious link between debt and growth until pub-
lic debt reaches a threshold of 90 percent. The
observations with debt to GDP over 90 percent
have median growth roughly 1 percent lower than
the lower debt burden groups and mean levels of
growth almost 4 percent lower. (Using lagged
debt does not dramatically change the picture.)
The line in Figure 2 plots the median in!ation for
the different debt groupings—which makes plain
that there is no apparent pattern of simultaneous
rising in!ation and debt.

Table 1 provides detail on the growth experi-
ence for individual countries, but over a much
longer period, typically one to two centuries.
Interestingly, introducing the longer time-series
yields remarkably similar conclusions. Over the
past two centuries, debt in excess of 90 percent
has typically been associated with mean growth
of 1.7 percent versus 3.7 percent when debt is
low (under 30 percent of GDP), and compared
with growth rates of over 3 percent for the two
middle categories (debt between 30 and 90 per-
cent of GDP). Of course, there is considerable
variation across the countries, with some coun-
tries such as Australia and New Zealand experi-
encing no growth deterioration at very high debt
levels. It is noteworthy, however, that those high-
growth high-debt observations are clustered in
the years following World War II.

B. Evidence from Emerging Market Countries

We next perform the same exercise for 24
emerging market economies for the periods
1946–2009 and 1900–2009, using comparable
central government debt data to those we used
for the advanced economies.4 Perhaps surpris-
ingly, the results illustrated in Figure 2 and
Table 1 for advanced economies are repeated
for emerging market economies. The emerging

4 While we have pre-1900 in!ation, real GDP, and public
debt data for many emerging markets, nominal GDP data is
harder to “nd.

!

1.0

0.0

1.0

2.0

3.0

4.0

5.0

Average Median Average Median Average Median Average Median

G
D

P
g

ro
w

th

2

2.5

3

3.5

4

4.5

5

5.5

6

In
fla

ti
o

n

Debt/GDP
below 30%

Debt/GDP
30 to 60%

Debt/GDP
60 to 90%

Debt/GDP
above 90%

Inflation
(line, right axis)

GDP growth (bars, left axis)

Figure 2. Government Debt, Growth, and Inflation:
Selected Advanced Economies, 1946–2009

Notes: Central government debt includes domestic and
external public debts. The 20 advanced economies included
are Australia, Austria, Belgium, Canada, Denmark,
Finland, France, Germany, Greece, Ireland, Italy, Japan,
Netherlands, New Zealand, Norway, Portugal, Spain,
Sweden, the United Kingdom, and the United States. The
number of observations for the four debt groups are: 443
for debt/GDP below 30 percent; 442 for debt/GDP 30 to 60
percent; 199 observations for debt/GDP 60 to 90 percent;
and 96 for debt/GDP above 90 percent. There are 1,180
observations.
Sources: International Monetary Fund, World Economic
Outlook, OECD, World Bank, Global Development
Finance, and Reinhart and Rogoff (2009b) and sources
cited therein.

MAY 2010576 AEA PAPERS AND PROCEEDINGS

market equivalents of Figure 2 and Table 1 are
not reproduced here (to economize on space),
but the interested reader is referred to the
NBER working paper version of this paper.
For 1900–2009, for example, median and aver-
age GDP growth hovers around 4–4.5 percent
for levels of debt below 90 percent of GDP, but
median growth falls markedly to 2.9 percent
for high debt (above 90 percent); the decline is
even greater for the average growth rate, which
falls to 1 percent. With much faster population
growth than the advanced economies’, the impli-
cations for per capita GDP growth are in line (or
worse) with those shown for advanced econo-
mies. The similarities with advanced economies
end there, as higher debt levels are associated

with signi”cantly higher levels of in!ation in
emerging markets. Median in!ation more than
doubles (from less than seven percent to 16 per-
cent) as debt rises from the low (0 to 30 percent)
range to above 90 percent. Fiscal dominance is a
plausible interpretation of this pattern.

Because emerging markets often depend so
much on external borrowing, it is interesting to
look separately at thresholds for external debt
(public and private). In Figure 3, we highlight
the connection between gross external debt as
reported by the World Bank and growth and
in!ation. As one can see, the growth thresholds
for external debt are considerably lower than the
thresholds for total public debt. Growth dete-
riorates markedly at external debt levels over

Table 1—Real GDP Growth as the Level of Government Debt Varies:
Selected Advanced Economies, 1790–2009

(annual percent change)
Central (federal) government debt/GDP

Country Period
Below 30
percent

30 to 60
percent

60 to 90
percent

90 percent
and above

Australia 1902–2009 3.1 4.1 2.3 4.6
Austria 1880–2009 4.3 3.0 2.3 n.a.
Belgium 1835–2009 3.0 2.6 2.1 3.3
Canada 1925–2009 2.0 4.5 3.0 2.2
Denmark 1880–2009 3.1 1.7 2.4 n.a.
Finland 1913–2009 3.2 3.0 4.3 1.9
France 1880–2009 4.9 2.7 2.8 2.3
Germany 1880–2009 3.6 0.9 n.a. n.a.
Greece 1884–2009 4.0 0.3 4.8 2.5
Ireland 1949–2009 4.4 4.5 4.0 2.4
Italy 1880–2009 5.4 4.9 1.9 0.7
Japan 1885–2009 4.9 3.7 3.9 0.7
Netherlands 1880–2009 4.0 2.8 2.4 2.0
New Zealand 1932–2009 2.5 2.9 3.9 3.6
Norway 1880–2009 2.9 4.4 n.a. n.a.
Portugal 1851–2009 4.8 2.5 1.4 n.a.
Spain 1850–2009 1.6 3.3 1.3 2.2
Sweden 1880–2009 2.9 2.9 2.7 n.a.
United Kingdom 1830–2009 2.5 2.2 2.1 1.8
United States 1790–2009 4.0 3.4 3.3 !1.8
Average 3.7 3.0 3.4 1.7
Median 3.9 3.1 2.8 1.9

Observations = 2,317 866 654 445 352
Notes: An n.a. denotes no observations were recorded for that particular debt range. There
are missing observations, most notably during World War I and II years; further details are
provided in the data appendices to Reinhart and Rogoff (2009b) and are available from the
authors. Minimum and maximum values for each debt range are shown in bolded italics.
Sources: There are many sources; among the more prominent are: International Monetary
Fund, World Economic Outlook, OECD, World Bank, Global Development Finance. Extensive
other sources are cited in Reinhart and Rogoff (2009).

VOL. 100 NO. 2 577GROWTH IN A TIME OF DEBT

60 percent, and further still when external debt
levels exceed 90 percent, which record outright
declines. In light of this, it is more understand-
able that over one half of all defaults on external
debt in emerging markets since 1970 occurred at
levels of debt that would have met the Maastricht
criteria of 60 percent. In!ation becomes signi”-
cantly higher only for the group of observations
with external debt over 90 percent.

III. Private Sector Debt: An Illustration

Our main focus has been on central govern-
ment debt and, to a lesser degree, external public
and private debt, since reliable data on private
domestic debts are much scarcer across countries
and time. We have argued here and elsewhere
that a key legacy of a deep “nancial crisis is
rapidly expanding public debt. Furthermore, we

have shown that public levels of debt/GDP that
push the 90 percent threshold are associated with
lower median and average growth.5 These obser-
vations, however, present only a partial picture of
the post-“nancial crisis landscape. Private debt,
in contrast, tends to shrink sharply in the after-
math of a “nancial crisis. Just as a rapid expan-
sion in private credit fuels the boom phase of the
cycle, so does serious deleveraging exacerbate the
post-crisis downturn. This pattern is illustrated in
Figure 4, which shows the ratio of private debt to
GDP for the United States for 1916–2009. Periods
of sharp deleveraging have followed periods of
lower growth and coincide with higher unem-
ployment. In varying degrees, the private sector
(households and “rms) in many other countries
(notably both advanced and emerging Europe)
are also unwinding the debt built up during the
boom year. Thus, private deleveraging may be
another legacy of the “nancial crisis that may
dampen growth in the medium term.

IV. Concluding Remarks

The sharp run-up in public sector debt will
likely prove one of the most enduring lega-
cies of the 2007–2009 “nancial crises in the
United States and elsewhere. We examine the
experience of 44 countries spanning up to two
centuries of data on central government debt,
in!ation and growth. Our main “nding is that
across both advanced countries and emerging
markets, high debt/GDP levels (90 percent and
above) are associated with notably lower growth
outcomes. Much lower levels of external debt/
GDP (60 percent) are associated with adverse
outcomes for emerging market growth. Seldom
do countries “grow” their way out of debts. The
nonlinear response of growth to debt as debt
grows towards historical boundaries is remi-
niscent of the “debt intolerance” phenomenon
developed in Reinhart, Rogoff, and Savastano
(2003). As countries hit debt tolerance ceilings,
market interest rates can begin to rise quite sud-
denly, forcing painful adjustment.

Of course, there are other vulnerabilities
associated with debt buildups, particularly if
governments try to mitigate servicing costs by

5 It is important to note that post-crises increases in pub-
lic debt do not necessarily push economies into the vulner-
able 90+ debt/GDP range.

!

1.5

!

0.5

0.5
1.5
2.5
3.5
4.5
5.5
Average Median Average Median Average Median Average Median
G
D
P
g
ro
w
th

10

11

12

13

14

15

16

17

Debt/GDP
below 30%
Debt/GDP
30 to 60%
Debt/GDP
60 to 90%
Debt/GDP
above 90%
Inflation
(line, right axis)
GDP growth (bars, left axis)
In
fla

tio
n

Figure 3. External Debt, Growth, and Inflation:
Selected Emerging Markets, 1970-2009

Notes: The 20 emerging market countries included are
Argentina, Bolivia, Brazil, Chile, China, Colombia, Egypt,
India, Indonesia, Korea, Malaysia, Mexico, Nigeria, Peru,
Philippines, South Africa, Thailand, Turkey, Uruguay, and
Venezuela. The number of observations for the four debt
groups are: 252 for debt/GDP below 30 percent; 309 for
debt/GDP 30 to 60 percent; 120 observations for debt/GDP
60 to 90 percent; and 74 for debt/GDP above 90 percent.
There is a total of 755 annual observations.
Sources: International Monetary Fund, World Economic
Outlook, World Bank, Global Development Finance, and
Reinhart and Rogoff (2009b) and sources cited therein.

MAY 2010578 AEA PAPERS AND PROCEEDINGS

shortening the maturing structure of debt. As
Reinhart and Rogoff (2009b) emphasize and
numerous models suggest, countries that choose
to rely excessively on short-term borrowing to
fund growing debt levels are particularly vul-
nerable to crises in con”dence that can provoke
very sudden and “unexpected” “nancial crises.
At the very minimum, this would suggest that
traditional debt management issues should be at
the forefront of public policy concerns.

REFERENCES

Reinhart, Carmen M., and Kenneth S. Rogoff.
2008. “The Forgotten History of Domestic

Debt.” National Bureau of Economic Research
Working Paper 13946.

Reinhart, Carmen M., and Kenneth S. Rogoff.
2009a. “The Aftermath of Financial Cri-
ses.” American Economic Review, 99(2):
466–72.

Reinhart, Carmen M., and Kenneth S. Rogoff.
2009b. This Time Is Different: Eight Centuries
of Financial Folly. Princeton, NJ: Princeton
University Press.

Reinhart, Carmen M., Kenneth S. Rogoff, and
Miguel A. Savastano. 2003. “Debt Intoler-
ance.” Brookings Papers on Economic Activ-
ity (1), ed. William C. Brainard and George L.
Perry, 1–62.

1916–1939 1946–2009
Years with debt/GDP declines 9.8 7.2

All other years 6.7 5.5

unemployment rate
Median

20

70

120

170

220

270

320

1916 1921 1926 1931 1936 1941 1946 1951 1956 1961 1966 1971 1976 1981 1986 1991 1996 2001 2006

Historical statistics
of the United States

Flow of funds

P
er

ce
n
t

Figure 4. United States: Private Debt Outstanding, 1916–2009
(end-of-period stock of debt as a percent of GDP)

Note: Data for 2009 is end-of-June.

Sources: Historical Statistics of the United States, Flow of Funds, Board of Governors of the
Federal Reserve, International Monetary Fund, World Economic Outlook, OECD, World Bank,
Global Development Finance, and Reinhart and Rogoff (2009b) and sources cited therein.

Growth in a Time of Debt
I. The 2007–2009 Global Buildup in Public Debt
II. Debt, Growth, and Inflation
A. Evidence from Advanced Countries
B. Evidence from Emerging Market Countries
III. Private Sector Debt: An Illustration
IV. Concluding Remarks
REFERENCES

1. Cit p_2: