summary and PowerPoint and case study

Accounting Research Center, Booth School of Business, University of Chicago
Earnings Management During Import Relief Investigations
Author(s): Jennifer J. Jones
Source: Journal of Accounting Research, Vol. 29, No. 2 (Autumn, 1991), pp. 193-228
Published by: Wiley on behalf of Accounting Research Center, Booth School of Business, University of
Chicago
Stable URL: http://www.jstor.org/stable/2491047 .
Accessed: 13/04/2013 16:12
Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at .
http://www.jstor.org/page/info/about/policies/terms.jsp
.
JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of
content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms
of scholarship. For more information about JSTOR, please contact support@jstor.org.
.
Wiley and Accounting Research Center, Booth School of Business, University of Chicago are collaborating
with JSTOR to digitize, preserve and extend access to Journal of Accounting Research.
http://www.jstor.org
This content downloaded from 177.24.16.14 on Sat, 13 Apr 2013 16:12:17 PM
All use subject to JSTOR Terms and Conditions
Journalof AccountingResearch
Vol. 29 No. 2 Autumn1991
Printedin U.S.A.
Earnings Management During
Import Relief Investigations
JENNIFER
J.
JONES*
1. Introduction
This study tests whether firms that would benefit from import relief
(e.g., tariff increases and quota reductions) attempt to decreaseearnings
throughearnings managementduringimport relief investigations by the
United States InternationalTrade Commission (ITC). The import relief
determination made by the ITC is based on several factors that are
specified in the federal trade acts, including the profitability of the
industry. Explicit use of accounting numbers in import relief regulation
providesincentives for managersto manageearnings in orderto increase
the likelihood of obtaining import relief and/or increase the amount of
relief granted.
While studies of earnings managementtypically examine situations in
which all contracting parties have incentives to “perfectly”monitor
(adjust)accounting numbers for such manipulation,import relief investigations provide a specific motive for earnings managementthat is not
* University of Chicago. I am especially indebted to my dissertation committee, John
Bound, Michael Bradley (cochairman), Michael Maher (cochairman), and Thomas Stober
for their assistance and to others who have provided helpful comments, notably an
anonymous referee, Victor Bernard, Dean Crawford, Linda DeAngelo, Bruce Ikawa, William
Kinney, Richard Leftwich, Robert Lipe, Peter Wilson, David Wright, and workshop
participants at Carnegie-Mellon University, Northwestern University, Ohio State University, University of Arizona, University of Chicago, University of Iowa, University of North
Carolina, University of Rochester, Washington University, and Wharton School. I am also
indebted to Richard Laulor from the United States International Trade Commission (ITC)
for his help in obtaining information about the import relief investigation process at the
ITC. This paper has been funded by the Institute of Professional Accounting at the
University of Chicago and dissertation fellowships from the Arthur Andersen & Co.
Foundation and the University of Michigan Dykstra Fund.
193
Copyright?, Journalof AccountingResearch1991
This content downloaded from 177.24.16.14 on Sat, 13 Apr 2013 16:12:17 PM
All use subject to JSTOR Terms and Conditions
194
JOURNAL OF ACCOUNTING RESEARCH, AUTUMN 1991
providedin other earnings managementstudies. Importrelief is a wealth
transfer from a groupof diffuse losers (consumers)to a groupof concentrated winners (all other contracting parties of domestic producers receiving import relief).’ I argue that consumers do not monitor earnings
management as effectively as losers examined in other studies because
the loss to each consumer is smaller, and their interests more diverse,
than for the contracting parties examined in these studies.2Regulators
have less incentive to adjust for managers’earnings manipulationssince
their ultimate payoff for such adjustment is less direct than in other
situations previously studied (e.g., union contract negotiations).3 Furthermore, interviews of ITC regulators indicate that the ITC does not
adjust financial data for accounting procedures used or for accrual
decisions made by firms.
This study documents the use of accounting numbers in a federal
governmentprogramas a basis for wealth transfers (i.e., import relief).
An estimate of the discretionarycomponent of total acruals is used as
the measure of earnings management rather than the discretionary
component of a single accrual (as used in McNichols and Wilson
[1988]). The discretionarycomponent of total accrualsis more appropriate in this context becausethe ITC is interested in earningsbefore taxes,
which includes the effects of all accrualaccounts, and, as such, managers
are likely to use several accruals to reduce reported earnings. Firmspecific expectations models are developed to estimate normal (nondiscretionary) accruals. The expectations models control for the effects of
economic conditions on the level of accruals.I conduct a cross-sectional
analysis to test whether estimated discretionaryaccruals (i.e., residuals
from the estimated expectations models) tend to be income-decreasing
duringthe importrelief investigationperiod.The methodologydeveloped
in this study extends the methodologyused in other earnings management studies; specifically, time-series models are developed to estimate
total nondiscretionaryaccruals and cross-sectional tests of the earnings
managementhypothesis are applied. The results of these tests are consistent with the hypothesis that managers decrease earnings through
earnings managementduring import relief investigations. This evidence
‘Examples of other contractingparties are management,shareholders,debtholders,and
employees. It should be noted that the long-run effects of import relief are not well
understood.It may be the case that the short-termcosts to consumers(i.e., wealthtransfers
dueto importrelief)areoffset in the long-runby benefitsfroma strongerdomesticindustry.
2
example,employeesduringunion contract negotiations (Libertyand Zimmerman
[1986]),shareholdersin setting managementcompensation(Healy [1985]),shareholdersin
managementbuyouts (DeAngelo [1986]), and shareholdersin proxy contests (DeAngelo
[1988]).
3As I discuss later, the ITC has been organizedin such a way as to minimizethe effect
of voters on the ITC’s actions; therefore,the ITC may have less incentive to adjust than
regulatorsin situations in which voters have a more direct influence. In either case, the
payoff to regulatorsfor adjustingfor managers’opportunisticaccounting choices is less
directthan it is for other contractingparties such as debtholdersand stockholders.
This content downloaded from 177.24.16.14 on Sat, 13 Apr 2013 16:12:17 PM
All use subject to JSTOR Terms and Conditions
protection.
The next section provides institutional backgroundfor import relief
determinations.Section 3 develops the hypothesis to be tested. Section
4 contains the sample selection procedures and descriptive statistics.
Section 5 reports the results of the empirical tests. The last section
providesconclusions.
2. Role of AccountingNumbers in Foreign Trade
Regulation
Foreign trade regulation provides avenues for granting import relief
throughtariffs, quotas, marketingagreements,and/or federaladjustment
assistance. In most cases, an increase in import protection results in a
wealth transfer from domestic consumers, domestic importers,and foreign suppliersto domestic producersof the protected good. Agents in the
domesticproducers’nexus of contracts, such as employees, stockholders,
debtholders,and suppliers,cannot be hurt directly by the importprotection and instead may benefit.4Managersof firms that wouldbenefit from
increased import protection have incentives to take actions to increase
the likelihood of obtaining such protection and/or increase the amount
of protection granted. The ways in which managers can increase the
expected value of the import relief depend on the factors consideredby
the regulatorswhen making import relief decisions. In the remainderof
this section, I review these factors.
2.1 STATUTORY PROVISIONS
OF THE FOREIGN TRADE ACTS
The majorstatutory provisions of the foreign trade acts that relate to
import relief are summarizedin Appendix A. The first three statutes,
which pertain to general escape clause, countervailing duty, and antidumpinginvestigations, are the primaryfocus of this study. Title VII of
the Tariff Act of 1930 was designed to protect domestic industries from
imports that are sold at less than fair value (antidumping) or are
benefiting from foreign subsidies (countervailing duty). The general
escape clause investigations are based on section 201 of the Trade Act of
1974, which was designed to aid domestic industries that are seriously
injuredby increasedimports. Section 201 is based on article XIX of the
General Agreement on Tariffs and Trade (GATT), which permits a
country to “escape”(hence the term “escape clause”) temporarilyfrom
its obligations under the GATT when increased imports of a specific
product are causing or threatening to cause serious injury to domestic
producersof a like or directly competitive product. ITC investigations
conductedunder section 201 provide a basis for the president to invoke
article XIX of the GATT.
4Except to the extent that the contractingparties are also consumers.
This content downloaded from 177.24.16.14 on Sat, 13 Apr 2013 16:12:17 PM
All use subject to JSTOR Terms and Conditions
utes requirethe ITC to make a favorable injury decision before import
relief can be granted.’In the case of countervailingduty and antidumping
cases, once the ITC has determinedthat an industry is being injuredby
imports,the Departmentof Commercedeterminesthe increase in tariffs
necessary to offset the dumping margin or foreign subsidy. If the ITC
rules favorablyin generalescape clause investigations,a recommendation
is made by the ITC to the president to grant the industry some specified
type of import relief. The president has 60 days to make his import relief
decision. If the president does not grant any import relief or grants relief
that differs from that recommendedby the ITC, Congress can override
the president’sdecision and accept the ITC’s recommendationby obtaining an affirmativevote in each House within 90 days after the president’s
decision. In each of these three types of investigations, the ITC must
find that the industryhas been injuredbeforeimportrelief can be granted.
The federaltrade acts specifythe factorsto be consideredwhen making
importrelief decisions. In the case of generalescape clause investigations,
the Trade Act of 1974 states that in determininginjury:
… the Commissionshall take into account all economic factors which it considers
relevant,including(but not limited to)(A) with respect to serious injury,the significant idling of productivefacilities in the
industry, the inability of a significant number of firms to operate at a
reasonable level of profit, and significant unemploymentor underemployment
within the industry;
(B) with respectto threat of seriousinjury,a decline in sales, a higher and growing
inventory, and a downward trend in production,profits, wages,or employment
(or increasingunderemployment)in the domestic industryconcerned… (19 USC
2251(b)(2)).[Emphasisadded.]
The factors to be considered in antidumping and countervailing duty
investigations are as follows:
(1) actual and potential decline in output, sales, market share, profits, productivity,
return on investments, and utilization of capacity,
(2) factors affectingdomesticprices, and
(3) actual and potential negative effects on cash flow,6 inventories, employment,
wages, growth,ability to raise capital, and investment (Trade AgreementsAct of
1979, section 771 (19 USC 1677(7)). [Emphasisadded.]
Since injury determinations specifically call for the use of accounting
numbers (i.e., profits, sales, and inventories) in foreign trade regulation,
the remainderof this paper addresses the three types of investigations
that requiresuch determinations.
‘The ITC is responsiblefor makingall injurydecisionsunderthe foreigntradestatutes.
6 The definition of cash flows used by the ITC is income before tax plus depreciation
expense.
This content downloaded from 177.24.16.14 on Sat, 13 Apr 2013 16:12:17 PM
All use subject to JSTOR Terms and Conditions
EARNINGS MANAGEMENT DURING INVESTIGATIONS
2.2 USE OF ACCOUNTING
NUMBERS
197
BY THE ITC
The use of accountingnumbersby the ITC is not only specified in the
trade acts but is also apparent in other ways. A review of 50 petitions7
filed with the ITC for import relief investigations reveals that most
petitioners cite the poor financial condition of the domestic producersas
an indication of the industry’s need for import protection. In an article
regardingthe copperindustry’spetition for import relief, the WallStreet
Journal states that “the eleven copperproducersthat filed the complaint
cited a $623 million loss last year” (Wall Street Journal [January 27,
1984], p. 45). Another indication of the ITC’s use of accountingnumbers
is reflected in the public version of its staff reports,8which include a
section devoted to industry financial performance.An analysis of the
income statement through net operatingprofit (or loss) before taxes for
the industry is always presented. The commissioners’ injury opinions,
which are included in the ITC staff reports, always include a discussion
of the financial performance of the industry. All commissioners and
commissioners’aides that I interviewedagreedthat the financial condition of the industry is a key factor consideredin injury determinations.9
The use of accounting numbers by the ITC provides an incentive for
managersto manageearnings in orderto increase the apparentinjuryto
the firm and, thereby, the industry.10
Staff membersat the ITC indicated that the footwear industry was a
prime candidate for inclusion in this study. Their reasoning was as
IIn the case of generalescape clause investigations,”apetition for eligibilityfor import
relief for the purposeof facilitatingorderlyadjustmentto importcompetitionmay be filed
with the InternationalTrade Commission. . . by an entity, includinga trade association,
firm, certified or recognizedunion, or group of workers, which is representativeof an
industry”(TradeAct of 1974, 19 USC 2251(a)(1)).Investigationscan also be instituted by
the president,the Special Representativefor Trade Negotiations,Congress,or the ITC. In
the case of antidumpingand countervailingduty investigations,petitions can be filed by a
domesticproducer,or wholesaler,union or groupof workers,or a tradeassociation(see the
TradeAgreementsAct of 1979, 26 USC 3083(9)(C-E)).
8 The staff report is prepared by the investigative staff at the ITC for use by the
commissionersin making injury determinations.The report includes a wide range of
informationincludinga discussionof the product,domestic producers,foreign producers,
level of imports,and all the economicfactorsspecifiedin the tradestatutes to be considered
in injury determinationssuch as production,capacity utilization, financial performance,
product prices, and other causes of injury. The public version of the staff report also
includesthe final opinions of the commissioners.
‘Interviews were conductedin December1986 with CommissionersSeeley G. Lodwick
and AlfredE. Eckes and staff aides Kenneth Novak (aide to CommissionerLodwick)and
Ron Blum (aide to CommissionerDavid E. Rohr).
10Managementdiscussion and analysis sections in annual reportsand 10-Ks may also
affect the expected value of the import relief. For example, the staff report for one
investigationlists the reasons for the firms’ poor financial performanceduringthe past
threeyears as describedin the firms’annualreportsand 10-Ks (InvestigationNumberTA201-32,Publicationno. 905, p. A-44).
This content downloaded from 177.24.16.14 on Sat, 13 Apr 2013 16:12:17 PM
All use subject to JSTOR Terms and Conditions
investigation in 1984. Later that year, the ITC ruled that the footwear
industry was not being injured.One reason for this finding was the fact
that the footwear industry was relatively profitable in the most recent
periods. The petitioners later took their case to their congressional
representatives. The federal trade statutes specify that, in the case of
general escape clause investigations, one year must elapse between the
ITC’s recommendationto the president and commencementof another
general escape clause investigation unless there is good cause to forego
the requiredwaiting period (Trade Act of 1974, 19 USC 2251(e)).11The
Senate Committee on Finance issued a resolution stating that they
believed circumstanceshad changed since the ITC’s recommendationto
the president in 1984;therefore,they requestedthat the ITC institute a
second investigation. ITC ChairwomanStern found in the second investigation (1985) that “whilethe data in the previous investigation showed
that producersof the majorityof domestic productionwere experiencing
strong profits, our most recent data show otherwise” (United States
International Trade Commission, Publication no. 1717, p. 19). The
sudden and dramatic drop in profitability of the footwear industry led
some staff membersat the ITC to wonderif managersof these firms had
taken steps deliberatelyto decrease reportedearnings duringthe second
investigationperiod. ITC staff membersdo not believe that the footwear
industry is the only case in which managers may have deliberately
reducedprofits during import relief investigations; instead, they believe
that it depicts an obvious case in which deliberatereductionof profits is
a possibility.
Managershave greaterincentives to make income-decreasingaccounting choices if they believe that the regulatorsdo not completely adjust
for these choices. Neither the public nor the regulators are necessarily
thought to be “fooled”by the accounting numbersreportedby domestic
producers.Instead, the regulatorsmay be “captured”or may simply not
regard “undoing”the reported numbers to be cost effective.12On the
11The waiting period differs if the industryis grantedimport relief by the presidentas
a result of the earlier investigation. A general escape clause investigation cannot be
undertakenunless two years have elapsed since the last day on which import relief was
provided(TradeAgreementof 1974, 19 USC 2253(j)).
12 It should be noted that in 1916 the ITC (then called the United States Tariff
Commission)was establishedas a nonpartisaninformation-gatheringagency.Many of the
early advocatesof the tariff commission had hoped to ‘take the tariff out of politics’ by
establishinga nonpartisanagencyto analyzethe impactof varioustariff structures(Dobson
[1976]).But as Dobson also notes, this was an impossibletask for an issue as emotionalas
foreigntrade regulation.Over the past 70 years, Congresshas attempted to insulate the
ITC from politics by limiting the terms of the six commissioners(who are presidential
appointees) to one nine-year appointment,limiting the number of commissionersfrom
each political party to three, and staggeringthe terms of the commissioners.The overall
effectiveness of these restrictions is not central to this study, but it appears as though
politics have not been completelyeliminatedfrom the ITC.
This content downloaded from 177.24.16.14 on Sat, 13 Apr 2013 16:12:17 PM
All use subject to JSTOR Terms and Conditions
EARNINGS MANAGEMENT DURING INVESTIGATIONS
199
other hand, voters may realize that they could personally benefit from
opposingspecial interest groups (i.e., domestic producersseeking import
relief), but this personal gain might not exceed the costs of information
search, lobbying, and forming coalitions. As stated by Peltzman [1976,
p. 212], “producerprotection representsthe dominanceof a small group
with a largeper capita stake over the large group (consumers)with more
diffused interests.” In summary, while regulators might not adjust the
accountingnumbersfor various reasons, consumerswould not be able to
form effective coalitions to oppose this practice because the potential
benefit to each consumer is too small, and their interests too diverse, to
make such opposition cost effective.
Peltzman [1976] also argues that regulatorswill not exclusively serve
one economic interest (e.g., domestic producers)but will instead help the
groupsthat providethe greatest amount of personal gain to the regulator.
Thus, not all domestic industries seeking import relief can be expected
to obtain the desired relief. If all domestic producersare granted relief,
there may be less incentive for them to increase the apparent injury to
the industry.13In a subsequent section, a descriptionof the sample used
in this paper is presented which is consistent with Peltzman’s theory
that the regulators(i.e., the ITC) will not exclusively serve one economic
interest (that of domestic producers).
2.3 ITC SOURCES OF FINANCIAL INFORMATION
The ITC obtains its financial informationfromthe domesticproducers’
audited financial statements, 10-Ks, and responses to an ITC questionnaire called the Producer’s Questionnaire. The financial information
collected in the Producer’sQuestionnaireis similar to that reported in
firms’ annual reports, but it is disaggregatedby product line and/or
establishment.14Appendix B contains a summary of the data requested
in the Producer’sQuestionnaire.A company official is requiredto affirm
that the questionnaire is complete and correct to the best of his/her
knowledge and beliefs. Domestic producers are required under law to
provide the requested data and subpoenas can be used to obtain the
informationif the producersfail to comply. The ITC attempts to obtain
information from all domestic producers unless there are a very large
numberof small producers,in which case the ITC requests information
from a subset of the small producers. The ITC aggregates the data
providedby the producersinto industry totals.
13 Domestic producersmay still have incentives to increase the apparent injuryto the
industryin orderto providean “excuse”for the ITC to recommendimport relief. Also, if
the industrygrantedimportreliefappearsto be injured,it mayreducethe risk of retaliation
by foreigngovernments.
14 An establishmentis definedby the ITC as “eachfacility in the United States in which
productA is produced,includingauxiliaryfacilities operatedin conjunctionwith (whether
or not physicallyseparatefrom)such productionfacilities.”This definitionwas taken from
a sampleProducer’sQuestionnaireprovidedby the ITC.
This content downloaded from 177.24.16.14 on Sat, 13 Apr 2013 16:12:17 PM
All use subject to JSTOR Terms and Conditions
200
JENNIFER
J. JONES
The Trade Act of 1979 provides for the verification of information
submittedto the ITC by domestic producers.The Act providesthat “. . .
the administering authority shall verify all information relied upon in
makinga final determinationin an investigation…. If the administering
authority is unable to verify the accuracyof the information submitted,
it shall use the best information available to it as the basis for its
determination .. .” (Trade Act of 1979, section 776 (19 USC 1677e)).
The following details of the verificationprocess used by the ITC were
obtained from interviews with membersof the investigative staff of the
ITC.1″The ITC does not verify any informationin the audited financial
statements or 10-Ks, nor do they make any adjustments to these data.
Verificationof the submitted data is restricted to the product line and/
or establishment data providedin the Producer’sQuestionnaire.Most of
the verification process is aimed at determining whether the cost allocation methodsused were appropriateand consistently applied.The ITC
does not attempt to adjust the financial data for accountingprocedures
used or for accrualdecisions made by the firms’ managers.
3. HypothesisDevelopment
3.1 EARNINGS
MANAGEMENT
HYPOTHESIS
The ITC’s use of reportedearnings in injury determinationsprovides
an incentive for managersto make accounting choices that increase the
apparent injury of the firm. By doing so, managers may increase the
probability of obtaining the desired import relief and/or increase the
amountof relief granted;therefore,the link between accountingnumbers
and injury determinations may result in managers’ accounting choices
having economic consequences (i.e., wealth transfers from consumersto
domestic producers due to import relief). This incentive leads to the
followinghypothesis.
EARNINGS MANAGEMENTHYPOTHESIS:Managersof domestic
producersthat would benefit from import protection make accounting
choices that reduce reported earnings during ITC investigation periods
as comparedto noninvestigationperiods.
3.1.1. Conflicting Incentives. An assumption underlying this hypothesis
is that the import relief incentive to decreasereportedearningsis greater
than other incentives the managers have to increase reportedearnings.
Prior research in other contexts (see n. 2) indicates that managersface
other economic consequences of their accounting choices that motivate
them to make income-increasingratherthan income-decreasingaccounting choices-for example,debt covenants and managementcompensation
15Interviewswere conductedin December1986 with RichardLaulor,supervisorof the
FinancialAnalysis and AccountingDivision of the Office of Investigations,ChandMehta,
accountant,and Dan Dwyer,investigator.
This content downloaded from 177.24.16.14 on Sat, 13 Apr 2013 16:12:17 PM
All use subject to JSTOR Terms and Conditions
EARNINGS MANAGEMENT DURING INVESTIGATIONS
201
plans.16By increasingreportedearnings,managerscan reducethe restric-
tiveness of the debt covenants and increase their own compensation
through higher bonuses.
Debtholderswould benefit by tolerating managers’income-decreasing
accountingchoices duringimport relief investigations, even if it requires
them to waive or amend covenants that are violated,17 since the financial
performanceof the firm can be expected to improve if import relief is
granted.Managerswould also benefit from obtaining import relief if the
future earnings of the firm are higher than without the relief and higher
earnings result in higher bonus payments. Thus, during import relief
investigationsmanagershave less incentive to increasereportedearnings
than they would at other times because it is in the best interests of all
contractingparties (except consumers)for the firm to obtain the desired
importprotection.
3.1.2. Free-RiderProblem.The fact that all domestic producerswithin
an industry stand to benefit if import relief is granted results in a freeriderproblem.As a result, managersof firms within an industry may not
have equal incentives to manage earnings during import relief investigations. The ITC evaluates the overall results of the industry but does
not requirethat all firms be injuredin orderto recommendimport relief;
therefore,the managersof some domestic producersmay decide that the
results of their operationswill not alter the ITC’s ultimate decision and,
consequently,they may not manage earnings during import relief investigations. Of course, if a large numberof managersadoptedthis attitude,
then, as a group,they couldpotentially affect the ITC’s ultimate decision.
Also, some firms may not decreasereportedearningsduringimportrelief
investigationsbecausetheir managementcompensationand/or debt covenant incentives to manage earnings upward override the import relief
incentive.
In orderto address the free-riderproblem, a supplementaltest of the
earnings management hypothesis restricts the sample to firms that
petitioned for import relief. The petitioners bear the costs, thought to be
quite substantial in many cases, of supporting their claims of injury
beforethe ITC. Thus, petitioners may have greaterincentives than other
domestic producers in the industry to maximize the probability of obtaining import relief and, as a result, petitioners may have greater
incentives to manage earnings during import relief investigations.
3.1.3. InvestigationTypes.The type of import relief investigation may
16 Evidence presented in Healy [1985] indicates that managementcompensationplans
do not motivatemanagersto makestrictlyincome-increasingaccountingchoicesbut instead
the accountingchoicesdependon the relationof the earningsnumber(beforethese choices)
to any upperor lowerlimits specifiedin the plans.
is an exampleof a domesticproducerthat obtaineddebtcovenant
17 Asarco,Incorporated
amendmentswhen the covenants were violated duringan ITC import relief investigation
of the copperindustry(Asarco,Incorporated,1984Annual Report).
This content downloaded from 177.24.16.14 on Sat, 13 Apr 2013 16:12:17 PM
All use subject to JSTOR Terms and Conditions
202
JENNIFER J. JONES
also affect managers’incentives to manage earnings duringimport relief
investigations.Antidumpingand countervailingduty cases are instituted
when there is evidence that unfair trade practices exist, whereas general
escape clause cases are instituted when there is no such evidence. Managers may have greater incentives to manage earnings in general escape
clause investigations (section 201) than in antidumpingor countervailing
duty cases because “to be found eligible for relief under section 201,
industries need not prove that an unfair trade practice exists, as is
necessary under the antidumping and countervailing duty laws and
section 337 of the Tariff Act of 1930. However, under section 201, a
greaterdegree of injury, ‘serious’injury,must be found to exist.””8Since
a greater degree of injury is necessary to obtain relief, managers may
have greater incentives to decrease reported earnings in general escape
clause investigations than in other types of investigations;therefore, in
an alternative and, perhaps, more powerfultest of the earnings management hypothesis, I restrict the sample to general escape clause investigations.
3.2 INVESTIGATION
PERIOD
The ITC normallyrequests informationfor five years priorto the date
the petition was filed for general escape clause cases, and three years for
antidumpingand countervailingduty cases.”9In some cases, data for the
most recent quarterare also requested.The actual informationrequested
by the ITC in each investigation is indicated in the staff report.
The length of time taken to complete an ITC investigation depends on
the type of investigation.20The ITC must complete generalescape clause
investigations within six months after the filing of the petitions. In
antidumpingand countervailingduty investigations, the ITC must complete their investigationswithin 120 days after an affirmativepreliminary
determination(dumpingor subsidy) or 45 days after an affirmativefinal
determinationby the Secretaryof Commerce.21
Tests of the hypothesis are restricted to the two most recent years
reviewedby the ITC;the year the investigation is completed (year 0) and
the prior year (year-1).22 It is not clear when managersfirst anticipate
18United States InternationalTrade Commission,1985 Annual Report, p. 1.
19The period for which informationis requestedcan be altered under special circumstances, for example,in the case of a cyclical industry.
20 Interviewswith membersof the investigation staff (see n. 15) indicate that the ITC
normallyuses the maximumamountof time allowableto investigatecases.
21 Each final antidumpingand countervailingduty investigationconductedby the ITC
is precededby a preliminaryinvestigationin which the ITC has 45 days after the petition
is filed to determine whether there is a reasonableindicationthat the industry is being
injuredby imports.If the ITC’s preliminaryinvestigationresults in a negative determination, the case is terminated.
22 In most cases, the ITC completes its investigation in the same year the petition is
filed, but in some cases the petition is filed late in the year and the investigationis then
completedin the followingyear.None of the firms in this studywas part of an investigation
This content downloaded from 177.24.16.14 on Sat, 13 Apr 2013 16:12:17 PM
All use subject to JSTOR Terms and Conditions
EARNINGS
MANAGEMENT DURING INVESTIGATIONS
203
a future import relief investigation but is unlikely that they would
anticipate it prior to year -1; thus, they would have no import-reliefrelated incentive to manage earnings during periods prior to year -1.
The ITC does not formally request information for year 0 (except, in
some cases, for the most recent quarter),but there is evidence that this
informationenters the ITC’s decision process either through the public
hearings or by voluntary submission of the data.23Interviews with the
ITC commissionersand commissioners’aides (see n. 9) and statements
made by commissioners in their injury opinions provide evidence that
they place greater reliance on results for years -1 and 0 than on prior
years. The commissionersare looking for a downwardtrend in financial
performanceor a drastic decline in the most recent periods, providing
greaterincentives for managersto decreasereportedearningsin the most
recent periods.24
3.3 LIMITATIONS
The empirical tests might not support the earnings management
hypothesis for several reasons. First, managers may believe the ITC
adjusts for their discretionaryaccounting choices, reducingtheir incentives to use accountingchoices to manageearnings.Interviewsconducted
at the ITC indicate that the ITC does not adjust for accounting choices,
and most of the informationthat the ITC uses in its injurydetermination
is publicly available;therefore, managers should be aware of the ITC’s
practices. Second, financial performanceof the affected firms may be so
bad that managers do not need to use accounting choices to manage
earnings. If the amount of injury found by the ITC impacts the amount
of relief granted, then firms will still have an incentive to manage
earnings. Third, managers may rely on cost allocations rather than
accrualsto manageearnings for the productline investigatedby the ITC.
Cost allocations can be used by managersto shift revenues and expenses
betweenthe productline being investigatedby the ITC and other product
lines. Finally, the power of the tests may not be sufficient to detect
in whichthe ITC did not formallyrequestinformationforthe yearprecedingthe completion
of the investigation.
23 For example, data through June 30, 1980 were formallycollected by the ITC in the
automobileinvestigationbut several of the commissioners’opinions refer to results from
operationssubsequentto June 30, 1980. CommissionerStern’s opinion states that “the
currentyear (ending December31, 1980) will be the first in recent history in which the
industryshows aggregatelosses” (United States InternationalTrade Commission,Publication no. 1110,p. 119).
” For example,in the generalescape clause investigationof automobileswhich covered
the period 1975 through 1979 and the first six months of 1980, ChairmanAlbergerstated
the following in his injury opinion: “In the aggregate most of the indices of the U.S.
automobileproducers’performanceduring the period of investigation reveal a healthy
picture from 1976 through 1978 and rapidly declining trends thereafter”(United States
InternationalTrade Commission, Publication no. 1110, p. 17). CommissionerAlberger
foundthat the industrywas injured.
This content downloaded from 177.24.16.14 on Sat, 13 Apr 2013 16:12:17 PM
All use subject to JSTOR Terms and Conditions
204
JENNIFER
J. JONES
managers’income-decreasingaccounting choices. The sample selection
procedures and empirical tests described in subsequent sections are
designedto mitigate as many of these limitations as possible.
4. Data and DescriptiveStatistics
4.1 SAMPLE SELECTION
The sample used in this study is restricted to import relief investigations that requirethe ITC to makean injurydetermination:antidumping,
countervailingduty, and general escape clause investigations. Further,
only investigations pertaining to a broad product line, such as automobiles and footwear, are examined because earnings management for
narrow product lines may not be material relative to the consolidated
financial statement data used in the empirical tests. Table 1 describes
the five industries included in the sample: automobiles, carbon steel,
stainless steel, copper, and footwear. These industries represent six
investigations (as discussed earlier, there were two footwear investigations). Five of the investigationswereconsideredunderthe generalescape
clause provision, and the other was consideredunderboth the antidumping and countervailingduty provisions. Favorablerulings were issued by
the ITC in three of the investigations (stainless steel, copper, and the
1985 footwearcases), and the president grantedrelief for two of the three
(the stainless steel and 1985 footwearcases).
Relief grantedto the footwearindustry was in the form of adjustment
assistance paid to displacedworkersand, as such, did not directlybenefit
the domestic producers. Stainless steel was the only industry that obtained a substantial amount of relief as a direct result of the ITC
investigations. Voluntary import limit agreements were reached with
foreign governmentsin the automobileand carbon steel industries.
Data from firms’ annual financial statements are used to construct a
proxy for firms’ earnings management. Data for overall operations are
used rather than segment data because the segment data do not provide
enough informationto compute an estimate of firms’ earnings management (i.e., accruals). Table 2 summarizesthe sample selection criteria.
Financial data for 49 firms in the five industries are available on Compustat. One firm with foreign ownership is excluded from the sample
because this firm may not have benefited from import relief. Also, two
firms that expressed opposition to import relief are omitted from the
sample because they have incentives to increase rather than decrease
reportedearningsduringthe investigation.25Five firms are excludedfrom
25 Firms’ positions on import relief were obtained or inferred from ITC staff reports,
transcriptsof ITC hearings,industrydata identifyingimporters,and WallStreet Journal
articles. Both of the opposingfirms were substantial importersof the goods in question;
therefore,import relief may hurt these firms. Thus, these firms may have incentives to
decreasethe apparentinjuryto the firm so as to decreasethe likelihoodof obtainingrelief
and/or amountof relief granted.
This content downloaded from 177.24.16.14 on Sat, 13 Apr 2013 16:12:17 PM
All use subject to JSTOR Terms and Conditions
Cd –
a~4o
4-~~~Oc~~~
w
G)
O
~~~~~~
—
ZE-4
W
~ ~~ ~ ~ ~ ~ ~ ~~~~~4
C12
cis
–Cl)0C>
~~~
“~~~~~~~~G)~i 0)
~4-4
04-00
H >4
–
C
cis
0
0 O
G)
G)
W
0
0
0
,~4a
0
~~~~~~~L
e,~~~~0_, ~~~~~~~~~~~~~~
~~~O
0
0
‘
Q~~~~~~~~~~C~~~~Q
C~~~~~~~C~~~~
Zo
~~~~~Cd C
>
0
CI
rJn
000
0
00
000000
>0
CO)-
0
~4-1 Z
–
0 00
cisci
ci
is
‘
4
0
~~
CD
—
H4>
>
Q0
-~~~~
r
~~~D-~~~~0r
CD
CD
CD
CD
(L)
cd
0
This content downloaded from 177.24.16.14 on Sat, 13 Apr 2013 16:12:17 PM
All use subject to JSTOR Terms and Conditions
Autos Copper CStebn Footwear Steineles Total
Description
Total number of domestic producers-on
Compustata
4
Firms omitted from the sample:
Foreign ownership …………..
Opposed relief (importers)
Division/subsidiary of a firm in another
line of business .(5)
Too few time-series observations
Highly diversified firms .(3)
Total number of firms omitted from the
sample ……………………..
11
12
(1)
(1)
(4)
0
(7)
(7)
15
7
49
(1)
(1)
(2)
(4)
(2)
(5)
(5)
(8)
(10)
(7)
(5)
(26)
4
Total included in the sample ……
4
5
8
2
…..
23
a The domesticproducerswereobtainedfromthe ITC staff reports,petitions filed with the ITC, and
correspondencefiles maintainedon microficheat the ITC.
the sample because they are divisions or subsidiariesof a firm in another
line of business. Firms with too few time-series observations (less than
14 years) are also excluded from the sample as estimation of the expectations models wouldbe hindered (eight firms). Domestic producerswith
operationsin several differentindustries (highly diversified)are excluded
since earnings management for the segment being investigated by the
ITC may not be detectable in the aggregateddata (ten firms). Also, the
ITC may rely less heavily on the overall results of these diversifiedfirms
when making injury decisions. The empirical tests are based on the
resulting sample of 23 firms from five industries.
4.2 MEASURE
OF EARNINGS
MANAGEMENT
Earnings management can be achieved by various means such as the
use of accruals, changes in accounting methods, and changes in capital
structure (e.g., debt defeasance, debt-equity swaps). This study focuses
on total accrualsas the sourceof earningsmanagement.Morespecifically,
discretionary accruals are used as a measure of managers’ earnings
manipulationsduring import relief investigations. Previous studies such
as DeAngelo [1986], Healy [1985], and McNichols and Wilson [1988],
which also use some type of discretionaryaccruals measure, discuss the
partitioning of total acruals into discretionary and nondiscretionary
components.
The discretionary portion of total accruals is used in this study to
captureearnings managementrather than the discretionaryportion of a
single accrualaccount (as used in McNichols and Wilson [1988]) because
total accrualsshouldcapturea largerportion of managers’manipulations.
This content downloaded from 177.24.16.14 on Sat, 13 Apr 2013 16:12:17 PM
All use subject to JSTOR Terms and Conditions
before income taxes payable less total depreciationexpense. The change
in noncash working capital before taxes is defined as the change in
current assets other than cash and short-term investments less current
liabilities other than current maturities of long-term liabilities and income taxes payable.26This accrual measure excludes accruals related to
income taxes because the ITC bases its evaluation on income before
taxes.
4.3 DESCRIPTIVE STATISTICS
The descriptivestatistics presented here are based on the expectations
model used by DeAngelo [1986]. DeAngelo used total accruals from a
prior period (t – k) as a measure of the “normal”total accrual. She
defines the “abnormal”total accrual (ATA) as the difference between
currenttotal accruals and normal total accruals, which, in turn, can be
separatedinto discretionaryand nondiscretionaryaccruals:
ATAt = (TAt –
TAt-k)
=
(DAt –
DAt-k)
–
(NAt –
NAt-k).
(1)
DeAngelo tested whether the average value of the abnormal accrual
was significantly negative duringthe event period. This test relies on the
assumption that the average change in nondiscretionary accruals,
(NAt – NAtk), is approximatelyzero, so that a change in total accruals,
(TAt – TAtk), primarily reflects a change in discretionary accruals,
(DAt- DAtk). Table 3 summarizesscaled changes in accruals,earnings,
cash flow, and revenue before taxes for the sample for years -5 through
+1. The scaled changes are computedas first differencesof the variables
(Xt – Xt1), divided by total assets at time t – 1. Table 3 presents the
mean and median change for each of the variables,as well as the number
of negative and positive changes, t-statistics (null hypothesis that the
average change is zero), and significance levels for the nonparametric
Wilcoxon signed-rankstest.
Changes in accruals scaled by total assets are reported in panel A of
table 3. Prior to year 0, all of the accrual changes are relatively small.
The change in accruals in year 0 is negative with a t-statistic of -1.824.
If the change in accruals is viewed in isolation, year 0 suggests that
managersare making income-decreasingaccrualdecisions. These results,
however, must be interpreted cautiously because panels B through D
indicate that changes in earnings, cash flows, and revenues are also
significantlyless than zero in year 0.27
The compositionof total accruals (TAt) is as follows: TAt = [ACurrentAssetst (4) ACasht(1)] – [ACurrentLiabilitiest(5) – ACurrentMaturitiesof Long-TermDebtt(44) AIncomeTaxes Payablet(71)] – Depreciationand AmortizationExpenset(14), where the
change (A) is computedbetween time t and time t – 1; Compustatdata item numbersare
indicatedparenthetically.
27 Cash flows are definedto be earnings(or income) less accrualsthroughoutthis paper.
26
This content downloaded from 177.24.16.14 on Sat, 13 Apr 2013 16:12:17 PM
All use subject to JSTOR Terms and Conditions
co
6 – o
-4
O
o
CI
HO
OL
I
‘
o
6o o
I n
cs
C) D
CI
t1O
-zW
L
o
~
r->
CD
LOLC
00
?
?
m
0M
0
COg
0
m0)?
?
o. o ot
o
?i
6ceo6
LO
000
e
0
l~
–
00
?
3)5
,-q
r–
?
>
C2a)
a)
C)
ooo
>
CD
?l ?l ?l
2 >
;
C .~
0.
0~~~~~~~~~~00
0200202~~>2202
0
F~~~02~~~~~
F-.
0~~~~0.~~~ F-~~~~~~~02c
0
02~~~~~~~~~~~~~
~ ~ ~ ~ ~ ~ ~~
O
-/
02
02
2
0
c0
10
CZ.
-(L
–
.02
.
.02
0202~~~~~~~~~~~
This content downloaded from 177.24.16.14 on Sat, 13 Apr 2013 16:12:17 PM
All use subject to JSTOR Terms and Conditions
218
JENNIFER J. JONES
and -3.459 (with a one-tailed significancelevel of 0.0003), respectively.37
Thus, year 0 provides support for the earnings management hypothesis
whereas year -1 does not.38The Z statistic for year +1 is-1.228 with a
one-tailed significancelevel of 0.109.39Based on this test, there does not
appear to be a reversal of discretionary accruals in year +1, which is
consistent with the descriptivestatistics presented in table 3.4?
Table 6 reports, by firm and industry, change in net income, change
in cash flows, estimated nondiscretionaryaccruals,estimated discretionary accruals, and change in revenues scaled by lagged assets. The Wilcoxon signed-rankstest reveals that the discretionaryaccrualfor year 0
is significantly less than zero, with a significance level of 0.001. The
industry data are presented in order to provide some information about
the relation between the financial variables and the ultimate ITC decision. A review of the data does not result in an obvious relation between
the financialvariablesand the ITC decision. For example,the automobile
industrynot only has the largest negative cash flow change, but also the
largest income-decreasingdiscretionary accruals, yet the industry was
not deemedto be injuredby imports (i.e., the ITC issues an unfavorable
decision).Due to the limited numberof industriesincludedin this sample,
statistical testing of the relation between the various financial variables
and the ITC decision was not possible.
” When the analysis is based on the accruals measure from section 4, discretionary
accrualsare also significantlynegativein year 0 but the significancelevels are substantially
lower (t-statistic of -2.275 with a significancelevel of 0.011 when 1984 is treated as year
0, and -2.249 with a significancelevel of 0.012 when 1985 is treated as year 0). The Z
statistics are negative and insignificant for year -1 and positive and insignificant for
year +1.
38 When 1985 is treated as year 0 for the footwear industry, the Z statistic for year 0
increasesin absolute magnitudeto -3.802 with a significancelevel less than 0.0001. This
result is consistent with the notion that the footwearindustrytook additionalsteps during
the secondinvestigation(1985) to decreasereportedearnings.Since the Z statistic is based
on time-seriesregressionsthat control for the effect of changingrevenueson accruals,the
decreasesin accrualsreportedin table 3 do not appearto be caused entirely by decreases
in revenues.
39 When 1985 is treated as year 0 for the footwearindustry,the Z statistic decreasesin
absolutemagnitudeto -0.799 with a one-tailed significancelevel of 0.212.
40 Z statistics are also computed for years -5 through +1 using regressionequations
estimatedover periodsusing all availabledata throughyear -6 as the basis for obtaining
the “normal”accrual.This allowedfor a wider comparisonof significancelevels for the Z
statistics obtainedfor investigationyears to those obtainedin noninvestigationyears. The
Z statistics for years -5, -4, -2, and -1 are negative and relatively small comparedto
their standarderrors,with Z statistics rangingfrom -0.274 to -1.021. The Z statistic for
year 0 is also negativebut moresignificant(-3.137 with a significancelevel of 0.0008)than
for any of the other years, lending supportto the earningsmanagementhypothesis.The Z
statistic is positive for year -3 but is not statistically significant.The Z statistic for year
+1 is negativeand relativelysmall (-1.552) which does not indicatethat managersreversed
their income-decreasingaccrualsin year +1. The results of this estimation are consistent
with those reportedin table 5.
This content downloaded from 177.24.16.14 on Sat, 13 Apr 2013 16:12:17 PM
All use subject to JSTOR Terms and Conditions
EARNINGS MANAGEMENT DURING INVESTIGATIONS
5.2 TEST
219
FOR MODEL MISSPECIFICATION
Scatter plots of the regression equation residuals do not exhibit a
nonlinear relation between abnormal accruals and changes in revenue;
therefore,the negative residualsfor year 0 do not appearto be the result
of this type of model misspecification. In order to obtain additional
information regarding possible model misspecification for periods in
which there are large decreases in revenues, the accruals expectations
model is estimated for 459 firms that are not in the ITC investigation
sample. The 459 firms represent all firms not in the ITC investigation
sample having 25 years of data available on Compustat(1961-85). The
residuals for 1980 through 1985 are each divided by the standard error
from the estimated regression model resulting in 2,754 Vitvalues. The
Vitsare divided into deciles based on the size of the change in revenues
scaled by assets.
The first two columns of table 7 reportthe averagechange in revenue
scaled by lagged assets (AREVitlAit-1)and Vit for each decile. Visual
inspection of AREVitl/Ait1and Vitacross the deciles does not reveal any
systematic relation between the two variables. A systematic relation
between the two variables might indicate that the nondiscretionary
accrualsmodel is misspecified.Since the year 0 mean change in revenues
scaled by lagged assets for the ITC investigation sample (-0.188 from
table 3) falls within decile 1 in table 7, it is also important to compare
the average Vitfor decile 1 to all other deciles to provide some evidence
that the nondiscretionaryaccrualsmodel is not misspecifiedfor extreme
decreasesin revenues.
Paired comparisonsfor the mean Vitsby decile are computedand the
resulting significance levels are reported in table 7. The question of
interest is whether the mean Vit for the decile containing the largest
decreasein revenues (decile 1) differs from the other deciles. If decile 1
is found to differ significantly from the other deciles, it could indicate
that the accruals expectations model is inappropriatewhen changes in
revenues are large and negative. The results in table 7 indicate that the
most significant differencebetween decile 1 and any of the other deciles
is significant at the 0.171 level. The mean Vitfor decile 1 is greaterthan
the means for deciles 3 and 5 and less than the means for the other
deciles. Decile 3 has the smallest mean Vitvalue and is the only decile
that differs from other deciles at significance levels of 0.10 or less. This
analysis provides some evidence that the significant negative Z statistic
in year 0 for the ITC sample is not due to the relative inability of the
expectationsmodel to predict accrualsduringperiods of severe economic
downturn.
5.3 SENSITIVITY
ANALYSES
Sensitivity analyses, not reportedhere, of the results reportedin table
5 were conductedto ascertainthe effect variouscompaniesand industries
This content downloaded from 177.24.16.14 on Sat, 13 Apr 2013 16:12:17 PM
All use subject to JSTOR Terms and Conditions
220
JENNIFER J. JONES
0)
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~-4~~~
1-
Cq
N
I
Cl)~~~~~~~~~~~~~~~~~~~~C
m
~~~~~~0
ii~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~C
CThW
cq
0)
00
~
~
~
-d
C~o
666I c;
0
0
4
>
0)~~~~~~~~~~~
0)1-Cl))
O
4.5~0
T
Cd
~ ~ ~ ~ ~ -C.C~t-.d
~ ~ 0l)~~~~~~0)~~~~~
~~~
CI)
CD0
0~~~~~~~~~~
00)
00
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~l
>) o
I~~~~~~~~~~~~~~~~~~~~~~~~~~~~~m0
LO Ce ~
Ce
U
a)-iS
Lo Cd
C01-LOCOOCO,-4~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~C
0
~
N
~
~
~
~ -1-COO~~~~~~~~a
Q4)
00’-
0
I
I
-u~-
1111.I;,-
U~ ooaor-~~~
~~~o~~~Co
~~~~~oooi-0~
N
l)
dC
0
1
~~~~~~~~~~~~~~~~0
00
0
—
1.
CeDt-LoCe
C
4
cd
CZ)
0
CD –
NN00
0
O-4
0
cu~~~~~~~~dl
C~~~~~~~0N
~ ~
cu
00
CO
ot
C
E-4
0Nc)
0
CN
~
0
CA)~~~~~~~~~~~~~~~~l
00
r.
U
0
“O~~~
l.l
Cl)0
~~~~~~
~~~~~~~~~~
elCD~o
)l
)C
LOI
al)
r
.-*Q
C 4 4lIo
40
cq
t-
t-
oo LO
C)
00
~Jo00
“t
cq
cu
4mm,
0~~~~~~~~-C
This content downloaded from 177.24.16.14 on Sat, 13 Apr 2013 16:12:17 PM
All use subject to JSTOR Terms and Conditions
CCO
0
EARNINGS MANAGEMENT DURING INVESTIGATIONS
221
had on the Z statistic. If firm 23 (an auto company) or the entire auto
industry is excluded from the analysis, the resulting Z statistics are
negative and significant at the 0.002 and 0.005 levels, respectively.If the
footwear industry is omitted from the sample because a second investigation was conducted in this industry in 1985 (year +1 in table 5), the
absolute magnitude of the Z statistic increases (-3.771) for year 0 and
decreases (-0.562) for year +1.
5.4 ALTERNATIVE TESTS
Table 8 presents results for two alternative tests. As discussed in
section 3, managers may have greater incentives to manage earnings if
they are petitioners or if the ITC investigation is being conductedunder
the general escape clause. The Z statistic for the petitioning firms for
year 0 (-2.833) is smaller in absolute magnitude than for the entire
sample.The lower Z statistic may be due to a smaller sample size. Thus,
the average Vis (Va) are comparedto the entire sample rather than the
Z statistics. In year 0, V, is greater in absolute magnitude for the
petitioners (-0.800) than it is for the entire sample (-0.760), indicating
TABLE
8
AlternativeTests of the EarningsManagementHypothesis
AverageVip
ZvDPt
Year -1b
Year0
Year +1 Year -1
Year0 Year +1
All companies, see table 5
(n = 23) ………………
-0.372
-3.459 -1.228 -0.082 -0.760 -0.270
Include only petitioners
(n = 14) ………………
-0.264
-2.833 -1.097 -0.074 -0.800 -0.310
Include only escape clause
cases (n = 18) ………….
-2.772 – -1.058 -0.196 -0.690 -0.263
-0.788
The Zvpstatistic is calculatedas Zi’j V,, [EN, (T, – 3)(Ti – 5)]- , where T, is the numberof years
in the estimationperiod. Vi,,is computedas ui,,/(si(l + C-‘)), whereCii,= [X,(X’X)-‘X,’] in which X
is the matrix of independentvariablesfor the estimation period,Xp is the matrix for the prediction
period,uipis the predictionerror,p is the predictionyear,and si is the standarderrorfromestimatesof
the followingregressionmodel:
= aj[11/Ajt_]
+ 3,j[AREVat/Ajt_,]
+ 32i[PPEit/Ait-.] + fit,
TAj,1Aj,_j
where:
TAit= total accrualsin year t for firm i;
AREVit= revenuesin year t less revenuesin year t – 1 for firm i;
PPEit= grossproperty,plant, and equipmentin year t for firm i;
Ait-1= total assets in year t – 1 for firm i;
Ej,= errorterm in year t for firm i;
i= 1, . . , 23 firm index;
t = 1, . . ., Ti,year index for the years includedin the estimationperiodfor firm i.
The compositionof total accruals (TAt) is as follows: TAt = [ACurrentAssetst (4) – ACash,(1)] [ACurrentLiabilities,(5)] – Depreciationand AmortizationExpense, (14), where the change (A) is
computedbetweentime t and time t – 1; Compustatdata item numbersare indicatedparenthetically.
The regressionequationsare estimatedover all availableyearspriorto year -1.
b Year0 is the yearthe ITC completedits investigation.In this table,year0 for the footwearindustry
is 1984.
cThe numberof observationsincludedin the Z statistics is n.
This content downloaded from 177.24.16.14 on Sat, 13 Apr 2013 16:12:17 PM
All use subject to JSTOR Terms and Conditions
JENNIFER
222
J. JONES
that discretionaryaccruals for the petitioners are more income-decreasing. When the sample is limited to general escape clause investigation
firms,both the Z statistic (-2.772) and Vp(-0.690) for year 0 are smaller
in absolute magnitude. As such, the results do not indicate that these
firms made more income-decreasingaccruals than firms being investigated under the countervailingduty and antidumpingstatutes.
5.5 PORTFOLIO TEST
As noted earlier,cross-sectionalcorrelationresults in violations of the
assumptionsunderlyingthe Z test statistic. Since the firms are clustered
by industry and, within industry, by time there is potential for crosssectionally correlatedaccruals.One way to addressthe problemof crosssectional correlationis to groupthe firms by industry and to analyze the
discretionary accruals for the industries. One such method of testing
portfoliopredictionerrorsis set forth in Mandelker[1974].The residuals
(i.e., discretionary accruals) from equation (2) are averaged across all
firms within an industry for each time period during the regression
estimation period.4″The estimated standard deviation of these average
residualterms is computedfor each portfolio (industry).Averageprediction errors are computed for each portfolio for year 0 and standardized
by the estimated standarddeviation for the portfolio as follows:
USffi = Ufp/af,
(6)
where:
usfp= averagestandardizedpredictionerrorfor portfoliof at time p;
ufp= averageprediction errorfor portfoliof at time p;
af = estimated standarddeviation of the portfoliof residuals;
f = portfolio (industry);
p = time period.
A t-statistic is computed to test whether the average prediction errors
are different from zero as follows:
Tp= usp/[S/(n_)1/2]
(7)
where:
usp = averageof usfpacross all portfolios f at time p;
S = estimated standarddeivation of lisp, (S = 1);
nf =
numberof portfolios.
41 The regressionequationsare estimatedover the same numberof time periodsfor each
firm within an industry, although the number varied across industries. The number of
periodsincludedin the estimation is limited to the numberof observationsfor the firm in
each industrywith the fewest availableobservations.When the same estimation periodis
used to computeZ statistics for each firm in an industry,the results are -0.754, -3.290,
and -0.953 for year -1, year 0, and year +1, respectively(footwearyear 0 is 1984). These
results are similar to those using the maximumnumberof availableobservationsfor each
firm,which are reportedin table 5.
This content downloaded from 177.24.16.14 on Sat, 13 Apr 2013 16:12:17 PM
All use subject to JSTOR Terms and Conditions
which is significant at less than the 0.001 level. When each industry is
omitted from the t-statistic calculation (one at a time), the resulting
four-industryt-statistics rangefrom -3.635 to -5.035. The portfoliotests
indicate that after the problemsrelatedto cross-sectionalcorrelationare
mitigated by groupingfirms into industry portfolios (which also results
in grouping by year), the discretionary accruals in year 0 are still
significantlyincome-decreasing.The results also indicate that the significant t-statistics are not the result of a single influential industry.
In summary, the empirical tests using total accruals indicate that
discretionaryaccrualsare income-decreasingin year 0, providingsupport
for the earnings managementhypothesis. Discretionaryaccrualsare not
significantlydifferent from zero in years -1 and +1.
6. Conclusions
The results of the empirical tests reported here support the earnings
managementhypothesis suggestingthat managersmake income-decreasing accruals during import relief investigations. Discretionary accruals
are more income-decreasing during the year the ITC completed its
investigation (year 0) than would otherwise be expected. Tests of the
earningsmanagementhypothesis are based on firm-specificexpectations
models used to estimate “normal”total accruals.These models allow for
changes in nondiscretionary accruals that are caused by changes in
economic conditions. The expectations models developedhere represent
an attempt to improveupon the measuresof discretionarytotal accruals
used in prior research;specifically, time-series models are developed to
estimate total nondiscretionaryaccruals and cross-sectional tests of the
earningsmanagementhypothesis are appliedto the resulting discretionary accrualsmeasure.
In addition to providing evidence that managers manage earnings
during import relief investigations, the results of this study may prove
useful to regulators at the ITC. The ITC may benefit by taking into
accountthe evidence providedherein that managersappearto be making
income-decreasingaccrualsduringimport relief investigationperiods. Of
course,the ITC relies on several factors in makingtheir injurydecisions
which may reduce the problem of relying on the reported earnings
numbers.42
42 For example, injury itself is necessary but not sufficient to obtain a favorable injury
decision. The injury must be shown to have been caused by foreign imports.
This content downloaded from 177.24.16.14 on Sat, 13 Apr 2013 16:12:17 PM
All use subject to JSTOR Terms and Conditions
224
JENNIFER J. JONES
~~~~~
~~~oQ
b0
rJ2~~~~~~
.a
‘-~
ca.o
a
P-1C –4
-0-
J4’2)
v
– 4
is
X-=eo
~ –.
0
I
0
4-a
QA)
~&
>~~~~~~~~
i?
Uok
~~~~~~
40
4
I-
0
M .
b
(;
~~~~~~~~~~~d0
a)
1
_
4
;4
-4~~~~~~~~~~~~~~~~~~~~~~or
–
Ci2~~~~~~~~~~~~~
–
4-
-4i
–.
~
4 4;4
~~~~~~~~~~~~$-
a)
~~~~~~~~~a)
0~~~~~~~4
)
P
“-4
,
2
–
a
cis
~~~~~~
Ci2~~~~~~~~~~~;4
.4
;4 a
0
)44-)
-4-~~~~~~a
-) 14> I L4 4-
-4-
QCI
This content downloaded from 177.24.16.14 on Sat, 13 Apr 2013 16:12:17 PM
All use subject to JSTOR Terms and Conditions
–
2-‘
ISC4 –
i
EARNINGS MANAGEMENT DURING INVESTIGATIONS
*
T
a)~~~~~z
a)
~~~~4-
I
TZ0)i
00-
o.
b
co0
CO,
(2~~~~~~~~~)
~~~~~a)
~ ~ ~~~4’
~
_
~
c~~~C)
cJ2~~~~~~c
*-
S
1a)~~~~~~i)Q
a)
p.4
ZI)
cis~~~~~.4
Cl
Ci2~~~~~~
.
Z Z~~~~ci
~~~~0~~~~~s
~ i
Z
CZcoa)Q-~1a )
rn
Q h
L
nb
This content downloaded from 177.24.16.14 on Sat, 13 Apr 2013 16:12:17 PM
All use subject to JSTOR Terms and Conditions
225
226
JENNIFER J. JONES
.
0
cis
o
i
0)0
Q
3)
4
X
S
c12gr
5
;;
:
–
-4~~~~~~-
~~~~~;-
0)
I
4-~QL4
I
+
c
4z’Su0
X (2)
o IL.4
c12
Z O ~~~~~~~~~~~~~~I0c
~~~~~1
0
cis
~
0)
Hoi
X
~~~~~~~~~~~~C
cis
cis
?
P X R
0
cis
4-4~~~~~~~~~~~-
.4
u.
z
z
This content downloaded from 177.24.16.14 on Sat, 13 Apr 2013 16:12:17 PM
All use subject to JSTOR Terms and Conditions
0
EARNINGS MANAGEMENT DURING INVESTIGATIONS
227
APPENDIX B
InformationCollectedin the Producer’sQuestionnairefor ITC Final
Investigations
A. Practical Annual Capacity43
B. Production
C. End-of-PeriodInventories
D. Purchases
1. Imports
2. Other Purchases
E. Shipments
1. Intracompanyand IntercompanyTransfers
2. Domestic Shipments
3. Export Shipments
4. Amounts Made to End Users and Distributors
F. Employmentand Wages
G. Income-and-Lossand Other Financial Information
1. Methods of Allocation
2. Income from Operations
3. List of Unusual or NonrecurringExpenses
4. Asset Valuation:OriginalCost and Book Value
5. CapitalExpenditures
6. Capital and Investment-Actual and potential negative effects of
imports on firm’s growth, investment, and ability to raise capital
7. Research and Development Expenses
H. Prices
L Competitionfrom Imports-Price Suppression/Depression
J. Competitionfrom Imports-Lost Sales
REFERENCES
DEANGELo,L. E. “AccountingNumbers as Market Valuation Substitutes: A Study of
ManagementBuyouts of Public Stockholders.”The AccountingReview61 (July 1986):
400-20.
. “ManagerialCompetition,InformationCosts, and CorporateGovernance:The Use
of AccountingPerformanceMeasures in Proxy Contests.”Journal of Accountingand
Economics10 (January1988):3-36.
DOBSON,J. M. Two Centuriesof Tariffs: The Backgroundand Emergenceof the U.S.
InternationalTradeCommission.Washington,D.C.:GovernmentPrintingOffice, 1976.
HEALY,P. “The Impact of Bonus Schemes on the Selection of Accounting Principles.”
Journalof Accountingand Economics7 (April 1985):85-107.
43 Companiesmust report all informationby establishment, and if 85% or less of the
establishmentis devoted to the product line in question, the informationmust also be
providedby productline. An establishmentis defined by the ITC as “eachfacility in the
United States in which product A is produced,including auxiliary facilities operated in
conjunctionwith (whetheror not physicallyseparatefrom) such productionfacilities.”For
preliminaryinvestigations,less detailedinformationis collectedin parts B, C, D, E, and G
of the questionnaire.This informationwas obtained from a sample copy of a Producer’s
Questionnaireprovidedby the ITC.
This content downloaded from 177.24.16.14 on Sat, 13 Apr 2013 16:12:17 PM
All use subject to JSTOR Terms and Conditions
and InvestmentDecisions.”Ph.D. dissertation,Universityof Michigan,1988.
KAPLAN, R. S. “Commentson Paul Healy: Evidence on the Effect of Bonus Schemes on
AccountingProcedureand AccrualDecisions.”Journal of Accounting and Economics 7
(April1985):109-13.
-. “Developinga Financial Planning Model for an Analytical Review:A Feasibility
Study.” In Symposium onAuditingResearch III, pp. 3-34. Urbana-Champaign: University
of Illinois, 1979.
KMENTA, J. Elements of Econometrics. New York:Macmillan,1986.
LIBERTY, S. E., AND J. L. ZIMMERMAN. “LaborUnion ContractNegotiationsand Accounting Choices.”The AccountingReview61 (October1986):692-712.
LIPE,R. C. “The Information Contained in the Components of Earnings.”Journal of
Accounting Research 24 (Supplement 1986): 37-64.
MANDELKER, G. “Risk and Return: The Case of Merging Firms.”Journal of Financial
Economics1 (1974):303-35.
M., AND G. P. WILSON. “Evidence of Earnings Management from the
Provisionfor Bad Debts.”Journalof AccountingResearch26 (Supplement1988):1-31.
PATELL, J. M. “CorporateForecasts of Earnings per Share and Stock Price Behavior:
EmpiricalTests.”Journal of Accounting Research 14 (Autumn1976):246-76.
PELTZMAN, S. “Towarda More General Theory of Regulation.”Journal of Law and
Economics19 (August1976):211-40.
RAYBURN, J. “The Association of Operating Cash Flows and Accruals with Security
Returns.”Journal of Accounting Research 24 (Supplement1986):112-33.
THEIL, H. Principles of Econometrics. Santa Barbara,Calif.:Wiley, 1971.
UNITED STATES CONGRESS (93d). TradeAct of 1974, Public Law 93-618,January3, 1975.
UNITED STATES CONGRESS (96th). Trade Agreements Act of 1979, Public Law 96-39,
July 26, 1979.
UNITED STATES INTERNATIONAL TRADE COMMISSION. InvestigationNumberTA-201-32,
Publicationno. 905, August 1978.
InvestigationNumberTA-201-44,Publicationno. 1110, December1980.
InvestigationNumberTA-201-55,Publicationno. 1717, August 1985.
Annual Report,Publicationno. 1847,December1985.
McNicHoLs,
This content downloaded from 177.24.16.14 on Sat, 13 Apr 2013 16:12:17 PM
All use subject to JSTOR Terms and Conditions
THE ACCOUNTING REVIEW
Vol. 91, No. 2
March 2016
pp. 559–586
American Accounting Association
DOI: 10.2308/accr-51153
Managing for the Moment: The Role of Earnings
Management via Real Activities versus Accruals in SEO
Valuation
S. P. Kothari
Massachusetts Institute of Technology
Natalie Mizik
University of Washington
Sugata Roychowdhury
Boston College
ABSTRACT: We assess the role of both accruals manipulation (AM) and real activities manipulation (RAM) in
inducing overvaluation at the time of a seasoned equity offering (SEO). Our results reveal that earnings management
is most consistently and predictably linked with post-SEO stock market underperformance when it is driven by RAM;
in particular, the opportunistic reduction of expenditures on R&D and selling, general, and administrative activities.
Thus, overvaluation at the time of the SEO is more likely when managers actively engage in more opaque channels
to overstate earnings. Our findings are particularly relevant because managers exhibit a greater propensity for RAM
at the time of SEOs, even though RAM is more costly in the long run.
Keywords: earnings management; discretionary accruals; real activity manipulation; seasoned equity offering;
SEO overvaluation; SEO return reversal; earnings management opacity.
JEL Classifications: G14; G31; M4; M41.
I. INTRODUCTION
M
anagement is an important source of financial information to investors. The voluminous earnings management
literature demonstrates that managers misrepresent, typically positively, the firm’s financial information in the hope
of skewing the firm’s stock market valuation upward. The extent to which earnings management strategies succeed
in misleading investors depends in part on their relative opacity, i.e., the degree to which external investors can detect and
unravel earnings management. Earnings management can occur through two channels: accruals management (AM) and real
activities management (RAM). Our objective is to assess whether the market is indeed misled by earnings management in a
setting characterized by considerable scrutiny of the firm’s financial statements and its operations—the seasoned equity offering
(SEO). Of particular interest is the relative opacity of real earnings management versus accruals management.
Our primary result is that evidence of overvaluation at the time of an SEO is robust among firms overstating earnings
through real activities management, but not accruals management. Accruals management is reliably associated with negative
future returns only when it is simultaneously accompanied by real activities management. Both real and accrual earnings
management strategies, individually and jointly, forecast negative future operating performance. Against this backdrop, our
findings suggest that investors’ ability to detect earnings management and to assess its consequence for future performance is
impaired more severely when real activities are the basis for earnings management.
The authors are grateful for comments from John Harry Evans III (editor), two anonymous reviewers, Amy Hutton, Robert Jacobson, Shiva Rajgopal, Ewa
Sletten, Susan Shu, seminar participants at Boston College, and conference participants at the 2012 Nick Dopuch Annual Conference at Washington
University in St. Louis.
Editor’s note: Accepted by John Harry Evans III.
Submitted: January 2013
Accepted: May 2015
Published Online: May 2015
559
560
Kothari, Mizik, and Roychowdhury
In accruals-based earnings management, managers intervene in the financial reporting process by exercising
discretion and judgment regarding accounting choices. Importantly, accruals management misrepresents the firm’s
underlying operating performance, but does not generally involve altering operations themselves. Real activities
manipulation, on the other hand, entails departures from normal operations with the intent to mislead at least some
stakeholders into believing that the reported financial performance has been achieved in the normal course of operations
(Roychowdhury 2006). Real activities manipulation can manifest in many forms, including decreased investment in
research and development (R&D), advertising, and employee training, all for the purpose of meeting short-term goals
(Graham, Harvey, and Rajgopal 2005; Roychowdhury 2006). Marketing strategies, tactics, and budgets are often at the
center of implementing real activity-based earnings management, as well (Moorman, Wies, Mizik, and Spencer 2012;
Chapman and Steenburgh 2009).
A crucial difference between accruals and real earnings management arises because generally accepted accounting
principles (GAAP) provide a framework for ‘‘acceptable’’ accounting principles enforced by the Securities and Exchange
Commission (SEC), but no such framework for real operations exists. Auditors look for whether firms’ accounting practices
meet GAAP. Detectable departures from GAAP can impose substantial costs on firms, managers, and auditors via regulatory
investigations, restatements, and personal penalties. Real activities-based management, in contrast, arises from managers’
investment and operating decisions. Shareholders delegate to managers the rights to make investment decisions because
managers possess superior information relative to external stakeholders in making those decisions. Thus, the detection of RAM
is conceivably more challenging for investors than that of accruals management.
Existing literature documents that real manipulation is more likely when accounting practices are under greater scrutiny.
For example, Cohen, Dey, and Lys (2008) find greater incidence of real earnings management in the period following the
implementation of the Sarbanes-Oxley Act (SOX) in 2002, which sought to limit questionable accrual choices. Cohen and
Zarowin (2010) and Zang (2012) report that firms whose auditors are likely to be more vigilant with respect to accounting
choices engage more in RAM. While regulators and auditors understandably focus on a firm’s accounting choices, the larger
financial community includes those who analyze a company’s operations—for example, financial analysts and investors.
Financial analysts spend considerable resources to analyze not just managers’ accrual choices, but also their operational and
strategic decisions. Further, existing literature suggests that the presence of institutional investors constrains real activities
earnings management (Bushee 1998; Roychowdhury 2006; Zang 2012). Definitive evidence on the market’s relative ability to
unravel real versus accruals management conditional on the presence of such manipulation is lacking, however. If equity
investors are indeed misled by overstated earnings, then the firm’s stock price would be artificially inflated and, in future years,
we would observe return reversals as the market eventually learns that prevailing valuations are unsustainable. Relative opacity
determines the extent to which real earnings management and accruals management are differentially associated with future
return reversals.
In a related study, Cohen and Zarowin (2010; hereafter, CZ), find that earnings manipulation via both accruals and real
activities is associated with poor future earnings performance, although the association is stronger when real activities are
involved. Our study differs from CZ in two respects. First, CZ focus on the costs of real earnings management, in terms of its
association with declines in future operating performance. In contrast, we examine the opacity of earnings management, in
terms of the extent to which the market is unable to detect the earnings management and is resultantly misled into
overvaluing the stock. Our objective requires a stock return-based test: if the market is adept at unraveling earnings
management, then any future earnings decline would be anticipated and factored into current prices, which would have no
further implication for stock returns. However, earnings management that is opaque to market participants will lead to
overvaluation of firms managing earnings, and these firms will exhibit future stock return reversals as the overvaluation is
‘‘corrected.’’
Second, we recognize that the opacity of both accruals and real activities manipulation can depend on whether managers
engage in the two strategies individually or in conjunction. Our research design accommodates these multiple possibilities, and
in doing so, we discover strategies that investors (markets) find most difficult to unravel.
SEOs provide an excellent setting for assessing the capital market consequences of earnings inflation via real activities or
accruals. An SEO is a well-identified event around which there is a significantly increased emphasis on a firm’s operations and
its stock price. Managers clearly have incentives to inflate earnings, since overvaluation at the time of an SEO results in a
wealth transfer from prospective shareholders to the firm and its current shareholders (see Teoh, Welch, and Wong 1998;
DuCharme, Malatesta, and Sefcik 2004; Rangan 1998; Cohen and Zarowin 2010). On the other hand, financial analysts and the
investment community also subject firms to considerable scrutiny at the time of securities issues (see Ball and Shivakumar
2008). Therefore, only highly opaque earnings management strategies are likely to escape detection. Thus, the SEO provides a
powerful setting to examine the relative opacity of real activities versus accruals manipulation.
Simple frequency analyses reveal that the incidence of firms attempting to overstate earnings via RAM is significantly
higher in SEO years than in non-SEO years. In contrast, the likelihood of firms engaging solely in accruals management to
The Accounting Review
Volume 91, Number 2, 2016
Managing for the Moment: Earnings Management via Real Activities versus Accruals in SEO Valuation
561
inflate earnings (without any RAM) is similar in SEO and non-SEO years. The patterns suggest that managers’ propensity to
engage in RAM is higher in SEO years than in other years, and it is higher than the propensity for accruals-based earnings
management.
In subsequent analysis, we examine the operating performance of SEO firms in the years following the SEO. We find that
firms that overstate earnings through RAM experience negative operating performance relative to their Barber and Lyon (1997)
matched counterparts in each of the three years following the SEO. Firms relying solely on accruals (but not real activities) to
overstate earnings at the time of the SEO exhibit evidence of negative operating performance in future years as well, although
not as severe as those engaging in RAM. The results on operating performance are largely consistent with Cohen and Zarowin
(2010).
Our returns-based tests underscore the salience of real activities-based earnings management as the driver of future
negative stock price performance. First, future price performance is significantly negative over each of the three post-SEO
years f…