review

The three articles Summarize the articles Identify the main arguments, findings, or results of each article Compare and contrast the articles, focusing on their commonalities and differences 800 – 1000 words

Mobility and Globalization
Habibul Haque Khondker

abstract
Mobility, viewed empirically as the movement of people, capital, technol-
ogy, institutions, ideas, ideological systems, and knowledge, is the most
visible face of globalization. Mobility is central to the process of globaliza-
tion marking ageless continuity. Although some writers define the present
phase (twentieth and twenty-first century) of globalization as the “age of
migration,” historical human migration—both involuntary and volun-
tary—characterized earlier phases of history just as well. The so-called free
movement of labor in the present phase of globalization hides forms of
slavery and bonded labor that continue to characterize twenty-first-cen-
tury globalization. An examination of the mobility of people in the first
quarter of the twenty-first century will illustrate the seesaw-like tendency
of a borderless and bordered world, which reveals the contradictions of
globalization with implications for both mobilities of people and dissemi-
nation or mobility of scientific knowledge and technology. The present
chapter takes an interdisciplinary perspective to examine the intersection-
ality of mobility and globalization.

keywords
globality, globalization, knowledge, migration, mobility

The link between mobility and globalization, two key concepts that straddle several
social sciences, can be understood in several ways. Studies on both mobilities and
globalization combine geographical, sociological, political-economic, and histori-
cal approaches. Terms such as mobility and circularity have been used with great
frequency in the discussion of globalization and global history in recent times
(Gänger, 2017). Titles such as connected history (Subrahmanyam, 1997, 2022), fol-
lowed by connected sociology (Bhambra, 2014), have come into circulation in recent
decades. Connectivity, mobility, and globality have become part of a conceptual

60 Globalization: Past

assemblage. A turn to mobility and the emergence of a mobility paradigm took
place in the first decade of the twenty-first century (Sheller & Urry, 2006).

Both globalization and mobility are polysemous terms. The various approaches
to globalization are narrowed into two. In one approach, often held by journal-
ists, some economists, and politicians, globalization is viewed as a new label for
neoliberal capitalism writ large. In this view, globalization is an enabler of the
mobility of people, capital, and technology. Yet the process is not without con-
tradictions. While the advocates of neoliberal globalization applaud the mobility
of the professional classes, the mobility of the working class is a source of concern
for them. They view unrestricted mobility of capital as a boon since it fuels glo-
balization, but unrestricted mobility of labor might jeopardize political stability,
giving rise to xenophobia (Steger, 2003: 118).

A broader, holistic, historically rooted sociological view of globalization views
mobility of labor, capital, technology, ideas, religions, ideological systems, knowl-
edge, lifestyles, and cultural products as a natural process that has come to be
restricted or moderated by the rise of the states and the ideologies of extreme
nationalism. In the present chapter, we adopt a sociological view of globalization
and consider mobility in broader terms that include both artifacts and ideas, yet
human mobility remains its central component.

Human mobility can be viewed in three broad phases. For millennia of foraging
and pastoral living, human mobility was the norm, as life was unencumbered by
the boundaries of states or functional equivalents of such organizations. Histori-
cally, migration has been the normal condition (Manning, 2005). People migrated
in large numbers from Africa to the rest of the world. Asian migrants populated
the Arctic and reached North America well before the continental drift. Since the
end of pastoral society and with the advent of agricultural societies, people have
been largely sedentary, but not immobile, as mobility was limited to seasonal or
short-distance migration. With the advent of modernity triggered by industrial-
ization, mobility has become commonplace. Mobility, in that sense, can be seen
as a major marker of modernity. The “mobility hypothesis,” which was advanced
by Wilber Zelinsky (1971) and supported by Charles Tilly (1978), argued that the
imperatives of capitalism created mobile free labor through a process of proletari-
anization. The view that links population mobility to industrialization has come
under critical examination (Hochstadt, 1989; Lucassen & Lucassen, 2009).

The debate, to some extent, hinges on the definition and scale of migration. The
movement of people from one place to another is nothing new. Historically, there
has been a natural flow of people, for example, during harvests, which would
be regarded today as seasonal migration. Indeed, people moved from place to
place, individually or as a group, for a better life. Emigration to the so-called
New World, rural to urban migration, and the movement of soldiers and sail-
ors predated industrialization (Lucassen & Lucassen, 2009). At least in theory,
people could move more freely across geographical boundaries until the invention

Mobility and Globalization 61

of the passport and the monopolization of control of the means of movement
by the state in the nineteenth century (Torpey, 2000). In Europe until World War I,
the movement of people took place quite freely. During and after World War I, as
mass travel expanded and borders became more rigorously guarded, the regula-
tion and monitoring of human movement by the state became more determined
(Torpey, 1998: 254). At the beginning of the twenty-first century, global citizens
traveled a total of 23 billion kilometers; by 2050, that figure will have risen to
105 billion (Schafer & Victor, 2000: 171).

MOBILIT Y AND GLOBALIT Y

In the current global circumstances, mobility has become one of the central fea-
tures of globality. Mobilities of technology, tools, and ideas have a long history.
With revolutionary improvements in transportation and communication, a vast
number of people around the world have become highly mobile. For certain
classes of people, geographical constraints have lost significance, and societies and
nation-states have become more permeable. While there is a class dimension to
mobility, since only the middle class and upper classes can travel, a growing num-
ber of working-class people are also resorting to travel across national boundaries
and within their respective societies in search of livelihoods. As such, mobility and
migration have become synonymous.

The COVID-19 pandemic, by denting both globalization and mobility, unwit-
tingly revealed the close relationship between the two. However, if globalization
is defined in a sociological way, focusing on historical encounters among civiliza-
tions, societies, and cultures, mobility becomes a defining feature of globalization.
Consider the definition of globalization by Roland Robertson (1992), for whom
globalization entails that the world becomes a single place followed by a growing
awareness of the shrinking of the world. In this process, mobility is both a cause
and consequence of globalization. John Urry (2000), among others, reframed
modern society as a mobile society, emphasizing the mobility of corporeal bodies,
that is, people, but also a way of life, that is, ideas (Urry, 2002). Corporeal mobil-
ity, a distinctive feature of modern global society, was interrupted by immobility
caused by the coronavirus pandemic in 2020–21. Even without a huge public
health emergency, one cannot overlook the contradictions of the modern world
(circa the second decade of the twenty-first century): the co-presence of mobility
and the increasingly bordered and fenced nature of the world, which go hand in
hand with the advancing march of globalization (Turner & Khondker, 2010).

Mobility, viewed empirically as the movement of people, capital, technology,
institutions, ideas, ideological systems, and knowledge, is the most visible face of
globalization. Mobility is not only central to the process of globalization, given
its historical depth, but it also helps mark the differences in the various histori-
cal phases of globalization. Enhancing mobility has become a central feature of

62 Globalization: Past

contemporary globalization. At the end of the twentieth century, close to 2.6 billion
people traveled by the world’s airlines each year (Hobsbawm, 2007: 86). In 2019, a
year before the outbreak of COVID-19, the International Civil Aviation Organiza-
tion’s preliminary compilation of annual global statistics put the total number of
passengers carried on scheduled services at 4.5 billion (ICAO, 2019). Asia and the
Pacific accounted for 34.7 percent of the traffic, while Europe and North America
accounted for 26.8 percent and 22.2 percent of the traffic, respectively.

An increase in air travel is a good indicator of both intra- and international
mobility. Businesspeople, workers, students, and people traveling back and
forth to meet family members in different parts of the globe constitute most
of the air passengers. The mobility of students has been an interesting trend
in recent decades, especially since the economic reforms in China. An increase in
the number of tourists is another aspect of enhanced mobility. In 1896, author
Mark Twain landed in Bombay (now renamed Mumbai) and stayed at Watson’s,
Bombay’s leading hotel at the time. His sojourn was part of his global travels that
took him to Europe, India, and Australia (Twain, 1897). Tourism can be traced
back to the seventeenth and eighteenth centuries, when it was a privilege limited
to the nobles, when inns and hotels were being established in various European
cities. As the Industrial Revolution unfolded, tourism proper began in the
nineteenth century. The first travel agency, Thomas Cook & Son, dates back to
the nineteenth century, offering excursions and holidays. Tourism flourished in the
twentieth century with the expansion of railway lines, the advent of the automo-
bile, and later planes. Being able to travel, particularly for nonwork reasons, was
only available to a narrow elite and was itself a mark of status (Urry, 1990: 24). The
frequency of foreign travel prompted Omhae to declare the idea of a “borderless
world” (Ohmae, 1989). In 1988, nearly 90 percent of all Japanese honeymooners
went abroad.

The World Tourism Organization (UNWTO) estimates that internationally,
there were just 25 million tourist arrivals in 1950. Sixty-eight years later, this num-
ber has increased to 1.4 billion international arrivals per year. This is a fifty-six-fold
increase. Europe accounts for over 51 percent of all the tourists in 2018 (Our
World in Data, n.d.). Tourism and travel’s direct contribution to GDP globally was
approximately US$9.1 trillion in 2019, which dropped to US$4.7 trillion in 2020
due to the COVID-19 pandemic (Statista, 2021).

The precursor of tourism is a pilgrimage where people motivated by religious
belief undertook visits to holy sites at a regular interval. Pilgrimage provides
continuity in the historicity of globalization. One of the world’s largest religious
gatherings is at Hajj, where about 2.5 million Muslim pilgrims took part in 2019
(Government of Saudi Arabia, 2019). The Kumbh Mela, considered the most pop-
ulous pilgrimage, attracts tens of millions of Hindu devotees to dip in the Ganges,
defying COVID-19 fears (Srinivas, 2021).

Mobility and Globalization 63

MOBILIT Y AS MIGR ATION

Migration, which is simply the physical movement of people from one place to
another, has become the most visible face of globalization today. One cannot ignore
the fact that an increasing number of residents of almost every country today are
foreign-born, revealing global mobility. Although some writers define the present
phase (twentieth and twenty-first century) of globalization as the “age of migra-
tion” (Castles & Miller, 1993), historical human migration—both involuntary and
voluntary—marked the salience of migration in earlier phases of history just as
well. Pitrim Sorokin, a Harvard sociologist and an immigrant from what was then
the Soviet Union, used the term mobility in the 1940s and differentiated between
horizontal mobility, which implied migration, and vertical, or social mobility.

Although human migration is not new, one of the most interesting features of
twenty-first-century globalization has been an increase in mobility, where more
people tend to move more frequently from place to place. Mobile people include
tourists, commuters, and migrants. We differentiate between migrants and other
mobile people on the grounds that migrants live for a certain period in a destina-
tion country for work, education, or business or to join family members; are subject
to the rules, regulations, and customs of the receiving country; and are required to
make some degree of cultural accommodation. All modern societies allow free
movement of citizens within their national borders and the national laws ensure
that they enjoy the freedom to travel. An exception has been the hukou system in
China, which was an attempt to regulate and restrict internal migration. In recent
years, reforms have taken place in the hukou system to facilitate industrial produc-
tion in the coastal cities of the South and Southeast (Zhao & Fu, 2010). Although
the system was officially phased out in 2014, this practice is unlikely to be phased
out anytime soon (Goodburn, 2014). In premodern Europe, there were restric-
tions on rural people’s ability to move into the cities without work. Vagabondage
was a punishable offense in early modern Europe (Kamen, 1986; Perry, 2002).

Migration raises interesting issues about the relationship between market
forces and the authority of the state. It is the classic law of supply and demand that
dictates the movement of people across national boundaries. However, the state
in the receiving countries, in aiding the market forces, usually complies with the
dominant economic forces, but the state is responsible for its citizens. The sending
states also play a role in sometimes promoting the interests of migrant workers by
providing all kinds of assistance or blocking their movement if there are possibili-
ties of workers falling into exploitative situations overseas.

Although there has been a significant increase in the migrant population in
recent decades, earlier centuries saw greater mobility of people in terms of per-
centage of total population and degree of freedom—if not ease—of movement.
The last decade of the twentieth century was declared the “age of migration” by

64 Globalization: Past

Stephen Castles and Mark J. Miller (2003). Yet the nineteenth century allowed
easier migration flows. In the euphoria of discussions on globalization, writers
like Kenichi Ohmae (1989), a business thought leader, announced the coming of a
“borderless world.” It is ironic that when celebratory remarks about the so-called
borderless world were being made, walls were being erected in many parts of the
world to restrict the movement of people. Contrary to popular perception, in
the twentieth century, the world has become more bordered than at any other time
in the past. We live in a state-centered world. The contradiction of globalization
is best revealed in the free flow of capital and the restricted movement of labor.

The uneven forces of globalization have made the state more central. The
relationship between globalization and the state has taken a new turn, making
the state more powerful insofar as border control is concerned. The power of the
state is often felt in its ability to control and regulate the movement of people,
goods, technology, and intellectual property. In the discourses on globaliza-
tion, several ways of conceptualizing the processes of globalization are available.
One of the conceptualizations views globalization as a series of flows: flows
of capital, technology, ideas, and population. This view is most relevant in the
discussion of mobility. Since the emergence of state systems, migrants can be
viewed as either internal or international. Most migrants in the present world
are internal migrants, which reflects uneven development within the country, as
international migration is rooted in uneven global development. Both internal
and international migrants constitute close to one billion migrant workers in the
world. Here, migrant is defined—minimally—as a person who lives in a place other
than where she or he was born. In the world of nation-states, some people move
within the country from rural to urban areas or from small towns to metropolitan
cities in search of work, education, career enhancement, better living conditions,
and so on. They constitute around 740 million people (UNDP, 2009, quoted in
IOM, 2015).

The global estimate of international migrants stood around 286 million in 2022,
which included 32.5 million refugees in mid-2022 (World Bank, 2022). Overall,
the estimated number of international migrants has increased over the past five
decades. One hundred ninety-one million people, or 3 percent of the world’s pop-
ulation, lived outside their country of birth in 2005, according to the UN Depart-
ment of Economic and Social Affairs. The equivalent figure in 1960 amounted to
75 million people or 2.5 percent of the world population. Almost one in every
ten people living in more developed regions is a migrant. The total estimated 281
million people living in countries other than their countries of birth in 2020 was
128 million more than in 1990 and over three times the estimated number in 1970
(IOM, 2022).

China accounts for the highest number of internal migrants in the world.
According to the 2020 census, the floating, or migrant, population—defined as
those without local household registration (hukou)—has increased to 376 million,

Mobility and Globalization 65

up from 155 million in 2010 (Chan, 2021). Uneven industrialization is the cause of
both internal and international migration. In China, internal migration has health
impacts, especially on young migrants aged 16–35 (Lu, Kandilov, & Zhu, 2020).

What is unique about migration in the age of globalization is that more and more
people are moving to and from more and more countries, and their movements
are closely documented and surveyed. Controversies over the undocumented
movement of people are an affirmation of the importance of documentation and
surveillance. Migration is the result of a complex set of social, economic, political,
and cultural processes. There are several types of migration: voluntary, involun-
tary or forced, and economic or political. Often, a combination of factors drives
people to migrate. While migrants exercise their volition in choosing to move, for
refugees and internally displaced people, such choices do not exist. COVID-19
restricted the number of people on the move but did not fully restrict mobility.
Displacement continued to occur and grow, with 1 in 95 people displaced at the
end of 2020, up from 1 in 159 in 2010 (UNHCR, 2022).

CAUSES OF MIGR ATION

John Maynard Keynes, the famous English economist, said that “migration is the
first act against poverty.” According to a United Nations report, three D’s account
for the majority of migration today: demography, development, and democracy.
People tend to move out of so-called overpopulated countries to less populated
countries, from less developed to more developed countries, and from authoritar-
ian to democratic countries. Less populated but rich countries such as Canada
and Australia remain popular destinations for migrants. While some migrants
move permanently, the oil-rich Gulf countries remain destinations for temporary
migrant workers. The United Arab Emirates has the highest proportion of tempo-
rary migrants, who constitute over 88 percent of the population.

According to Adam McKeown (2004), world migration reached new peaks in
the 1920s, and the immigration restrictions of the 1920s were also part of a much
longer trend of regulation, border control, and nationalism that had grown con-
currently with migration since the middle of the nineteenth century. From 1846
to 1940, there were three main circuits of long-distance migration. During this
century of migration, 55–58 million Europeans and 2.5 million from India, China,
Japan, and Africa migrated or were taken to the Americas. During the same
period, the other main destination was Southeast Asia, the Indian Ocean Rim, and
the South Pacific, where 48–52 million Chinese from China and Indians moved.

In the twentieth century, alongside forced involuntary migration caused by
war, voluntary migration grew enormously. At the beginning of the twenty-first
century, owing to a combination of factors such as relatively cheaper air travel,
the expansion of job opportunities, falling birth rates in some countries, and the
availability of surplus populations in other countries, more and more people were

66 Globalization: Past

becoming increasingly mobile. There are international migrants and refugees as
well as internally displaced persons. The latter group is often a product of civil war
or social unrest.

MIGR ANT L AB OR

Political-economic globalization can be traced back to the slave trade in the six-
teenth century with the forced movement of African slave labor to the Caribbean
and North American plantations. Such forced and exploitative labor transfers are
still practiced in various parts of the world. Colonialism and the European land
grab marginalized the poor of many colonies who were eventually driven by the
economic necessity to become indentured laborers. In the nineteenth century,
the migration of Europeans to various parts of the world created white-settler soci-
eties such as Australia, New Zealand, and South Africa. Here, the migration issue
was connected with racism and the marginalization of various Indigenous com-
munities. Many of those problems remain, especially around the status of aborigi-
nal people and their relationship to the land (Turner & Khondker, 2010: 107).

Mobile people are extremely heterogenous. They can be migrant workers, tour-
ists, international students, or refugees. In addition to migrants and refugees, there
were 48 million internally displaced persons in 2020 (UNHCR, 2021). One of the
important trends in migration is an increase in South-South migration (Hujo
& Piper, 2010; IOM, 2020). About 60 percent of all migrants are now found in
the world’s most prosperous countries and around 40 percent in the developing
regions (GCIM, 2005) Migrants to industrially developed countries often seek
permanent status and citizenship. Because of the high mobility of people across
nations, many countries now accept dual citizenship. Professionals in certain spe-
cialized fields are very much in demand, and some countries offer incentives to
attract these specialists. Indian software engineers can be seen in many different
countries. Some countries, such as the Philippines, have taken a proactive atti-
tude toward the export of labor or an out-migration strategy since the mid-1970s
under President Ferdinand Marcos. In the first decade of the twenty-first century,
Filipina women were employed as domestic workers in over 130 countries in the
world (Parrenas, 2001: 1). It is estimated that there are eight million workers—both
as domestic workers and in other trades—who are known as “overseas foreign
workers” (OFW) from the Philippines who play a vital role in the economy of the
country. The income they earn overseas helps sustain their families left behind in
their homeland.

The World Bank estimates that worldwide remittances will reach $689 billion
in 2021, with remittances to the developing world reaching $529 billion. India in
2021 again topped the list with the US$87 billion (World Bank, 2021). The idea
that mobility begets possibility (Giaveanu, 2020) is often realized with migrants
if this process is properly administered. However, often, poorly administered and

Mobility and Globalization 67

laissez-faire migration leads to a new category of vulnerable people known as
“irregular” or “undocumented workers.” Preying on their vulnerable existence and
exploiting their ignorance, human traffickers lead people on uncertain journeys.
According to the UN Refugee Agency, in 2021, 3,231 people died while trying to
cross the Mediterranean Sea (UNHCR, 2022).

HUMAN TR AFFICKING AND SL AVERY

Some people choose to leave home in search of better jobs and security elsewhere,
and such economic migration is often in response to the push and pull of the
forces of globalization. They leave their home to avoid poverty and repression,
while others are allured by the prospect of a better life in other places. Some people
make choices on their own, but others are forced to migrate because of a host
of factors ranging from economic deprivation to political repression to outright
expulsions (or so-called compulsory repatriation). An extreme form of forced
migration is human trafficking; children and women are often kidnapped, stolen,
and sold into slavery.

Human trafficking has been identified as a new form of slavery in the pres-
ent world, exposing some of the ill effects of uncontrolled globalization. The
International Organization for Migration calls human trafficking the “most
menacing form of irregular migration due to its ever-increasing scale and com-
plexity, involving, as it does, arms, drugs, and prostitution.” In this shady world,
precise figures are difficult to come by. According to the International Labour
Organization (ILO), there were 24.9 million victims of human trafficking around
the world in 2016, of whom 10.9 million were women and 3.3 million were chil-
dren. ILO also estimated there were 4.8 million sex-trafficking victims subjected to
commercial sexual exploitation around the world in 2016. Ninety-nine percent of
the victims were women, while 3.8 million were adults and 1 million were children
(Ecker, 2022).

In the Trafficking in Persons Report published by the State Department of the
U.S. government, “Trafficking in persons is a modern-day form of slavery, a new
type of global slave trade.” Perpetrators prey on the most vulnerable among us,
primarily women and children, for profit and gain. Female victims continue to be
overrepresented in trafficking in persons. “In 2018, for every 10 victims detected
globally, about five were adult women and two were girls. About one-third of the
overall detected victims were children, both girls, and boys, while 20 percent were
adult men” (UNODC, 2021: 31).

Slavery was officially abolished in 1833 in the British Empire, in 1865 in the
United States by the Thirteenth Amendment, and in 1886 in Cuba, but the practice
goes on in our society under a different name. Human trafficking is the twenty-
first-century version of slavery. The extent of slavery in the contemporary world
is extensively documented, for example, in the works of Kevin Bales (1999, 2005).

68 Globalization: Past

According to Bales, since its general abolition in the late nineteenth century,
slavery has slipped easily into the shadow economy. Slavery may be defined as the
complete control of a person for economic exploitation by violence or the threat of
violence (Bales, 2000: 461). With slavery, the person becomes a mere commodity
or thing.

According to another authority, “Slavery exists today on an unprecedented
scale.” In Africa, tens of thousands are chattel slaves, seized in war or tucked away
for generations. Across Europe, Asia, and the Americas, traffickers have forced
as many as two million into prostitution or labor. In South Asia, which has the
highest concentration of slaves on the planet, nearly ten million languish in
bondage, unable to leave their captors until they pay off their “debts” (Skinner,
2008: 64). Bales estimates that there are twenty-seven million slaves in the world
today, of whom fifteen to twenty million are in India, Pakistan, and Nepal (Bales,
2005). The positive view of the free movement of labor in a global economy will
not want to deal with these issues and will instead attempt to focus on the mate-
rial improvement of people in a deregulated global economy, where people have
in principle the freedom to cross borders at ease in a borderless world. In general
terms, the science of economics does not deal effectively with black markets and
criminal activity in the marketplace, concentrating instead on the formal mar-
ket in which goods and services are transacted according to formal, public rules.
Consequently, academic economics dealing with formal and legal exchanges does
not normally include criminal activity, which may keep a large section of the
community in employment and the gross domestic product. Slavery, trafficking,
and the informal, undocumented movement of people often remain unnoticed
and unaccounted for. These people remain permanently marginalized.

BR AIN DR AIN TO THE BR AIN IN CIRCUL ATION

In the 1970s, “brain drain” was a popular slogan, and it was often regarded as one
cause of the poor economic performance of those countries that were exporting
their most talented doctors, engineers, and scientists to the rich, developed coun-
tries. Indeed, many talented young men and women migrated from the periphery
to the core of the world economy for better educational and career opportuni-
ties. Universities in rich countries, such as the United States, were magnets for
attracting foreign-born talent. This process was reversed in the 1990s and in the
first decade of the twenty-first century. The notion of “brain drain” was supple-
mented by “brain gain,” and Analee Saxsenian (2005) introduced the concept of
“brain in circulation.” Countries such as India produced many talented men and
women in various scientific and engineering fields, only to be absorbed by the
sluggish Indian economy of the 1960s, 1970s, and 1980s, which led to a serious
outpouring of the Indian creative class to North America and elsewhere. In the
1990s and especially in the first decade of the twenty-first century, many such

Mobility and Globalization 69

talented Indians experienced in high-tech industries began to return to their own
countries, which by then created conditions that would allow them to pursue their
professional careers.

Brain drain remains a global problem from which the rich countries in the
Global North benefit at the expense of the poor countries in the South. Reasons
for brain drain are mostly economic: poor working conditions in the origin coun-
tries and the attraction of a better quality of personal and professional life in the
destination countries. In the second decade of the twenty-first century, foreign-
born people represented 10 percent of the workforce in OECD countries, a twofold
increase since 1960 and a twofold increase since 1990. High-skill migration grew
even faster, with a twofold increase during the 1990s alone (Alesina, Harnoss, &
Rapoport, 2016: 102).

The idea that national borders have become more porous for the creative
class was popularized by Richard Florida (2002). The footloose nature of the cre-
ative classes and the frequent movement of professionals have led some writers
to develop the notion of “flexible citizenship” (Ong, 1999). Although Aihwa Ong
developed the notion to describe the global flexibility of the Chinese business
class, the phrase has become popular in the literature on global migration. While it
may be the case for a small number of skilled professionals to move around at ease,
often having more than one passport, for most working-class migrants, national
boundaries remain a hard reality.

Some countries, such as Taiwan and Israel, have had effective incentive poli-
cies to reverse the outward trend of the migration of talented people. A reverse
brain drain ensued. In the case of Taiwan, many Chinese from overseas—mainly
from the United States—returned not only with swathes of cash, but many of
them brought valuable scientific and technological knowledge with them, which
assisted Taiwan’s remarkable economic development. India has also been suc-
cessful in capitalizing on these trends, and many Indians with years of overseas
experience are now returning to India. Several Indian professionals in the infor-
mation and technology industries have left Silicon Valley to return to Bangalore,
the Indian information hub. Many Indian professionals are now returning to
Bangalore and other economic hothouses in India with rich experience in tech-
nology, education, and finance. This rapid economic growth has also given India
a prominent international status. Following trade liberalization and the open-
ing of the economy to investments from outside, China has received huge funds
and expertise from overseas Chinese communities. To attract people of Chinese
origin, China maintained an ethnicity-based in-migration policy. In the rush
to capitalist development, this economic stimulus was timely and important. In
India too, NRIs, which meant “nonresident Indians,” played an important role
in India’s high-tech development.

Diasporic communities rarely sever their links with the country of their origin,
although, in many instances, these links may stretch over several generations. The

70 Globalization: Past

close community bonds of the diasporic groups sometimes have unfavorable con-
sequences. The radicalization of young people often takes place in the diasporic
environment of alienation and strangeness. Modern technology has played an
important role in maintaining links between families and communities in the
world of the diaspora. Unlike the diasporic situation of the past, migrants main-
tain a close link with their home countries, thus rendering the meaning of “home”
tenuous. Diasporic Islamic groups often display a heightened sense of religiosity
bordering on radicalism. Research has shown how some diasporic communities
come under the influence of radicalized religion, thereby becoming the source of
religious extremism in their countries of origin (Kibria, 2008).

MOBILIT Y OF IDEAS, IDEOLO GIES, AND KNOWLED GE

Historically, transnational intellectuals have been purveyors of ideas, ideologies,
and knowledge. Scientific knowledge in the present phase of globalization tends to
travel from the Global North to the Global South, whereas in the last phase of the
first millennium mathematical, philosophical, and physiological science knowl-
edge traveled from China, India, and the Middle East to the West. Ideologies such
as Marxism evolved in the West and became a dominant force in the so-called
East as intellectuals-turned-leaders from Lenin and Trotsky in the former USSR
to Chou En-Lai of China and others were schooled in Marxist ideology in the
capital cities of the West. Later Asian nationalist leaders such as Mahatma Gan-
dhi and Jawaharlal Nehru, their African counterparts such as Kwame Nkrumah
and Jomo Kenyatta, and Ho Chi Minh of Vietnam were schooled in the ideas of
nationalism, democracy, and socialism in the universities in Britain and France
who, in turn, purveyed these ideas in their native countries. As the knowledge
economy becomes the dominant mode of the global economy, globalization of
knowledge becomes imperative. Universities have freely played a critical role in
the dissemination of intellectual traditions, humanities, and social sciences. How-
ever, contradictions arise when it comes to sharing knowledge of sciences and
technologies with immediate relevance to profit-making. Intellectual property
laws have been designed to protect patent rights, but they can be an obstacle to
knowledge sharing.

During the COVID-19 crisis of 2020–21, debates ensued when strong argu-
ments were provided for and against relaxing intellectual property laws so that
vaccines could become the global public good (Rake, 2021). The World Trade
Organization on the one hand and South Africa and India on the other demanded
relaxation of intellectual property laws for three years on humanitarian grounds so
that vaccines could be produced during this time in multiple locations, especially
in the vaccine-deficit part of the Global South, an effort that was opposed by the
Big Pharma industries (Jecker & Aturie, 2021).

Mobility and Globalization 71

C ONCLUSION

Globalization processes are intrinsically uneven in both their form and their
effect. One of the challenges of the globalized world is that economy and poli-
tics are driven by divergent interests or logics. Modern economies require a flex-
ible labor market in which workers can move rapidly and easily between different
work sites depending on the local demand for labor inputs. Political imperatives
and the state’s need for sovereignty and security outweigh the economic needs for
labor mobility. These controls inevitably involve greater negotiation and manage-
ment of migration, and the result is the seesaw of labor mobility and immobility.
Such contradictions are also present in the dilemma of free flow of scientific and
technological knowledge dissemination and protection of information and copy-
rights via intellectual property laws. Setting aside such contradictions, mobility, as
such—especially in knowledge, ideas, and technology—remains an intrinsic part
and a defining feature of the global age, which is likely to be augmented by the
AI and the new generative technology in years to come.

REFERENCES

Alsenia, A., J. Harnoss, & H. Rapoport (2016). “Birthplace Diversity and Economic Prosperity.” Journal of
Economic Growth 21 (2): 101–38. https://doi.org/10.1016/j.jegrowth.2016.02.001.10.1007/s10887-016
-9127-6.

Bales, K. (1999). Disposable People: New Slavery in the Global Economy. Berkeley: University of
California Press.

———. (2000). “Expendable People: Slavery in the Age of Globalization.” Journal of International
Affairs 53 (2): 461–84. https://www.jstor.org/stable/24357761.

———. (2005) Understanding Global Slavery: A Reader. Berkeley: University of California Press.
Bhambra, G. K. (2014). Connected Sociologies. London: Bloomsbury.
Castles, S., & M. Miller (2003). The Age of Migration: International Population Movements in the

Modern World. New York: Guilford Publications.
Chan, K. W. (2021). “China’s Hukou Reform Remains a Major Challenge to Domestic Migrants in

Cities.” World Bank Blogs, December 17. https://blogs.worldbank.org/peoplemove/chinas-hukou
-reform-remains-major-challenge-domestic-migrants-cities.

Ecker, E. (2022). “Breaking Down Global Estimates of Human Trafficking: Human Trafficking Aware-
ness Month 2022.” Human Trafficking Institute, January 12. https://traffickinginstitute.org/breaking
-down-global-estimates-of-human-trafficking-human-trafficking-awareness-month-2022/.

Florida, R. (2002). The Rise of the Creative Class. New York: Basic Books.
Gänger, S. (2017). “Circulation: Reflections on Circularity, Entities, and Liquidity in the Language of

Global History.” Journal of Global History 12 (3): 303–18. https://doi.org/10.1017/s174002281700016x.
GCIM (Global Commission on International Migration) (2005). “Migration in an Interconnected

World: New Directions for Action.” www.iom.int/global-commission-international-migration.
Goodburn, C. (2014). “The End of the Hukou System? Not Yet.” Chinese Policy Institute, Policy Paper

no. 2.
Government of Saudi Arabia (2019). Hajj Statistics, 2019–1440. www.stats.gov.sa/sites/default

/files/haj_40_en .
Hobsbawm, E. (2007) Globalisation, Democracy, and Terrorism. London: Little Brown.

https://doi.org/10.1016/j.jegrowth.2016.02.001.10.1007/s10887-016-9127-6

https://www.jstor.org/stable/24357761

https://blogs.worldbank.org/peoplemove/chinas-hukou-reform-remains-major-challenge-domestic-migrants-cities

https://traffickinginstitute.org/breaking-down-global-estimates-of-human-trafficking-human-trafficking-awareness-month-2022/

https://doi.org/10.1017/s174002281700016x

http://www.iom.int/global-commission-international-migration

http://www.stats.gov.sa/sites/default/files/haj_40_en

72 Globalization: Past

Hochstadt, S. (1999). Mobility and Modernity: Migration in Germany, 1820–1889. Ann Arbor: University
of Michigan Press.

Hujo, K., & N. Piper, eds. (2010). South–South Migration: Implications for Social Policy and Develop-
ment. London: Palgrave Macmillan.

ICAO (2020). “The World of Air Transport in 2019” (annual report). www.icao.int/annual-report
-2019/Pages/the-world-of-air-transport-in-2019.aspx, accessed April 12, 2022.

IOM (International Organization for Migration) (2015). World Migration Report 2015. https://world
migrationreport.iom.int/what-we-do/world-migration-report-2015.

———. (2020). World Migration Report 2020. https://publications.iom.int/system/files/pdf/wmr_2020 .
———. (2022). World Migration Report 2022. https://publications.iom.int/books/world-migration-report

-2022.
Jecker, N. S., & C. A. Atuire (2021). “What’s Yours Is Ours: Waiving Intellectual Property Protections for

COVID-19 Vaccines.” Journal of Medical Ethics 47 (9): 595–98. https://doi.org/10.1136/medethics
-2021-107555.

Kamen, H. (1986). European Society, 1500–1700. London: Hutchinson.
Kibria, N. (2008). “The ‘New Islam’ and Bangladeshi Youth in Britain and the US.” Ethnic and Racial

Studies 31 (2): 243–66.
Lu, H., I. T. Kandilov, & R. Zhu (2020). “The Impact of Internal Migration on the Health of Rural Mi-

grants: Evidence from Longitudinal Data in China.” Journal of Development Studies 56 (4): 840–55.
https://doi.org/10.1080/00220388.2019.1590553.

Lucassen, L., & J. Lucassen (2009). “The Mobility Transition Revisited, 1500–1900: What Europe
Can Teach Us about Global History.” Journal of Global History 4 (4): 347–77. https://doi:10.1017
/S174002280999012X.

Manning, P. (2005). Migration in World History. London: Routledge.
McKeown, A. (2004). “Global Migration, 1846–1940.” Journal of World History 15 (2): 155–89.
Ohmae, K. (1989). “Managing in a Borderless World.” Harvard Business Review, May–June.
Ong, A. (1999). Flexible Citizenship: The Cultural Logic of Transnationality. Durham, NC: Duke

University Press.
Our World in Data (n.d.). “Tourism.” https://ourworldindata.org/tourism, accessed June 12, 2023.
Parrenas, R. S. (2001). Servants of Globalization. Stanford, CA: Stanford University Press.
Perry, M. (2002). Marxism and History. London: Palgrave Macmillan.
Rake, B. (2022). “Waiving Intellectual Property Rights: Boom or Bust for Medical Innovation.” Drug

Discovery Today 27 (2): 384–89. https://doi.org/10.1016/j.drudis.2021.10.015.
Robertson, R. (1992). Globalization: Social Theory and Global Culture. London: Sage.
Saxsenian, A. (2005). “From Brain Drain to Brain Circulation: Transnational Communities and

Regional Upgrading in India and China.” Studies in Comparative and International Development
40 (2): 35–61.

Schafer, A., & D. G. Victor (2000). “The Future Mobility of the World Population.” Transportation
Research Part A: Policy and Practice 34 (3): 171–205.

Sheller, M., & J. Urry (2006). “The New Mobilities Paradigm.” Environment and Planning A: Economy
and Space 38: 207–26.

Skinner, E. B. (2008). A Crime So Monstrous: Face-to-Face with Modern-Day Slavery. New York: Free
Press.

Srinivas, T. (2021). “India Prepares for Kumbh Mela, World’s Largest Religious Gathering, amid
COVID-19 Fears.” The Conversation, April 8. https://theconversation.com/india-prepares-for-kumbh
-mela-worlds-largest-religious-gathering-amid-covid-19-fears-158364.

Statista (2021). “Total Contribution of Travel and Tourism to Gross Domestic Product (GDP) World-
wide from 2019 to 2021.” www.statista.com/statistics/233223/travel-and-tourism—total-economic
-contribution-worldwide/.

Steger, M. B. (2013). Globalization: A Very Short Introduction. Oxford: Oxford University Press.

http://www.icao.int/annual-report-2019/Pages/the-world-of-air-transport-in-2019.aspx

https://worldmigrationreport.iom.int/what-we-do/world-migration-report-2015

https://publications.iom.int/system/files/pdf/wmr_2020

https://publications.iom.int/books/world-migration-report-2022

https://doi.org/10.1136/medethics-2021-107555

https://doi.org/10.1080/00220388.2019.1590553

https://doi:10.1017/S174002280999012X

https://ourworldindata.org/tourism

https://doi.org/10.1016/j.drudis.2021.10.015

https://theconversation.com/india-prepares-for-kumbh-mela-worlds-largest-religious-gathering-amid-covid-19-fears-158364

www.statista.com/statistics/233223/travel-and-tourism—total-economic-contribution-worldwide

Mobility and Globalization 73

Subrahmanyam, S. (1997). “Connected Histories: Notes towards a Reconfiguration of Early Eurasian
History.” Modern Asian Studies 31 (3): 735–62.

———. (2022). Connected History: Essays and Arguments. London: Verso.
Tilly, C. (1978). “Migration in Modern European History.” In W. H. McNeill and B. Adams, eds., Human

Migrations: Patterns and Policies. Bloomington: University of Indiana Press.
Torpey J. (1998). “Coming and Going.” Sociological Theory 16 (3): 239–59.
———. (2000). The Invention of the Passport: Surveillance, Citizenship, and the State. Cambridge:

Cambridge University Press.
Turner, B., & H. Khondker (2010). Globalization: East and West. London: Sage.
Twain, M. (1897). Following the Equator. New York: American Publishing Company.
UNDOC (2021). Global Report on Trafficking in Persons 2020. New York: United Nations. www.unodc

.org/documents/data-and-analysis/tip/2021/GLOTiP_2020_15jan_web .
UNDP (2009). Human Development Report 2009: Overcoming Barriers: Human Ability and Develop-

ment. New York: United Nations Development Program.
UNHCR (2022). Global Report on Law and Policy on Internal Displacement: Implementing National Re-

sponsibility. UN High Commissioner for Refugees. www.globalprotectioncluster.org/sites/default
/files/2023-03/UNHCR%20-%20Global%20Report%20on%20Law%20and%20Policy%20on%20
Internal%20Displacement%20Hi-Res%20Version%20%281%29 .

———. (2022). Global Trends: Forced Displacement in 2020. UN High Commissioner for Refugees. www
.unhcr.org/flagship-reports/globaltrends/.

———. (2022). “ UNHCR Data Visualization on Mediterranean Crossings Charts Rising Death Toll
and Tragedy at Sea.” UN High Commissioner for Refugees. www.unhcr.org/news/briefing-notes
/unhcr-data-visualization-mediterranean-crossings-charts-rising-death-toll-and.

Urry, J. (1999). “The ‘Consumption’ of Tourism.” Sociology 24 (1): 23–35.
———. (2000). Sociology beyond Societies: Mobilities for the Twenty-first Century. London: Routledge.
World Bank (2016). Migration and Remittances Factbook (2016), 3rd ed. www.worldbank.org/en/research

/brief/migration-and-remittances.
———. (2022). Migration. www.worldbank.org/en/topic/migration/overview.
Zelinsky, W. (1971). “The Hypothesis of the Mobility Transition.” Geographical Review 61: 219–49.
Zhao, L., & C. R. Fu (2010). China’s Hukou Reform: The Guangdong and Shanghai Cases. Singapore: East

Asian Institute, National University of Singapore.

http://www.unodc.org/documents/data-and-analysis/tip/2021/GLOTiP_2020_15jan_web

http://www.globalprotectioncluster.org/sites/default/files/2023-03/UNHCR%20-%20Global%20Report%20on%20Law%20and%20Policy%20on%20Internal%20Displacement%20Hi-Res%20Version%20%281%29

http://www.unhcr.org/flagship-reports/globaltrends/

http://www.unhcr.org/news/briefing-notes/unhcr-data-visualization-mediterranean-crossings-charts-rising-death-toll-and

http://www.worldbank.org/en/research/brief/migration-and-remittances

http://www.worldbank.org/en/topic/migration/overview

American Political Science Review Vol. 97, No. 4 November 2003

Security and the Political Economy of International Migration
CHRISTOPHER RUDOLPH University of California, Los Angeles

How does migration affect the security of advanced industrial states, and how does the security
environment shape the way states deal with international migration? Migration rests at the nexus
of three dimensions of security, including geopolitical interests, material production, and internal

security. I argue that migration policy is an integral instrument of state grand strategy in this context, and
that examining levels of threat on each facet of security at a given point in time can largely explain variation
in policy. I test a series of hypotheses drawn from this security framework using a case-study method that
examines policy development in four advanced industrial states, including the United States, Germany,
France, and Great Britain in the period 1945–present.

Does the security environment affect immigra-
tion and border policies among advanced in-
dustrial states? The tragic events of September

11, 2001, made the connection between international
migration and security quite clear: all 19 of the terror-
ists exploited loopholes in existing laws to infiltrate the
United States. Less recognized is the fact that inter-
national migration has had significant implications for
security long before 9/11—a process inherently linked
to the multiple facets of security in an age of rapid glob-
alization. Understanding the politics of international
migration and border control policies not only is impor-
tant in terms of national security and economic growth
among advanced industrial countries but reveals chang-
ing conceptions of sovereignty and the role of the state
in policy development.

TWO WORLDS

In an important sense, 9/11 has established two distinct
worlds. Prior to 9/11, conceptions of the “new world
order” were based on an optimistic view of the triumph
of political and economic liberalism over Communism
at the close of the Cold War. This new world order
was envisioned as the perfection of a system based
on trading states, where power is increasingly based
on Ricardian notions of comparative advantage, factor
mobility, and free trade (Rosecrance 1986). As gains
from trade and interdependence increase, as they have
under the past half-century of American hegemony, the
use of a military–territorial strategy is a less appealing
means of maximizing state power, especially since war
has become increasingly costly, complicated, and de-
structive. In place of military conquest, the trading state
logic suggests that liberal trade policies and laissez-faire
treatment of international factor flows move economies

Christopher Rudolph is Lecturer and Faculty Fellow Researcher, De-
partment of Political Science, University of California, Los Angeles,
4262 Bunche Hall, 405 Hilgard Avenue, Los Angeles, CA 90095-1472
(crudolph@polisci.ucla.edu).

Thanks go to James Hollifield, David Lake, Marc Rosenblum, and
Arthur Stein for comments and suggestions made on early drafts. The
financial support of the Center for International Studies, University
of Southern California, is gratefully acknowledged. Previous versions
of this paper were presented at the Institute of Governmental Stud-
ies, University of California, Berkeley, 26 November 2001, and at
the Annual Meeting of the International Studies Association, New
Orleans, 25 March 2002.

toward a Pareto-optimal frontier, one that will create
a “rising tide to lift all boats” (Bhagwati 1998). Con-
siderable evidence supports the argument that trad-
ing state globalization has emerged as a global norm
and as a widely accepted basis of state grand strategy
since World War II. Since the 1940s, successive rounds
of the GATT (now the WTO) have resulted in con-
sistently lower tariff rates that have helped stimulate
world trade. From 1980 to 1998, the rate of growth
in world trade ranged from 4.2% to 10.3%, and from
1990 to 1999 it grew at over three times the rate of
global output (World Bank 1998; World Trade Orga-
nization 2000). Moreover, financial transactions, once
an adjunct of trade, have grown even faster and now
tower over trade flows by a ratio of 50:1.

Given the gains available through free market poli-
cies, it would seem logical that states would seek eco-
nomic maximization through labor mobility as well
as trade and capital flows (Ghosh 2000). There has
been growing appreciation of the fact that economic
transformations resulting from the IT revolution and
changes in the global division of labor have created a
tremendous surge in demand for services and skilled
labor, and knowledge-based human capital is emerg-
ing as an ever more important component of state
power (Cornelius, Espenshade, and Salehyan 2001;
Mittelman 2000). In addition, as the domestic labor
force in advanced industrial economies shifts toward
higher-skill professions, demand for low-skilled labor
in industry, agriculture, and domestic services has risen
concomitantly and is a crucial component to a well-
functioning economy and society (Cornelius 1998a).
Economic historians examining trade and migration
during the nineteenth century have offered compelling
evidence that migration, both high- and low-skilled,
provided higher gains than trade alone and have also
suggested that migration may be a necessary condition
to receive the overall economic gains of openness to
trade and capital flows, due to total specialization or
locational economies of scale (Hatton and Williamson
1998; O’Rourke and Williamson 1999). A neoliberal
approach to global economic efficiency would call for
unrestricted migration, allowing labor to move freely to
the country where it earns the highest return and where
its marginal product is the highest (Chang 1998). More-
over, in his testimony before a special Senate com-
mittee on aging U.S. Federal Reserve Board chairman
Alan Greenspan pointed out that generally low fertility

603

ht
tp

s:
//

do
i.o

rg
/1

0.
10

17
/S

00
03

05
54

03
00

09
0X

P
ub

lis
he

d
on

lin
e

by
C

am
br

id
ge

U
ni

ve
rs

ity
P

re
ss

https://doi.org/10.1017/S000305540300090X

Security and International Migration November 2003

rates and the accompanying aging of the population in
many advanced industrial welfare states, including the
United States, have called into question the viability
of social security programs without a major increase in
immigration.

Openness toward migration flows makes perfect
sense given these links among migration, the labor
demands of the new global economy, trade, and the
pursuit of economic maximization. Surprisingly, how-
ever, while the contemporary logic of trade and inter-
national finance has been one of generally increasing
degrees of openness, the logic of migration policies has
increasingly been one of closure since the mid-1960s
(Cornelius, Martin, and Hollifield, 1994). In contrast,
the events of 9/11 would seem to have reversed the
prevailing logic. As the events of 9/11 revealed, terror-
ists need not physically transport weapons of mass de-
struction across borders but can improvise using instru-
ments not normally associated with military weaponry.
Moreover, even controlled immigration through offi-
cial entry channels may not impede the creation of
so-called “sleeper cells” in receiving countries. Yet,
while additional security measures have been taken
since 9/11 to better control and monitor entry of for-
eign migrants, we have not witnessed a clampdown at
the border on either side of the Atlantic. The policy
developments evident in both the pre- and the post-
9/11 worlds present two important puzzles: (1) Prior to
9/11, why would neoliberal strategies be applied pri-
marily to trade and capital mobility and not labor? and
(2) In a post-9/11 world where a handful of individ-
uals can unleash profound devastation by delivering
weapons of mass destruction on military or civilian
targets, why would states “risk migration” (Hollifield
1998)?

In this article, I offer a single model of state be-
havior that can account for policy developments in
each environment. I begin by exploring the limita-
tions of interest group-based models of policy and
counter that explaining the behavior of states requires
that we first reexamine the “national interest,” con-
ceptions of security, and the role of the state. This
is followed by empirical evidence of policy develop-
ment in the United States and Europe since 1945,
which is divided into four security environments: early
Cold War, détente, new world order (post-1989), and
post-9/11.

CONCEPTIONS OF SECURITY AND THE
ROLE OF THE STATE

Immigration as Domestic Policy

Immigration has long been seen solely as a domes-
tic public policy issue. Accordingly, most of the exist-
ing literature on the determinants of immigration and
border policy focuses exclusively on domestic factors.
Such approaches generally identify the potential do-
mestic “winners” and “losers” regarding changes in
immigration levels and model politics as a “bottom-
up” process that ends with dominant interests re-

flected in policy development (Gimpel and Edwards
1999; Kessler 1999; Money 1999). The state is largely
epiphenomenal to political processes. Marxist vari-
ants argue that immigration policies consistently fa-
vor business interests because of their privileged ac-
cess to the capitalist state, noting that use of foreign
labor can be used as a mechanism to keep wages low
and confound union action by dividing the working
class (Castells 1975; Castles and Kosack 1973). Oth-
ers argue that the dominance of business interests is
the product of collective action dynamics and Olso-
nian interest groups (Freeman 1995). Specifically, be-
cause the economic benefits of migration are concen-
trated (benefiting employers) while the costs of mi-
gration are diffuse (general public), employers have
a greater incentive to mobilize politically than those
who bear its costs. Moreover, because employers face
fewer collective action problems, they can more easily
mobilize and generate interest-group pressure on pol-
icy makers and create a clientist relationship with the
state.

If the state were truly beholden to the interests of
domestic employers, we would expect policy to dis-
play three consistent characteristics. First, because of
the advantages of foreign labor—including controlling
domestic wages, elasticity, willingness to perform jobs
that natives are reluctant to accept, and favorable work
ethic—one would expect a continual process of liber-
alization in policies concerning labor mobility. This has
clearly not been the case in any advanced industrial
state. In fact, policies over the past 30 years have been
overtly hostile to migrants and have stressed a strong
political will to “regain control” of the national borders
(Cornelius, Martin, and Hollifield 1994). Second, policy
would not display racial or ethnocultural preferences
in the allocation of immigrant visas or temporary work
permits. From a strictly economic standpoint, “a worker
is a worker,” regardless of their skin color or ethnocul-
tural background. Thus, we would expect preferences
to be based solely on economic need and the skill-set
a given migrant possesses. Instead, migration and bor-
der policies in most advanced industrial countries have
displayed an acute sensitivity to the racial, ethnic, and
cultural composition of flows. Finally, we should expect
relative openness and closure to conform to the busi-
ness cycle. However, this has not consistently been the
case in many immigrant-receiving countries, especially
in the post-1945 era (Hollifield and Zuk 1998; Potts
1990).

Migration is an increasingly global phenomenon
with economic, social, and foreign policy implications
(Rudolph n.d.). Though an internationally oriented ap-
proach would seem ideally suited to addressing issues
of policy development, migration has until very re-
cently been largely ignored by the mainstream inter-
national relations theory discourse (cf. Andreas 2000;
Hollifield 1992; Koslowski 2000; Meyers 2003; Rudolph
n.d.; 2003b; Rosenblum 2002). As will be shown, in-
ternational relations theory provides useful tools that
promote theory building capable of addressing fac-
tors at several levels of analysis and does not pre-
clude an active role for the state as political agent.

604

ht
tp

s:
//

do
i.o

rg
/1

0.
10

17
/S

00
03

05
54

03
00

09
0X

P
ub

lis
he

d
on

lin
e

by
C

am
br

id
ge

U
ni

ve
rs

ity
P

re
ss

https://doi.org/10.1017/S000305540300090X

American Political Science Review Vol. 97, No. 4

I begin with international relations theory’s most ba-
sic question: What interests lie at the core of state
behavior?

Three Dimensions of Security

While contemporary international relations scholars
have generally privileged military security as the pri-
mary raison d’etat (“high politics”), it is important to
recognize that security has both external (geopoliti-
cal) and internal orientations. Taken in this context,
the “national interest” of states can be defined largely
along three dimensions: (1) geopolitical security,
(2) the production and accumulation of material
wealth, and (3) social stability and cohesion. Grand
strategy can then be defined as the mix of policies
that provide aggregate maximization along these three
facets of security. Focusing on the politics of immigra-
tion and border policies is particularly useful in expli-
cating shifts in grand strategy since migration processes
affect all three domains.

Of the three domains, internal security has received
the least attention by scholars. Ole Wæver (1993, 23)
offers the concept of “societal security” to capture the
social–internal dimension of security that he defines
in terms of “the sustainability, within acceptable con-
ditions for evolution, of traditional patterns of lan-
guage, culture, association, and religious and national
identity and custom.” Whereas change is integral to
economic rationalization (increasing productivity, ex-
panding markets, etc.), national identity is often based
on a myth of primordial origins that prove enduring
over time. Change is thus inherently threatening. Pre-
cisely defining “societal security” is inherently prob-
lematic, for, as the extensive literature on nationalism
makes clear, social identities are often highly complex,
continually evolving, and often politically contested
(A. Smith 1991; R. Smith 1997). Societal security is of-
ten presented in solely ethnocultural terms and is mea-
sured by public reactions to demographic changes that
significantly alter a polity as an “ethnic community”
(A. Smith 1986; Wæver et al. 1993). However, few of the
world’s states are ethnically homogeneous and most of
today’s advanced industrial states are based at least in
part on including liberal ideology as a prominent com-
ponent of the national identity. A more accurate use
of “societal security” must acknowledge and account
for the political effects such “multiple traditions” have
on political discourse and policy outcomes (R. Smith
1997).

By introducing large numbers of people of di-
verse ethnocultural and ideological backgrounds to
a host-society, the globalization of migration repre-
sents a potentially significant threat to notions of stable
national identities, culture, and ways of life—regardless
of whether ethnocultural or ideological criteria are em-
phasized. Societal security is both a useful and a nec-
essary concept in order to understand the ideational
interests that can strongly influence state grand strat-
egy, especially when geopolitical threats are low. How-
ever, operationalizing such variables presents consid-

erable obstacles to social-scientific analysis. I draw on
four primary sources to operationalize threat percep-
tion in the area of societal security, including public
opinion polls, public and policy maker discourse anal-
ysis, symbolic voting practices (e.g., voter initiatives,
support for right-wing parties, etc.), and deconstruc-
tion of the policies themselves. Immigration and bor-
der policy is much more nuanced than terms such as
“open” and “closed” can capture. While both increased
border surveillance and reduction of immigrant visas
can be classified as “restrictive,” paying attention to
where such restriction is placed can tell us important
information regarding ulterior motives, as is often the
case with issues of race and culture in so-called multi-
cultural democracies. Admittedly, some of the indices
used here to determine societal insecurities have weak-
nesses. While numerous public opinion polls have been
taken in all case nations surveyed here, systematic and
consistently worded polls of opinion across the time
period examined here simply do not exist. Moreover,
racialized discourse is often reflected publicly only by
the extremist minority, though such ideas may strike a
chord among the mainstream who either are unwilling
to be publicly associated with such groups or are ex-
periencing cognitive dissonance with these notions and
their more liberal sense of social identity. Finally, while
right-wing political parties may present largely single-
issue platforms (such as anti-immigration), mainstream
voters may not mobilize on single issues. As such, us-
ing voting patterns for such parties as a proxy for so-
cietal insecurities may present a more limited indi-
cator of public sentiments than are actually present
in a given polity. However, even though no single
proxy for measuring “societal insecurity” exists, care-
ful analysis of these variables in aggregate provides the
most tangible means of understanding the political pro-
cesses involved where societal security interests are at
stake.

What, then, drives state behavior? Stephen Walt
(1987) suggested that state behavior regarding al-
liance formation is driven largely by the perception of
threat(s). I argue that threat perception lies at the core
not only of alliance behavior, but of other elements
of state grand strategy, including migration and border
policies. Prior to 9/11, we could expect the presence of
acute external (geopolitical) threats to result in a grand
strategy heavily favoring military and material dimen-
sions. Economic production is necessary for military
production as well as to reduce the risk of domestic un-
rest and political subversion, while military protection
is a prerequisite for a functioning industrial base and
stable economy. A general preference for policies of
openness to correspond with material and military in-
terests is also aided by reduced demands for closure re-
sulting from societal insecurities. Wartime nationalism
produces a “rally-round-the-flag” effect that bolsters
social cohesion by mitigating internal differences in the
presence of external enemies (groups associated with
external enemies notably excluded). However, when
external threats decline, such rally effects no longer
exist and domestic sensitivities to societal differences
often become more acute, and we could expect grand

605

ht
tp

s:
//

do
i.o

rg
/1

0.
10

17
/S

00
03

05
54

03
00

09
0X

P
ub

lis
he

d
on

lin
e

by
C

am
br

id
ge

U
ni

ve
rs

ity
P

re
ss

https://doi.org/10.1017/S000305540300090X

Security and International Migration November 2003

strategy to reflect such changes in the security environ-
ment.

Given this framework, several hypotheses can be for-
warded regarding migration and border policy devel-
opment based on elements that generate threat:

Threat Hypothesis—As external threats increase, mi-
gration policy will be more open, both in terms of vol-
ume and type, because trading states will seek to maxi-
mize economic gains available through labor mobility.
It should also correspond with foreign policy interests.

Rally Hypothesis—High degrees of external threat
result in more open migration policies with a low em-
phasis on ethno-cultural criteria, but when migrant
flows are associated with a military adversary, increased
closure directed at such groups will result (Adversary
Corollary).

Societal Proximity Hypothesis—Ethno-cultural pro-
ximity results in lower degrees of societal threat
(i.e. more toleration). Ethno-cultural distance is posi-
tively correlated with societal threat and more restri-
ctive policy.

Concentration Hypothesis—Geographic and tempo-
ral concentration of migrants (entry routes and/or set-
tlement patterns) increases the visibility of migration
flows and increases societal insecurity. This should re-
sult in more closed policy.

Assimilation Hypothesis—Ability and willingness of
migrants to assimilate into the dominant culture de-
creases societal insecurity and increases openness.

Legacy Hypothesis—Policy legacies with unintended
consequences are associated with societal insecurities
because they establish patterns of migration (“chain
migration”) that make it more difficult for states to con-
trol flows and creates perceptions of a loss of control.
This should result in more closed policy.

Liberal State Hypothesis—Socio-political liberalism
embedded in social and legal institutions limit restric-
tive state responses to ethno-cultural aspects of societal
in security. This limits the degree of closure given soci-
etal pressures for such policies.

9/11 Hypothesis—Newly established links between
migration and military threat will swing grand strategy
sharply toward closure.

To test these hypotheses, I examine immigration and
border policies in four advanced industrial states in
the context of threat environments since 1945. Because
the hypothesized model is based on the economic and
political interests of advanced industrial trading states
and the political implications of immigration flows,
all fit this classification, including the United States,
Germany, France, and Great Britain. While developing
states also have strong interests at stake in processes of
global migration—particularly the use of emigration as
an unemployment “safety valve” and the use of mi-
grant remittances as sources of much needed foreign
exchange—the security agenda of these states is suffi-
ciently different to require separate in-depth analysis.
The four cases examined here represent excellent “crit-
ical cases” for theory development not only because of
common economic and geopolitical interests, but also

because “societal security” is socially contested and is
based on competing visions of national identity. Het-
erogeneous, liberal, multicultural societies represent
the most difficult tests of models incorporating societal
security as an explanatory variable.

More difficult than case selection may be the ques-
tion of analytical focus. The complexity of global migra-
tion suggests that analysis focus not solely on structural,
domestic, or individual levels alone but, rather, on the
inter relationships between forces at various levels us-
ing a multilevel approach. International and internal
warfare are structural factors that shape global migra-
tion flows, as are the economic disparities that exist be-
tween developing and developed states. These patterns
largely determine the ethnocultural proximity of mi-
grant inflows, the temporal and spatial density of these
flows, and their magnitude. We must also recognize that
individual migrants are active players whose decisions
also have significant effects on social and political pro-
cesses. Migrants’ decisions regarding entry channel, set-
tlement patterns, and willingness to assimilate into host
cultures may either exacerbate or mitigate societal in-
securities in receiving countries (Heisler 1998). Sociol-
ogists have documented migrant preferences to follow
routes traveled by prior migrants (“chain migration”),
causing entry and settlement patterns to be geographi-
cally concentrated rather than diffuse, and hence, more
visible as well as more challenging for states to control
(Cornelius, Martin, and Hollifield 1994; Massey and
Garcia Espana 1987; Massey et al. 1998).

Although both structural changes and aggregate in-
dividual decisions set in motion patterns that drive po-
litical processes, closer scrutiny reveals a central role of
the state not only in shaping forces at other levels, but
also as an agent in its own right. Migration processes
initiated by structural forces are often the result of the
foreign and economic policies of the state (or group of
states), and policies initiated by the state can alter a
given decision calculus among individual migrants that
has similarly powerful effects. Policy legacies—such as
labor recruitment programs—establish initial links in
migration chains that are difficult to alter when politi-
cal interests change. Moreover, state policies regarding
one channel of migration can have direct effects on
flows through other channels. Given continuing eco-
nomic disparities, reducing legal visa quota limits sim-
ply shifts migration pressures to other channels, includ-
ing clandestine entry or through exploitation of refugee
or asylum policies. Such shifts in patterns initiated by
state policies have greatly contributed to contempo-
rary anxieties regarding migration more generally. In
addition to affecting structural patterns and individual
migrant decisions, the state plays a key role in framing
issues and providing information. It is also often in-
strumental when it acts to preempt political demands
from the general public—the most powerful, though
often latent, interest group. Policy makers have often
shaped policy in anticipation of rising societal threats,
reflecting their own respect for the potential politi-
cal volatility migration represents. In such instances,
the state functions as the sole political determinant of
policy.

606

ht
tp

s:
//

do
i.o

rg
/1

0.
10

17
/S

00
03

05
54

03
00

09
0X

P
ub

lis
he

d
on

lin
e

by
C

am
br

id
ge

U
ni

ve
rs

ity
P

re
ss

https://doi.org/10.1017/S000305540300090X

American Political Science Review Vol. 97, No. 4

What patterns should be present if the hypotheses
proposed here are valid? High degrees of geopoliti-
cal threat stemming from early Cold War instability,
belligerence, and mistrust should skew grand strategy
toward the military and material poles and generally to-
ward more open policies; openness stems from the com-
bination of a desire for higher aggregate economic re-
turns as well as the societal rally effect. As such threats
decline with the emergence of détente and later the end
of the Cold War, societal dimensions of security should
become increasingly more salient and policies should
move toward increasing degrees of closure. This shift
toward closure should become even more pronounced
after 9/11, as military interests increasingly converge
with societal rather than material interests.

GRAND STRATEGY IN THE EARLY COLD
WAR PERIOD

The acute sense of external threat posed by the Soviet
Union after WWII, in terms of both military power
and bellicosity, left Western nations vulnerable and
made the production of material power a state imper-
ative. In such an environment, the Threat Hypothe-
sis would predict that this high degree of geopolitical
threat would result in (1) more open policies in order
to maximize material power and reduce domestic insta-
bility and (2) societal insecurity subordinated to mili-
tary security interests. In terms of migration and bor-
der policies, geopolitical and economic interests should
predominantly shape outcomes.

Security in PostWar Europe

External threat in Europe was significant in two re-
spects. Geographic proximity and apparently expan-
sionist intent of the Soviet Union, coupled with the
devastating effects of the war on Western Europe, made
the possibility of Soviet military invasion a credible
threat to the security of continental Europe. Moreover,
the bleak domestic economic situation in Europe, espe-
cially during the severe winter of 1946, made the possi-
bility of domestic unrest and Communist insurgency all
too likely. European security was initially based on two
associated strategies: (1) integration for economic de-
velopment and collective security and (2) management
of trade and factor mobility to maximize economic
gains and speed the pace of postwar reconstruction.
Population management emerged as a key element of
grand strategy because of the high level of war casual-
ties and a drop in Western European birthrates from
35 per 1,000 in 1910 (3.5%) to 15 per 1,000 (1.5%) dur-
ing WWII.

Geopolitical threats were acutely felt in Germany,
whose infrastructure was obliterated during the war,
suffered significant casualties (9.1% of the total pre-
war population) and was an occupied nation at that
time. A widespread belief among Western policy mak-
ers that economic weakness would induce Communist
insurgency (and/or a possible Soviet invasion) soon
led to Marshall Plan aid (Larson 1985). Germany’s

use of foreign labor soon became a primary factor
in Germany’s tremendous post-war economic growth
as foreign aid facilitated a rapid reconstruction of the
country’s infrastructure (Hollifield 1992; Kindleberger
1967). Initially, no government-sponsored recruitment
was necessary to secure needed labor. From 1945 to
1950, West Germany received 8.3 million refugees, ex-
pellees, and displaced persons from the Soviet Union
and Eastern Europe, and they soon constituted one-
sixth of the population in West Germany (Martin 1994,
198). Although this large influx of migrants represented
a potentially significant threat to societal security, the
fact that most were ethnic Germans (Volksdeutsche)
reduced potential frictions (consistent with the Societal
Proximity Hypothesis). In fact, the presence of the Ver-
triebene actually served to strengthen German national
identity. Maintaining strong blood ties with ethnic Ger-
mans kept the notions of the German Volk and the
Kulturnation intact, even as the country was partitioned
into occupation zones and, later, into separate political
entities (Brubaker 1992).

Because the skill levels of these initial migrant in-
flows were generally high, economists described them
as one of the finest sources of additional labor in all
of Europe (Kindleberger 1967, 31; Ulrich, 1994). From
1950 to 1960, West German economic productivity rose
by an average of 6.7% annually, while unemployment
in key industrial areas such as Baden-Württemberg
and North Rhein-Westphalia remained very low (2.2%
and 2.9%, respectively) (Korte 1985, 30). However,
as Vertriebene migration from the East slowed over
time and the need for labor increased as the postwar
economy surged, Germany established a program of
worker recruitment through a series of bilateral agree-
ments. Initially, these agreements were established with
other European states, including Italy (1955), Spain,
and Greece (1960). Encouraged by the success of these
programs and the positive effect worker recruitment
was having on economic development, policy makers
gained confidence that labor inflows could be man-
aged effectively to maximize economic development.
As the available supply of European labor declined,
Gastarbeiter recruitment programs were extended to-
ward more ethnoculturally distant sending countries,
including Turkey (1961), Morocco (1963), and Tunisia
(1965). A temporary suspension of labor importation
in response to the 1966–67 recession resulted in a 25%
decline in the number of foreign workers in Germany.
The responsiveness of flows to this change in policy sup-
ported attitudes that the Gastarbeiter were truly tem-
porary migrants, not permanent immigrants (Brubaker
1992). These beliefs, coupled with the rapid growth of
the West German economy during the 1960s, ensured
that the economic primacy of Germany’s immigration
and labor recruitment policies were never questioned.

In France, the postwar emphasis on population
management was perhaps most conspicuous, stem-
ming from some 600,000 war casualties and an inter-
war fertility rate that was below replacement level
(less than two). Population was widely considered the
number one problem facing economic reconstruction
and growth, and it was also associated with military

607

ht
tp

s:
//

do
i.o

rg
/1

0.
10

17
/S

00
03

05
54

03
00

09
0X

P
ub

lis
he

d
on

lin
e

by
C

am
br

id
ge

U
ni

ve
rs

ity
P

re
ss

https://doi.org/10.1017/S000305540300090X

Security and International Migration November 2003

preparedness, signaling a predominant “populate or
perish” attitude (Tapinos 1975, 16). Even though fer-
tility rates in the immediate postwar period increased
remarkably, from under two in 1945 to nearly three by
1950, rapid economic recovery required viable labor
immediately. Thus, the first five-year plan adopted by
the newly created Conseil Général du Plan made man-
agement of migration a key element of postwar French
grand strategy. To manage labor recruitment, the
Office National d’Immigration (ONI) was established
on 2 November 1945 and opened offices in Italy in 1945,
followed by a bilateral labor recruitment agreement in
1947. French elites were confident that all immigrants
could be assimilated into French society regardless of
their origins (Brubaker 1992), although the ease of in-
tegration was facilitated by the fact that initial flows
were dominated by other Europeans (consistent with
the Societal Proximity Hypothesis). In 1946, 27, 831 of
the 30, 171 foreign workers who entered France came
from Italy (Freeman 1979, 79). Moreover, though colo-
nial legacies resulted in a rising Algerian population in
France, immigrants from the Maghreb accounted for
only 12.9% of France’s foreign population by 1954,
while Italians, Spaniards, and Portuguese accounted
for 46.25% (Hargreaves 1995, 11; INSEE 1985, 20;
Silverman 1992, 41).

The rapid economic growth in France during the
1950s not only generated increased demand for foreign
labor but also created incentives for employers to cir-
cumvent the bureaucratic apparatus set up by the ONI.
Employers soon began to engage in direct labor recruit-
ment with workers abroad, who were then later “regu-
larized” by government officials, creating a laissez-faire
approach to migration control (Freeman 1979). While
early CGP plans promoted governmental control over
migration to meet the nation’s labor demands, attitudes
among policy makers seemed comfortable with allow-
ing the free market to regulate supply and demand.
During these early stages of labor recruitment, eco-
nomic gains from migration were substantial, and few
sociopolitical tensions were associated with these initial
inflows. From 1955 to 1973, the total inflow of foreigners
rose from 58,952 to 347, 160, and much of this labor was
utilized in key economic sectors including mining, man-
ufacturing, construction, and services (Hollifield 1992;
Kindleberger 1967). As competition for foreign work-
ers increased with other European states (especially
Germany), bilateral recruitment agreements were di-
rected farther from the French Metropole. Agreements
with Morocco and Tunisia were reached in 1963, while
immigration from Algeria increased after the Evian
agreement of 1962 that established independence for
the former colony.

In Britain, government forecasts estimated a labor
shortage of some 600,000 to 1.3 million workers. Cou-
pled with a balance of payments deficit, declining mon-
etary reserves, and a large debt, Britain’s economic
situation was indeed dire. Initial efforts to supple-
ment the labor supply were manifest in policies that
enabled 335,000 European POWs to be employed in
Britian in 1945, and another 180,000 migrants to enter
in 1947 through the Polish Resettlement Act and the

European Volunteer Workers Program (Holmes 1988,
210–214). However, in contrast to its continental neigh-
bors, Britain’s postwar policies regarding economic de-
velopment (including labor requirements) were based
on its prewar imperial grand strategy (Heinlein 2002;
Kennedy 1989). Britain’s focus on the Commonwealth
as the source of material power was based as much
on the defense of the declining Pax Britannica and the
nation’s status as an elite world power, as it was on
concerns regarding the emerging Soviet threat. Trade
opportunities and acquisition of raw materials within
the sterling area provided a financial buffer against
Britain’s accumulated wartime debt and the financial
requirements needed to rebuild its war-torn infras-
tructure. Canada accounted for over 40% of Britain’s
exports, while Australia and New Zealand provided
dollar-free sources of meat, wheat, timber, and dairy
products (Paul 1997, 1). Although Britain’s “dollar
shortage” stemmed in large part from its balance of
payments deficit outside the sterling area, it maintained
a significant current account balance within the Com-
monwealth from 1950 to 1954. Moreover, London’s
role as central banker within the sterling area generated
reserves used to finance a significant portion of Britain’s
balance of payments deficit in the early postwar pe-
riod (Krozewski 2001). Given the economic incentives,
maintaining an imperial basis to grand strategy made
perfect sense.

When the Canadian Citizenship Act threatened
Commonwealth cohesion in 1945, Britain responded
by establishing the British Nationality Act of 1948.
The Act provided nearly identical rights for individ-
uals that fell within either of its two major categories—
Citizens of the United Kingdom and Colonies (CUKC)
and Citizens of Independent Commonwealth Coun-
tries (CICC)—including the right of free movement
(Hansen 2000; Money 1999). The British Nationality
Act established a citizenship regime that allowed Com-
monwealth states to define their own citizenship, but
ensured that such definitions could not take place with-
out reference to the United Kingdom. Richard Weight
suggests that “by legally codifying the concept of a far-
flung British family, the Act was in a sense the high point
of British imperialism” (Weight 2002, 137). Although
the Act did not include provisions for labor impor-
tation, the establishment of a free movement regime
allowed for market forces to drive migration patterns
consistent with Britain’s labor needs and shifts in the
business cycle. More importantly, its symbolic purpose
in establishing a sense of equality among Common-
wealth members was crucial to Britain’s economic in-
terests by reinforcing ties with its key sources of exports,
dollar-free imports, and capital accumulation.

External and Internal Security in the
United States

As the only state capable of keeping potential Soviet
expansionism in check, postwar American policy was
understandably focused on boosting material produc-
tion and increasing military defense. In terms of trading

608

ht
tp

s:
//

do
i.o

rg
/1

0.
10

17
/S

00
03

05
54

03
00

09
0X

P
ub

lis
he

d
on

lin
e

by
C

am
br

id
ge

U
ni

ve
rs

ity
P

re
ss

https://doi.org/10.1017/S000305540300090X

American Political Science Review Vol. 97, No. 4

state grand strategy, the empirical evidence generally
confirms the Threat Hypothesis—with some important
caveats. Though the societal rally effect bolstered na-
tional cohesion against the Soviet bloc, ethnonation-
alist insecurities were also mitigated by the existing
immigration regime in place since 1924. Moreover, in
the American case, the binary external–internal secu-
rity dynamic was more complex than initially theorized
herein. Ideological elements dominated political dis-
course, while ethnocultural aspects of societal security
were deemphasized. This was largely a function of how
the Cold War was initially framed by the Truman ad-
ministration.

The Truman Doctrine, outlined in a speech to the U.S.
Congress on March 12, 1947, defined the postwar inter-
national order as a struggle between two worlds, two
identities, and two opposing ways of life. This defining
moment of the early Cold War period not only shaped
conceptions of political order but, indirectly, shaped
conceptions regarding migration (Rosenblum 2002).
Because anti-Communist paranoia created internal
insecurities based on political ideology rather than
race or culture, the Cold War provided a lens through
which migrants were increasingly perceived in a more
positive light—a perspective actively cultivated by the
American executive. Postwar migrants—in the form of
refugees fleeing Communist countries—became a sym-
bolic element of the Truman Doctrine. Their trials and
tribulations not only supported the American concep-
tion of the Soviet sphere as oppressive, but, by allowing
them to enter the country, the United States was able to
affirm its self-conception as a defender of liberty and
personal freedom. In addition, because many of those
fleeing Communist countries were highly skilled, their
migration represented a net loss to the Soviet econ-
omy and military complex and a gain for the American
pursuit of material power.

The Displaced Persons Act (1948) established entry
provisions for 205,000 refugees fleeing persecution in
the communist bloc. In 1950, this level was increased to
415,744, and later the Refugee Relief Act of 1953 pro-
vided nonquota visas for an additional 214,000 people.
Confirming the link to U.S. foreign policy, a 1953 NSC
memorandum suggested that the Act served to “en-
courage the defection of all USSR nations and key per-
sonnel from the satellite countries” in order to “inflict
a psychological blow on communism” (Zolberg 1995,
123–24). Refugee waves following the Soviet invasion
of Hungary in 1956 and the Communist revolution in
Cuba in 1959 were admitted to the United States via
the Attorney General’s “parole” powers codified in the
Immigration and Nationality Act of 1952. The impact
of this positive image of migration created by Cold War
foreign policy can be reflected in the rapid reduction of
nativist sentiment in American public opinion. In 1947,
72% of poll respondents did not support the admis-
sion of 100,000 new refugees; however, by 1953, 47%
replied that they supported of the admission of 240,000
displaced persons (Fetzer 2000, 40).

Foreign policy considerations had strong effects on
other aspects of migration and border policies as well.
While the national origins quota system helped to keep

ethnocultural insecurities in check, racially discrimi-
natory policies strained political ties with key allies.
Because of this, Congress repealed the Chinese Ex-
clusion Acts in 1943 to improve ties with China, a
U.S. ally in the war against Japan. In 1946, Congress
established quotas for Asian allied nations, including
India and the Philippines. This was followed by the
McCarran–Walter Act (1952), which attempted to de-
flect criticism of America’s national origins quota sys-
tem by including an official prohibition against utiliz-
ing racial bases for admissions decisions. Truman felt
that the 1952 Act did not go far enough to establish
equality and vetoed the legislation in order to maintain
a positive U.S. standing with Asian allies;1 however,
Congress passed the 1952 Immigration and Nationality
Act over the veto. As the Containment strategy devel-
oped during the late 1950s and 1960s, relations with
Third World nations increased in geopolitical impor-
tance and again placed political strains on America’s
existing racial segregation and immigration quota sys-
tem. Racial discrimination and violence at home were
a consistent embarrassment to the United States in the
context of the Cold War containment strategy, and Civil
Rights Movement leaders were able gain leverage by
exploiting this tension (Borstelmann 2002). The inter-
play between foreign policy considerations and domes-
tic pressure emanating from the Civil Rights Movement
contributed to the profound liberalization of immigra-
tion policy in 1965 (King 2000).

U.S. policy makers placed a high emphasis on
America’s relationship with Mexico in terms of the
economic dimensions of early Cold War grand strategy
(Rosenblum 2002). On March 10, 1947, the Bracero
program of Mexican contract labor begun during
WWII was renewed. Mexican contract labor during the
war not only provided needed manpower in agriculture,
but also enabled domestic workers to shift into indus-
trial production jobs. As domestic labor shifted toward
higher-paying industrial production jobs, opportunities
in agriculture became less appealing to native work-
ers in terms of relative wages, working conditions, and
social status. Due to high demand, workers admitted
to the United States through the Bracero program in-
creased rapidly from 1947 to 1959, reaching a peak of
nearly 450,000 in 1956. In general, the program was
not politically contentious and did not generate soci-
etal insecurities until an increasing number of Mexican
migrants began to circumvent bureaucratic procedures.
From 1946 to 1954, apprehensions of illegal immigrants
increased from 99,591 to 1,089,583, trends character-
ized in the U.S. media as being tantamount to a foreign
invasion. Consequently, negative perceptions of migra-
tion countered positive conceptions generated by the
dynamics of the Cold War. In response to public outcry
over increasing levels of illegal immigration in 1954, the
Eisenhower Administration decided that a pure laissez-
faire approach was politically unstable, even in a Cold
War environment, and established a crackdown at the

1 Quota provisions were not allocated equally, as northern and west-
ern European nations received 85%, while quotas for Asian nations
were numerically insignificant.

609

ht
tp

s:
//

do
i.o

rg
/1

0.
10

17
/S

00
03

05
54

03
00

09
0X

P
ub

lis
he

d
on

lin
e

by
C

am
br

id
ge

U
ni

ve
rs

ity
P

re
ss

https://doi.org/10.1017/S000305540300090X

Security and International Migration November 2003

border dubbed “Operation Wetback” (Reimers 1985).
The INS apprehended over one million illegal aliens in
1954, projecting a strong image of a government that
had not lost control of the nation’s borders. The empir-
ical evidence suggests that reduced societal insecurities
evident during periods of high external threat required
a sense of public confidence in government regulation
of migration flows, implying an important caveat to the
Threat Hypothesis and the Rally Hypothesis. This dy-
namic has remained a dominant underlying theme in
U.S. immigration policy ever since.

To summarize, external threat and the need for mate-
rial production moved the security equilibrium sharply
toward the economic pole in both the United States
and Europe during the early Cold War period. Policy
developments that increased migration levels without
generating strong nativist public sentiment generally
support the Threat Hypothesis, though the specific
approaches taken varied somewhat due to intervening
variables that are case specific. Governments either
facilitated trade and factor flows directly, or did so
indirectly by adopting a laissez-faire approach. While
Britain and the United States displayed more sensiti-
vity to ethnocultural threats presented by migration
than Germany or France, all adopted a remarkably
open trading state approach to trade and factor mobi-
lity. Many of these policies, however, would prove
to have unintended consequences that contributed to
societal insecurities later on.

THE RISE OF SOCIETAL INSECURITIES

Perceptions of geopolitical threat began to decline
around the mid-1960s for several reasons: The bipolar
Cold War system had stabilized over time, confidence in
policy makers’ ability to effectively manage the East–
West relationship increased, and economic growth in
the United States and Europe bolstered Western ma-
terial power. However, although trading state openness
resulted in unprecedented economic growth among
western nations, it also served to increase the disparity
between North and South, increasing global migration
pressures and altering migration patterns. In such an
environment, the hypotheses offered here would pre-
dict increased societal insecurities and associated poli-
cies of closure. The conventional wisdom suggests that
closure was primarily a function of the 1973 oil em-
bargo; however, such economic explanations alone can-
not account for the timing of restrictionist sentiments
and policies, the ethnocultural focus of policies, or that
such attitudes and political trends were sustained as
economic conditions changed.

Societal Insecurity in Europe

In France, laissez-faire recruitment of foreign labor
led to a gradual transformation of migration inflows.
From 1946 to 1955, Italians accounted for 66.8% of mi-
grant inflows, but by 1973 they accounted for only 4.6%
(Hollifield 1994, 153). In contrast, by 1975 over 30% of
France’s foreign population came from the Maghreb

(INSEE, 1992). Moreover, uneven geographic distri-
bution of these new immigrants served to increase their
visibility and raise awareness of societal changes among
the general public (consistent with the Societal Prox-
imity and Concentration Hypotheses). In contrast with
prior inflows, migrants entering France in the 1960s
were concentrating in the industrial Paris and Lyon re-
gions where industries that utilized foreign labor were
concentrated (Money 1999, 120–23; see also Hollifield
1992; Ogden 1989, and White 1995). The lack of suitable
housing and the emergence of ethnic enclaves created
a rise of urban ghettos and shantytowns (bidonvilles)
in many of these industrial centers. Where immigrants
were previously associated with economic growth dur-
ing the Trente Glorieuses, they increasingly became as-
sociated with urban plight and ethnic segregation.

A rise in family-oriented migration (rather than sin-
gle males) also served to increase the visibility of chang-
ing migration patterns. In addition to representing a
more permanent rather than temporary or cyclical flow,
family migration from North Africa led to deeper pen-
etration of the housing market, a larger presence of
foreign children in school, and increased opportuni-
ties for interaction with mainstream French society.
All of these factors served to increase visibility of the
foreign population (Hargreaves 1995). In addition, the
fact that Maghribi migrants were largely Muslim also
served to generate societal fears due to a perception of
cultural distance and growing doubts regarding their
ability and willingness to assimilate (Oriol 1992; Roy
1994). In 1968, domestic political and social unrest
prompted a reevaluation of migration policies by the
French government. Reports commissioned by the gov-
ernment suggested that even though migration offered
considerable economic gains, it threatened social co-
hesion. These reports argued that while French soci-
ety was based on liberal principles founded during the
revolution, it had a “threshold of tolerance” (seuil de
tolérance) for assimilating immigrants, and neglecting
this dynamic could have significant negative sociopo-
litical effects. It must be noted, however, that this seuil
de tolérance applied primarily to Maghribi migration,
not European migration (Silverman 1992, 75–76).

Responding to societal insecurities, the Franco–
Algerian Accords of 1968 were signed to limit mi-
gration from North Africa (Hollifield 1992, 68–69). In
addition, new sanctions were implemented against em-
ployers of illegal aliens to deter would-be migrants. On
3 July 1974 recruitment of foreign workers was tem-
porarily halted, and options for reducing the country’s
non-European foreign-born population were consid-
ered. Consistent with the Liberal State Hypothesis,
forced deportations were not a viable policy option.
Instead, France offered financial incentives (10,000
francs) to foreign workers to voluntarily return to their
home country in 1977. Unfortunately for policy mak-
ers, the program met with limited success as the num-
ber of participants was relatively low (80,000) and was
dominated by Spaniards and Portuguese rather than
Maghribi, for whom it was primarily intended (Weil
1991). An unintended consequence of France’s new re-
strictionism was that as official recruitment channels

610

ht
tp

s:
//

do
i.o

rg
/1

0.
10

17
/S

00
03

05
54

03
00

09
0X

P
ub

lis
he

d
on

lin
e

by
C

am
br

id
ge

U
ni

ve
rs

ity
P

re
ss

https://doi.org/10.1017/S000305540300090X

American Political Science Review Vol. 97, No. 4

closed, migration flows increasingly shifted toward
clandestine immigration and asylum/refugee channels.
Future responses to societal insecurities increasingly
focused on perceived exploitation of this entry channel.

As in France, the sociopolitical problems associated
with the Gastarbeiter program became increasingly ev-
ident in Germany in the late 1960s. In the period from
1968 to 1972, the foreign work force in Germany rose
from one million to 2.6 million, which amounted to
an increase from 5% to 12% of the total German
work force (Martin 1994, 201). More important was the
change in sources of migration, as Turkish migration
increased from 2,500 in 1960 to 605,000 in 1973 (Bade
1983, 70). In addition, it soon became apparent that the
“temporary” guest workers were increasingly less inter-
ested in returning to their home countries and were be-
coming a permanent part of German society. From 1963
to 1967, the percentage of foreign workers who had
been in Germany for over three years increased from
22% to 45%. More permanent and family-oriented
migration led to changes in preferred housing among
immigrants. Although the communal housing offered
by the German government was initially appealing to
guest workers (largely single males), by the end of the
1970s more than 80% of the foreign workers chose to
live in individual apartments and rooms secured on
the open market (Heckmann 1985, 73). Similar to the
French case, this change brought the resident foreign
population into closer contact with German society, in-
creasing social visibility. The Gastarbeiter program was
halted in November 1973, coinciding with the OPEC
embargo. However, Ulrich Herbert (1990, 324–25) sug-
gests that “the oil embargo had been little more than
a supplementary factor; it had provided a useful occa-
sion to check the influx of foreign workers and reduce
their numbers—without any great resistance from the
worker-exporting countries and without wearying pub-
lic discussion on the social consequences of this mea-
sure.”

As was the case in France, closure of recruitment
channels simply redirected flows. Due to its extremely
liberal asylum policies, codified in Article 16 of the basic
law, Germany received a disproportionate percentage
of Western Europe’s refugee applicants (Joppke 1999).
It makes sense that migrants that were no longer able to
gain entry through labor recruitment programs would
seek to exploit these policies. In the year following the
recruitment stoppage, new asylum claims almost dou-
bled, from 5,595 to 9,424, and the number of applica-
tions increased annually, reaching a level of 100,000 by
1980 (Joppke 1997, 277). Although public demand to
amend Article 16 was vociferous, doing so was con-
sidered contrary to Germany’s desire to emphasize its
postwar liberalism and to distance itself from its total-
itarian past. Thus, initial responses centered only on
administrative and procedural changes, including pro-
visions articulated in the 1982 Asylum Procedures Law
that sought to “fast-track” dismissal of “patently un-
founded” claims. In addition, the government sought
to reduce the size of the foreign population by pro-
moting voluntary repatriation. At that time, the for-
eign population in Germany had reached a peak of

4.7 million. Under a new program begun in 1982, fi-
nancial assistance (DM10,500) was offered to former
guest workers that were willing to voluntarily return to
their homeland. However, as with France’s attempt at
similar policies, the program had little effect on the size
and composition of the foreign population.

Britain’s response to societal insecurity is slightly dif-
ferent from that of France or Germany. Britain’s free
movement regime did not initially generate significant
migration from NCW countries; however, over time
flows from these sources increased dramatically. Dur-
ing the first months of 1958, immigration from India
and Pakistan rose sharply, and in February, the number
of new arrivals from these sources was equivalent to
30% of the total flow in 1957 (Spencer 1997, 91). From
1959 to 1961, immigration from the West Indies surged
from 16,400 to 66,300 (Patterson 1969, 3). Even though
they were still relatively insignificant levels in relation
to the total population in Britain, they generated con-
siderable perceptions of threat among Britons. How-
ever, it appears that societal insecurities were based
more on fears of the size of future flows (if trends
continued unabated) rather than on the size of current
flows. These fears became especially acute during the
race riots in 1958, which increased perceptions among
policy makers that changing ethnic demographics had
volatile political consequences. Attitudes among the
general public mirrored these concerns. A Gallup poll
conducted in June 1961 found that 67% of the public
favored immigration restrictions (Patterson 1969, 18).

In the wake of rising New Commonwealth migration,
the 1962 Commonwealth Immigrants Act subordinated
entry of all Commonwealth citizens and CUKCs whose
passports were issued outside the United Kingdom to
immigration and labor market controls consisting of a
three-tiered voucher system. When it became apparent
that Indians and Pakistanis were disproportionately ap-
plying for all three categories of vouchers in 1963, the
government stipulated that no Commonwealth coun-
try could receive more than 25% of the available “C”
vouchers (non-skills-based). In August 1965, a La-
bor Government White Paper tightened the 1962 con-
trols by establishing a ceiling on New Commonwealth
immigration to 8,500 per annum and abolishing the
“C” voucher category altogether (Layton-Henry 1994,
284).

Subsequent amendments to the Commonwealth Im-
migrants Act in 1968 stipulated that only patrial
CUKCs remained free from immigration controls while
others were regulated via an entry-voucher system,
with initial levels set at 1,500 heads of household. Re-
strictions on Commonwealth immigration were also
strengthened with the passage of the 1971 Immigration
Act, which removed most remaining migration privi-
leges accorded to Commonwealth Citizens. The pro-
visions of the 1971 Act actually served to increase the
number of people entitled to enter the United Kingdom
without restriction, although these were composed pri-
marily of people of European extraction who, it was
believed, posed a lower degree of societal threat. In this
sense, the 1971 Act marks a significant shift in British
grand strategy, from one predicated on imperial legacy

611

ht
tp

s:
//

do
i.o

rg
/1

0.
10

17
/S

00
03

05
54

03
00

09
0X

P
ub

lis
he

d
on

lin
e

by
C

am
br

id
ge

U
ni

ve
rs

ity
P

re
ss

https://doi.org/10.1017/S000305540300090X

Security and International Migration November 2003

to one increasingly predicated on ties with its European
neighbors.

The passage of the British Nationality Act of 1981
was a milestone in the British shift toward a soci-
etal reactionist posture. Through the Act, the status
of “British subject,” which had long been synonymous
with “Commonwealth citizen,” was for the most part
abolished (Hansen 2000, 214). The 1981 Act abolished
the Anglo-Saxon tradition of jus soli, where member-
ship is primarily a function of place, and replaced it
with a citizenship regime based on the principle of
jus sanguinis, where membership is a function of de-
scent (Goulbourne 1998, 54; see also Brubaker 1992).
The Act also sought to shore up loopholes in migration
and border policy. It established the “primary purpose”
rule that prohibited the entry of fiancées or spouses of
British women citizens unless they could prove that
the primary purpose of the marriage was not to garner
settlement rights in the United Kingdom. A few years
later, the Tamil crisis of 1985 raised British sensitivi-
ties to the possible exploitation of refugee and asylum
policy by economic migrants, and Home Office officials
used their deportation powers with extreme prejudice
(Joppke 1997). As the number of asylum applicants
increased, societal calls for closure increased as well.

The empirical evidence from Europe supports many
of the hypotheses presented here. A changing geopo-
litical environment, increasingly diverse flow of migra-
tion, settlement patterns, and integration problems all
served to increase societal insecurities and push poli-
cies toward closure. Yet the evidence also supports
ideational and institutional constraints on policies sup-
porting ethnonationalist impulses. The Liberal State
Hypothesis is supported by the fact that draconian de-
portation measures were generally eschewed in favor
of financial incentives for voluntary repatriation, and
though augmented, refugee and asylum regimes were
still functioning to provide protection for bona fide ap-
plicants.

Societal Insecurity in the United States

The Rally and Threat Hypotheses predict that a re-
duction geopolitical threat and concomitant reduction
in focus on opposition to the USSR as a major com-
ponent of U.S. national identity would result in more
closed policies as societal insecurities become more
pronounced. Furthermore, hypotheses based on demo-
graphic indicators should also be positively correlated
with increased levels of societal insecurities. The un-
desired effects of prior policies compounded societal
insecurities, suggesting support for the Legacy Hypoth-
esis. However, in the United States the timing of policy
changes in response to growing societal insecurities is
less parsimonious than in the European cases, reflect-
ing perhaps a cognitive dissonance between such re-
strictionist impulses so soon after a major policy shift
symbolic of U.S. liberalism (the 1965 amendments).

In addition to a changing geopolitical environment,
two policy legacies had a significant impact on the gen-
eration of societal insecurities and, later, on changes

in policy. The 1965 changes in U.S. immigration law
shifted visa preferences from those based on national
origins to those predominantly based on family reuni-
fication. Policy makers skeptical of the potential so-
ciocultural changes of such a change in policy were as-
suaged by the notion that a system based on familial ties
would replicate the existing demographic characteristic
of the nation. In practice, however, the 1965 changes re-
sulted in pronounced increases in Mexican, other Latin
American, and Asian immigration flows. Over time
these changes significantly affected the ethnic com-
position of the country as a whole. From 1960 to
1990, the percentage of the population that was non-
Hispanic white decreased from 85.1% to 75.2%, while
the percentage of Hispanics and Asians increased from
3.5% to 9% and from 0.6% to 2.9%, respectively
(Bean, Cushing, and Haynes 1997, 126). Consistent
with the Societal Proximity Hypothesis, public opinion
reflected a growing sense of societal insecurity associ-
ated with such changes (Fetzer 2000). Roper Center
and Gallup polls show that the percentage of Amer-
icans who favored decreasing existing levels of immi-
gration increased from 33% in 1965 to 42% by 1977. By
1993, this percentage increased to 65% (Shanks 2001,
234).

The second policy legacy that affected attitudes and
later policy development involved the Bracero pro-
gram. As mentioned above, the increase in illegal im-
migration associated with the Bracero program had
a significant impact on attitudes about migrants—
particularly Mexican migration. Even after the la-
bor importation program was abolished, notions of
Mexican migrants as lawbreakers had lingering residu-
als in the American public consciousness even though
they offered economic benefits to the country. More-
over, the program established migration linkages that
resulted in chain migration that persisted long after it
was discontinued (Massey and Espana 1987; Massey
et al. 1998). Effects of such chain migration are mani-
fest both in legal and clandestine channels of entry. In
terms of the naturalization channel, past migration (and
jus soli citizenship rights afforded to children of immi-
grants) provided familial ties that gave family members
preference under the post-1965 immigration regime. In
terms of clandestine flows, past migration created so-
cial linkages that facilitated illegal immigration through
the availability and use of social capital. The result is
that from 1965 to 1975, apprehensions of illegal aliens
increased by roughly 400%.

Beginning with its first comprehensive migration
regime established in 1924, U.S. policy hinged on a no-
tion of government control (Torpey 2000). The chang-
ing characteristics of migration flows in the 1960s and
1970s suggested that structural rather than governmen-
tal factors were driving migration dynamics. As such,
discourse in the 1970s and 1980s revolved around no-
tions of “regaining control” of our “neglected borders.”
Faced with growing numbers of migrants from non-
European sources, initial efforts to respond to societal
insecurities came in the form of revised Western Hemi-
sphere limits on immigrant visas, as stipulated in the Im-
migration and Nationality Act Amendments of 1976.

612

ht
tp

s:
//

do
i.o

rg
/1

0.
10

17
/S

00
03

05
54

03
00

09
0X

P
ub

lis
he

d
on

lin
e

by
C

am
br

id
ge

U
ni

ve
rs

ity
P

re
ss

https://doi.org/10.1017/S000305540300090X

American Political Science Review Vol. 97, No. 4

The Act implemented per country limits of 20,000—a
change which most significantly affected Mexico, which
had previously only been subject to the hemispheric
limit of 120,000. This was just the beginning of a more
comprehensive reassessment of policy. In 1978, the Se-
lect Commission on Immigration and Refugee Policy
recommended a comprehensive control policy aimed at
both the border and the interior. The commission rec-
ommended (1) sanctions against employers that hire il-
legal immigrants, (2) the development of a secure iden-
tification system, (3) enhanced border enforcement,
and (4) amnesty for illegal aliens currently residing in
the country. Most of these recommendations comprise
the basis of the Immigration Reform and Control Act
(IRCA) passed in 1986. Declaring amnesty for exist-
ing illegal immigrants was intended to provide a “clean
slate” for the increased control provisions articulated
in IRCA, especially border control. From 1986 to 1990,
the budget allocated for the Border Patrol increased
from $154 million to $262.6 million, which constituted
the largest portion of the INS’s budget for enforcement
activities (Rolph 1992, 45).

The public discourse and politics leading up to IRCA
suggest that the rise of identity as a security object cre-
ated strong pressure for closure in the United States.
Increases in non-European migration and a concomi-
tant rise in illegal immigration, especially from Mexico,
were greeted with alarm and calls for firm measures to
halt such migration. Policies enacted in response to such
threats included both external (border) and internal
(sanctions) measures, even though IRCA’s employer
sanctions provisions were vehemently opposed by busi-
ness interests who saw this as a negative externality
on business processes. In both the United States and
Europe, the strong negative reaction to new migration
trends resulted in calls for closure and made policy
makers aware that U.S. grand strategy could not be
based on pure trading state openness. Societal security
was an increasingly volatile issue that would have to be
seriously taken into account.

As in the European cases, although the evidence sup-
ports hypotheses pertaining to rising societal insecuri-
ties and resultant policy closure, government response
to identity-centered issues was not unidirectional. If
government was completely responsive to ethnocul-
tural restrictionist impulses, the most direct approach
would have been to significantly scale-back quota levels
and repeal the 1965 family-based immigration regime.
Because of its symbolic importance to the liberal facet
of America’s multiple-traditions of national identity,
such measures were not politically viable options. As
such, it would seem that the evidence supports the Lib-
eral State Hypothesis as a constraint on policy options.

BALANCING SOCIETAL AND MATERIAL
INTERESTS

The collapse of the Berlin Wall in 1989, marking the
end of the Cold War, served to further decrease geopo-
litical threats in the United States and Europe. Given
that external threat declined significantly at the close

of the 1980s, the Threat Hypothesis would predict a
continued—perhaps even dramatic—rise in restriction-
ism. Indeed, several indicators suggest that societal in-
securities continued to increase in the 1990s, especially
in those areas most acutely affected by migrant flows.
In the United States, large numbers of illegal aliens
brazenly crossed the border at the San Ysidro port of
entry, fueling rhetoric of an “alien invasion.” Similarly,
in Europe, increasing violence levied against foreigners
and the rising popularity of anti-immigrant party plat-
forms signaled increasing perceptions of societal threat.
Yet European nations remained committed to open
migration within the EU, even though this left individ-
ual states dependent on the ability of other European
states to police their borders effectively. In the United
States, immigration quota levels remained steady, and
the first legislative change enacted in the 1990s actually
resulted in a 40% increase in overall legal admissions.
These developments would seem to strongly contradict
the Threat Hypothesis. It would appear that the geopo-
litical environment and demographic characteristics of
migration flows did in fact contribute to an increase
in societal insecurities as predicted; however, state re-
sponses to such pressures appear to be anomalous. How
can this anomaly be explained?

Policies in the late 1980s and 1990s must be taken in
the context of the continued transition to flexible pro-
duction modes and the growing importance of human
capital. Although societal insecurities may be gener-
ated by migration, the movement of labor has become
increasingly linked with the accumulation of material
power, creating a strong political tension between ma-
terial and societal interests. What has emerged from
this environment is a governmental response similar
to that utilized with trade issues. In the trade realm,
Judith Goldstein (1986, 166) has shown that in order
to finesse this political impasse, states need to appear
responsive, not necessarily to be responsive. The same
would appear to hold true for migration. Rather than
responding to public anxieties by stopping migration
flows, policies are directed specifically at the source
of societal threat, not necessarily on volume. This in-
volves addressing those elements of migration reflected
in the Societal Proximity, Concentration, and Assimi-
lation Hypotheses rather than reducing volume from
all sources.

Fortress Europe and Third Country Nationals

Societal insecurities inflamed political discourse in
Europe as the Cold War ended. A 1998 poll conducted
in France found that 75% of the population believed
that French national identity was in jeopardy if mea-
sures were not taken to limit the size and inflow of the
foreign population (Hargreaves 1995, 151). This sen-
timent increased nearly 10% from results taken in a
similar poll in 1985. In Britain, polls revealed increas-
ing public concerns regarding migration and, in 1999,
reached their highest levels in two decades (Migration
News, May 2000). In Germany, public opinion polls
taken in 1992 showed that 75% of Germans supported

613

ht
tp

s:
//

do
i.o

rg
/1

0.
10

17
/S

00
03

05
54

03
00

09
0X

P
ub

lis
he

d
on

lin
e

by
C

am
br

id
ge

U
ni

ve
rs

ity
P

re
ss

https://doi.org/10.1017/S000305540300090X

Security and International Migration November 2003

drastic action to control the wave of migration flow-
ing through its asylum channel (Joppke 1999, 92). In
addition, support for the antiimmigrant Republikaner
party increased sharply during the early 1990s, while
other radical expressions of societal unrest were mani-
fest in violent attacks against foreigners in the cities of
Hoyerswerda, Rostock, Mölln, and Solingen.

Analysis of European policy suggests the “problem”
of migration rests with the movement of “third coun-
try nationals” (TCNs), not necessarily with migration
per se (cf. Newland and Papademetriou, 1998;
Papademetriou 1996). EU cooperation has been en-
gaged not only to facilitate migration internally, but also
to help stem TCN migration from outside of what many
have characterized as the “fortress Europe.” This in-
cludes information sharing, linkage politics, border en-
forcement, and policy harmonization concerning asy-
lum and refugee policies (Koslowski 2000). Linkage
politics, including linking foreign aid and entry into the
EU in exchange for cooperation in controlling TCN
migration into the EU, has emerged as a common ap-
proach. Given its large land borders to the East, these
methods have been used vigorously by the Germans.
In 1993, Germany linked the continuation of visa-free
travel for Poles entering Germany with their accep-
tance of a readmission agreement. By accepting, the
Polish government was committed to “reaccepting” all
undocumented TCNs attempting entry into Germany
across the German–Polish border. Germany also of-
fered to provide $80 million of direct financial assis-
tance as an additional incentive and to offset costs
during the first two years of the implementation of
the agreement. In 1997, Germany provided another
$66 million of aid to Poland in order to help secure the
border and stop clandestine entry of TCNs. For Poland
and the Czech Republic, as well as other countries seek-
ing entry in the union, EU policy makers made it clear
that the acceptance of EU border policy is a require-
ment for accession (Andreas 1999), most of which has
already been negotiated and confirmed by candidate
countries.

When labor recruitment programs were halted in
the early 1970s, applications for asylum from TCNs
increased dramatically. Although Western states sup-
ported the establishment of a liberal refugee regime in
the wake of World War II, perceptions that economic
migrants were increasingly exploiting liberal policies
rose sharply during the 1980s and into the 1990s.
This connection between illegal immigrants (generally
TCNs) and the exploitation of asylum procedures was
supported by new evidence in Britain. In 1995 it was
found that two-thirds of the illegal aliens detected by
authorities applied for asylum to prevent deportation.
Because the asylum/refugee channel was specifically
associated with TCN migration, closure and control
mechanisms during the 1990s were focused on this
mode of entry. The Dublin Convention (1990) estab-
lished a mechanism for policy harmonization among
EU states and sought to close policy loopholes being
exploited by TCNs. It has been argued that acceptance
of the Dublin Convention “was one of the key precon-
ditions agreed upon by the EU member states in order

to proceed with consolidating the single European mar-
ket and eliminating border controls within the EU”
(Newland and Papademetriou 1998, 643). The Conven-
tion stipulates that asylum applicants must generally
submit their claim in the first EU country they enter,
who is then responsible for deciding their case. This
ensures that asylum decisions are made once—and only
once—since the decision of the first EU nation will be
considered binding in all other EU member states, and
eliminates the practice of “asylum shopping” among
potential applicants (Koslowski 2000).

New policies were established in Germany, France,
and the United Kingdom that provided the means to
fast-track asylum procedures and granted the state
considerable discretion in dismissing “patently un-
founded” claims. Moreover, the concept of the “safe
third country” emerged as another means of reduc-
ing asylum claims from TCNs. The third country op-
tion, adopted domestically by most European states,
automatically dismisses claims if applicants arrive from
other European nations or transit countries that prima
facie do not produce refugees. These lists of “safe
countries” were established in consultation with the
UNHCR and may vary—albeit generally slightly—
between member EU states. The combination of con-
trol measures has served to significantly reduce the
number of asylum applications filed in the EU dur-
ing the 1990s. In Germany, these various provisions
and procedures resulted in a 70% decline in asylum
applications from 1992 to 1995, while in France, appli-
cations from 1990 to 1996 declined over 30%. In the
United Kingdom, the percentage of asylum applicants
granted acceptance declined from 85.9% in 1990 to
23.8% in 1997 (Newland and Papademetriou 1998, 645;
Joppke 1999, 94).

The empirical evidence suggests that European
states have sought to quell nativist impulses by focusing
policy on exclusively on restricting the movement of
TCNs—those whose societal proximity is considered
more remote than those of other Europeans—while
maintaining a commitment to internal openness within
the EU. The key to “finessing” the tension between
societal and material interests seems to rest on the con-
struction of an image of border integrity on the regional
level and closing perceived loopholes that TCNs may
exploit to circumvent immigration controls.

The Gatekeeper Approach in the
United States

As in Europe, public concerns regarding immigration
continued to increase in the United States in the 1990s,
consistent with the Threat Hypothesis. Polls taken in
1993 showed that 65% of American favored decreased
levels of immigration, up from 42% in 1977 (Shanks
2001, 234). Moreover, highly publicized images of mi-
grant border dashes through the San Ysidro border
checkpoint fueled passionate anti-immigrant rhetoric.
In 1994, voters in California overwhelmingly passed
Proposition 187, an anti-illegal immigration bill rid-
ing the crest of societal insecurities. Organizers and

614

ht
tp

s:
//

do
i.o

rg
/1

0.
10

17
/S

00
03

05
54

03
00

09
0X

P
ub

lis
he

d
on

lin
e

by
C

am
br

id
ge

U
ni

ve
rs

ity
P

re
ss

https://doi.org/10.1017/S000305540300090X

American Political Science Review Vol. 97, No. 4

proponents often suggested that the bill was intended
to “send a message” to policy makers, even though the
law was almost immediately struck down in the courts
(Andreas 2000). In the 1990s, government responses
to societal insecurities were focused on adjusting for
the unintended effects of the 1965 amendments to the
INA with regard to European migration and crafting
an image of security along the U.S.–Mexico border.
In response to economic interests, legal immigration
quota levels remained steady and special provisions
were made to recruit highly skilled labor.

The Immigration Act of 1990 (IMMACT) resulted
in a 40% increase in overall legal admissions into the
United States; however, its provisions were shaped to
alter the qualitative characteristics of immigrant in-
flows by stressing economic criteria and augmenting
ethnocultural characteristics of inflows. IMMACT in-
creased the number of employment-based visas from
54,000 to 140,000 and established the H-1B visa pro-
gram for highly skilled workers. IMMACT also estab-
lished an allocation for “diversity immigration” visas
(DV-1), intended to “correct” for the adverse effects
of the 1965 amendments. Beginning in October 1994,
IMMACT provided 55,000 immigrant visas to be dis-
tributed among nationals or nations that had sent fewer
than 55,000 legal immigrants to the United States over
the previous five years—a move clearly intended to
“balance” the dramatic rise in the proportion of Latin
American and Asian immigration. In the first appor-
tionment of such visas, nearly 50% (24,550) were allo-
cated to European nations.

Similar to the approach taken in Europe, open-
ness regarding entry levels has been politically “pur-
chased” through highly symbolic closure policies
(Andreas 1999). In the United States, these were man-
ifest in border enforcement policies directed at mi-
gration from Mexico and Latin America. In the early
1990s, the INS initiated a series of border enforcement
programs along the U.S.–Mexico border, beginning in
El Paso (1993) and, later, in San Diego (1994), Nogales
(1995), and East Texas (Operation Rio Grande, 1997).
Given the highly visible flows crossing along the San
Diego–Tijuana region, “Operation Gatekeeper” be-
came the symbolic flagship for the new strategy. Op-
eration Gatekeeper, begun in October 1994, employed
a highly visible, multilayered approach to sealing this
porous border region. In 1996, additional funding for
such border enforcement measures was secured as part
of the Illegal Immigration Reform and Immigrant Re-
sponsibility Act (IIRAIRA), which also called for 1,000
additional Border Patrol agents to be hired annually.
From 1993 to 1997, the annual budget for the INS
increased from $1.5 billion to $4.2 billion, while the
annual budget for the Border Patrol increased from
$354 million to $877 million. Some scholars have sug-
gested that these methods are a failure, since they have
failed to stem the flow of illegal immigrants entering
from Mexico (Cornelius 1998b, 392). Indeed, exam-
ining apprehensions statistics at the sector level re-
veals that while apprehensions in the San Diego sector
decreased, apprehensions increased in zones East of
San Diego. In aggregate, Border Patrol apprehensions

actually increased between 1992 and 1998, from
1,199,560 to 1,555,776 (INS 1998, 209–10).

The border enforcement strategy that has emerged
during the 1990s focused on curbing flows in high-
traffic zones such as San Ysidro and dispersing them
geographically. While this may not reduce the over-
all volume, it has significantly reduced the visibility of
border transgressions and, thus, an element significant
to societal threat perception (consistent with the Con-
centration Hypothesis). Most of the rhetoric concern-
ing illegal migration has focused on Mexico and Latin
America, though most estimates suggest that about
40–50% of the illegal immigrant population in the
United States consists of “out-of-status” migrants—
those who enter legally, then overstay the terms of their
visa. The Gatekeeper approach is very successful in two
ways: (1) It is a highly visible government response that
projects a symbolic image of control (Andreas 1999),
and (2) it makes reduces the visibility of migration flows
by reducing concentration, in both time and space. In
an important sense, by making flows “disappear,” Gate-
keeper serves to make border crossers as politically un-
problematic as those who over stay the terms of their
visa. Thus, reducing societal insecurities in the 1990s
was essentially based on the principle of “out of sight,
out of mind.” While politically successful during the
1990s, this notion would be challenged after 9/11.

SECURITY AND MIGRATION AFTER
SEPTEMBER 11

The post-9/11 world challenges the policy equilibrium
crafted during the 1990s in several significant ways. Eas-
ily the most salient change is the widespread recogni-
tion of the link between military security and the ef-
fective control over global migration. In terms of the
framework presented here, this change complicated the
external–internal model of security in terms of migra-
tion. Prior to 9/11, security threats related to migra-
tion were generally an issue of internal/societal security
that pressed grand strategy toward closure, while the
linkage between military and material interests pushed
grand strategy toward openness. The 9/11 Hypothesis
predicts that this new link between military and societal
facets of security will initiate severe restrictionism with
respect to migration and border policies. The grand
strategy equilibrium crafted during the 1990s seems
predicated on addressing societal fears generated pri-
marily through the visibility of immigration flows (so-
cietal proximity, geographic and temporal concentra-
tion, etc.); however, the post-9/11 environment would
seem to preclude such an approach. What is threatening
about the clandestine entry of terrorists and the pres-
ence of sleeper cells in the homeland is essentially their
invisibility—a specter lurking in the shadows. Thus, se-
curity would seem to require policies that increase visi-
bility rather than decrease it. However, early empirical
evidence only partly supports the 9/11 Hypothesis and
reveals the continued salience of societal and material
interests in the construction of grand strategy.

615

ht
tp

s:
//

do
i.o

rg
/1

0.
10

17
/S

00
03

05
54

03
00

09
0X

P
ub

lis
he

d
on

lin
e

by
C

am
br

id
ge

U
ni

ve
rs

ity
P

re
ss

https://doi.org/10.1017/S000305540300090X

Security and International Migration November 2003

“Securitizing” Migration

Public attitudes and government policies on both sides
of the Atlantic responded quickly to the post-9/11 secu-
rity environment. According to a Fox News poll taken
in November 2001, 65% of Americans supported stop-
ping all immigration into the country—a clear indica-
tion of just how acute perceptions of threat had become
and the clear linkage made between security and im-
migration control among the general public after 9/11.
The actions of policy makers reflect similar convictions:
Prior to 9/11, Mexico was touted as America’s pri-
mary ally, yet soon after 9/11 previous suggestions for
continued bilateral negotiations regarding amnesty for
undocumented workers and a new temporary worker
program were tabled indefinitely, much to the dismay
of Mexican president Vicente Fox.

Consistent with the 9/11 Hypothesis, the first signifi-
cant policy developments of the post-9/11 period were
restrictionist, including the USA Patriot Act (Octo-
ber 2001) and the Enhanced Border Security and Visa
Entry Reform Act (EBSVERA) (May 2002). The USA
Patriot Act called for fortifying the country’s northern
border, increased law enforcement powers for surveil-
lance and detention when dealing with suspected ter-
rorists, and increased the grounds for inadmissibility
for entry to those involved with terrorist organizations
by expanding the legal definition of “terrorist activi-
ties.” Likewise, EBSVERA stipulated an increase of
3,000 immigration inspectors and investigators, called
for increased scrutiny of visa applications originating
in countries suspected of supporting terrorism, and
required U.S. universities to better account for for-
eign students attending their institution. In addition
to border security, internal control measures were also
initiated in the name of homeland defense. In 2002,
Attorney General John Ashcroft initiated a pro-
gram requiring male foreign visitors from “politically
sensitive”—largely Muslim—countries to register with
the INS. In addition, information technology, such as
the National Automated Immigration Lookout System
(NAILS), increasingly has been utilized to properly
identify visa applicants, match names with international
databases of known or suspected terrorists and/or ter-
rorist supporters, and track the movement of foreigners
working or studying in the country (Koslowski 2002).

Initial responses in Germany, Great Britain, and
France would also seem to support the 9/11 Hypothesis.
Spanish Foreign Minister Josep Pique articulated the
post-9/11 sentiment that, “the fight against illegal im-
migration is also the reinforcement of the fight against
terrorism” (Migration News, Nov. 2001). Prior to 9/11,
the EU Commission was recommending liberalization
of immigration policies among member nations, con-
tinuing the trends evident during the late 1990s. After
September 11, most member nations immediately sus-
pended such plans, and it appeared that the new focus
of grand strategy would come in the form of new an-
titerrorism legislation with special focus on controlling
illegal immigration. Following 9/11, the governments
of Britain, France, and Germany granted additional
authority to police and intelligence agencies to monitor

and detain suspected terrorists and their associates.
In addition, the new security environment prompted
government action to better control migration. In
Britain, the Nationality, Immigration and Asylum Bill
(November 8, 2002) increased existing carrier and em-
ployers sanctions laws by increasing incarceration pe-
riods for those convicted of organizing the smuggling
of illegal immigrants into Britain. In addition, the law
restricts the appeals period for those seeking asylum
and permits the government to detain asylum seekers
for up to six months and also prohibits nationals of
EU countries (and accession countries) from applying
for asylum in the United Kingdom. The British gov-
ernment has also sought cooperation from the French
government in increasing security at the Channel Tun-
nel and has added extra police and video surveillance
cameras to this vital crossing route.

More significant, perhaps, was Britain’s success in
garnering French acceptance to close the Sangatte
Red Cross Center, a suspected staging point for ille-
gal immigrants and asylum seekers, in December 2002.
British Home Secretary David Blunkett optimistically
declared, “This agreement not only closes Sangatte. . .
[but] it will also shut off the routes used by illegal im-
migrants to get to the UK from France. It effectively
pushes our border controls across the channel to the
French coast, where stronger controls and tighter secu-
rity will mean we can prevent illegal immigrants from
getting to the UK in the first place” (M2 Presswire,
December 2, 2002). Increased French police deploy-
ment in the area resulted in the apprehension of hun-
dreds of illegal aliens, including some 250 migrant
smugglers. Cooperative security efforts related to im-
migration and border control have also increased in
Germany, where authorities have been working with
representatives of other EU countries to increase co-
operation and coordination of border policing.

Societal and Material Aspects of Policy

Although the establishment of these new security poli-
cies would seem to support the 9/11 Hypothesis, the em-
pirical evidence suggests that military security interests
do not wholly dominate migration policy as one might
predict. Societal security remains salient, and economic
interests continue to serve as a powerful counterweight
to the political interests that push for closure, though
they may have declined in relative importance after
9/11. As has long been the case, the state continues
to seek a grand strategy that best maximizes security’s
three dimensions in aggregate, a process that has be-
come increasingly difficult since 9/11.

In contrast to the 1990s, where “out of sight, out of
mind” seemed to rule the day, it is the very invisibility
of the undocumented or out-of-status migrant popu-
lation that generates vulnerability in terms of poten-
tial terrorist activity. Given the security environment
in the United States, one might expect a more expan-
sive effort to document the 8–11 million undocumented
aliens currently living in the country and to more ac-
curately regulate the movement of temporary workers.

616

ht
tp

s:
//

do
i.o

rg
/1

0.
10

17
/S

00
03

05
54

03
00

09
0X

P
ub

lis
he

d
on

lin
e

by
C

am
br

id
ge

U
ni

ve
rs

ity
P

re
ss

https://doi.org/10.1017/S000305540300090X

American Political Science Review Vol. 97, No. 4

In 2000, immigrant advocacy groups vigorously lobbied
Congress to pass the Latino and Immigrant Responsi-
bility Act, which would have established an extended
amnesty program for undocumented workers. How-
ever, consistent with the strategy employed during the
1990s, societal security interests held the day, and policy
makers were only willing to pass an extension of the
H1-B program for highly skilled workers (American
Competitiveness in the 21st Century Act). Renewing
amnesty legislation would make perfect sense post-
9/11, since it would provide a means to identify the
millions of “invisible” migrants living in the country, yet
such legislation has generated no serious political mo-
mentum. If increasing the ability to detect and track for-
eigners is a security imperative in the post-9/11 environ-
ment (consistent with the 9/11 Hypothesis), we might
also expect new efforts to establish a stricter regime
to control the movement of foreign workers, perhaps
in the form of a new guest worker program. However,
the lessons of Germany’s postwar Gastarbeiter program
(which resulted in unwanted permanent rather than
temporary migration) continue to guide government
attitudes regarding such proposals, consistent with
societal security interests. In contrast, if reluctance to
engage largely economic migration from Mexico re-
flects American ethnonationalism, the liberal aspect of
its national identity has confounded government efforts
at increasing internal monitoring and controls of for-
eign immigrants. A policy to establish a national ID
card utilizing biometric data has received little public
support and has been largely ignored by policy makers,
even though experts argue that it would be necessary to
establish the kind of monitoring necessary to maximize
internal security and thwart terrorist-aliens (consistent
with the Liberal State Hypothesis).

As in the United States, Europe’s post-9/11 grand
strategy does not appear myopic, but again reflects an
active state balancing various facets of the national in-
terest. Given the connection between terrorist cells op-
erating in the country and the events of September 11, it
is not surprising that some of the most significant policy
changes have taken place in Germany. What is surpris-
ing is that the new immigration laws signed into law in
April 2002 deal less with issues of military security than
with the continuing trend of balancing societal security
demands with macroeconomic interests. Although the
German Federal Constitutional Court in Karlsruhe de-
clared the new laws unconstitutional in December 2002,
examining the form and politics associated with the pro-
cess again illustrates how policy makers construe the
national interest. In terms of macroeconomic interests,
two factors continue to push Germany toward liber-
alization rather than closure. First, low fertility rates
(1.39 in 2002) may require that Germany initiate and
maintain relatively large levels of immigration over the
next 20 years in order to keep the social security sys-
tem solvent. Second, economists have steadily argued
that macroeconomic growth is dependent on immedi-
ate increases in the availability of highly skilled labor.
To address economic need, the laws established provi-
sions though which foreign entrepreneurs could obtain
a temporary residence permit, provided that they could

supply an initial investment of one million Euros and
that their business would create at least 10 jobs. Un-
limited residence permits would be available for highly
skilled applicants whose services were considered es-
sential to economic growth. Interior Ministry estimates
anticipated these laws would initially admit about 5,000
highly skilled workers and 500 entrepreneurs, though
no specific quota levels were included in the legislation.
The new laws also allowed foreign students who studied
at Germany universities to remain in the country and
work, providing yet another source of highly-educated
and skilled labor.

In addition to economic interests, societal security in-
terests also figured prominently in the 2002 legislation,
both ethnic-nationalist and civic-nationalist in orien-
tation. For those who have maintained that postwar
German identity must break with its ethnonationalist
past and embrace liberal ideology, these new laws—
Germany’s first immigration regime—represented the
manifestation of such ideals. Interior Minister Otto
Schily announced, “With this law Germany shows
itself to be an open country” (Migration News,
April 2002). However, the legislation also reflected an
interest in mollifying ethnonationalist societal insecu-
rities initiated by the immigration of Third Country
Nationals and the presence of a large, unnaturalized
foreign population living in Germany, many for over
a quarter century. The new laws contained provisions
to promote the integration of existing foreigners by
requiring them to take courses on the German lan-
guage, its social institutions, and its governance/legal
structure. Failure to attend such courses could result
in being denied an unlimited residence status. While
accepting the presence of the existing foreign popula-
tion can be interpreted as a remarkable transforma-
tion toward openness for a country reluctant to accept
its position as a “country of immigration,” the inclu-
sion of integration requirements provides some mea-
sure of symbolic weight for conservatives in defense of
the German Leitkultur. The heated debate regarding
a new immigration regime reveals how two visions of
German nationhood seek to defend themselves from
perceptions of societal threat and will certainly figure
prominently in the coming policy discourse.

In France, societal security has remained at the cen-
ter of political discourse even amid rising concerns
about global terrorism. Indeed, 9/11 may have in-
flamed the debate, given that the Al Queda terrorists
were from Muslim countries—the source of much of
France’s ethnonationalist societal insecurities. The un-
precedented support garnered by Jean-Marie Le Pen
in the April 2002 elections is evidence of a contin-
ued sense of ethnocultural societal threat. Such sen-
timents appeared especially strong in areas with high
concentrations of immigrants where Le Pen’s National
Front party received many of its votes. Yet, the rise
of France’s ethnonationalist underbelly has also mo-
bilized its republican liberal component. In October
2002, a rally of some 5,000 people marched through
the streets of Paris demanding residency permits for
undocumented foreigners living in France. Consistent
with the approach taken during the late 1980s and

617

ht
tp

s:
//

do
i.o

rg
/1

0.
10

17
/S

00
03

05
54

03
00

09
0X

P
ub

lis
he

d
on

lin
e

by
C

am
br

id
ge

U
ni

ve
rs

ity
P

re
ss

https://doi.org/10.1017/S000305540300090X

Security and International Migration November 2003

1990s, President Jacques Chirac has proposed integra-
tion contracts that would provide migrants with French
language instruction and job training in order to bolster
their integration into French society and reduce societal
tensions.

Likewise, Britain’s post-9/11 grand strategy also re-
flects responsiveness to societal and material interests,
not simply to military security and counter-terrorism.
In January 2002, Britain began a program to recruit
highly skilled migrants into the country. Applicants
who possess an advanced degree, have at least three
years of work experience, and can verify that they have
earned at least £40,000 a year are eligible to enter
the country without a prior job offer as required under
the normal work permit system. In terms of societal
insecurities and migration, the new British law is con-
sistent with the approach taken by its continental neigh-
bors. The Nationality, Immigration and Asylum Bill
stresses the social integration of new immigrants by re-
quiring citizenship applicants to take English language
classes and to demonstrate a thorough knowledge of
British society. As in the case with France and Germany,
new laws reflect a modest tightening of control policies
in response to new military threats associated with
migration, yet are not dominated by them, suggesting
a limited affirmation of the 9/11 Hypothesis.

Clearly, post-9/11 security measures related to mi-
gration at least partially support the 9/11 Hypothesis.
However, it is important to recognize that grand strat-
egy is not dominated by policy myopia. The evidence
also supports the Adversary Corollary of the Rally Hy-
pothesis, in that the identification of a particular ethnic
group with the source of external threat has focused
policy closure on migrants from those sources. The
widespread association of Muslim’s with the escalation
of global terrorism has sharpened perceptions of threat,
especially in Europe, where Muslims have long been
associated with societal insecurity. Conversely, the ev-
idence also suggests considerable support for the Lib-
eral State Hypothesis. Each of the cases examined here
represent national models that have both ethnic- and
civic-nationalist impulses. Restrictionism in the post-
9/11 environment has been tempered by a combination
of a reluctance to appear overtly authoritarian (and/or
racist) and the institutionalization of liberal ideals in ex-
isting policy and legal precedent. Germany’s apprehen-
sion at heavy-handed responses to counter-terrorism
and border policies in light of its desire to distance
itself from its authoritarian past is a good example.
The reluctance of American policy makers to seriously
discuss new laws requiring the creation of a national
ID card is another good example. Finally, the post-9/11
security environment has not radically altered existing
preferences for dealing with societal insecurities de-
veloped during the 1990s. There is little to suggest that
the Societal Proximity Hypothesis (and those related to
it) has lost salience, though its relative influence may
have declined somewhat since 9/11. What we see in the
post-9/11 world in similar to that which preceded it—
an active state seeking to balance interests and craft
an optimal grand strategy given a fluctuating security
environment.

CONCLUSION

In addition to the globalization of trade and capital
flows, the growing value of services and knowledge-
based capital in the emerging IT economy and the
benefits of an elastic labor supply make effective man-
agement of international migration an increasingly im-
portant component of trading state grand strategy. This
may be even more significant given a growing belief
among economists that migration complements trade—
a view that breaks with traditional beliefs regarding
the substitutability of trade for factor mobility (Collins,
O’Rourke, and Williamson 1999). Yet globalization
has served to generate forces that have created new
security interests among these countries that signif-
icantly constrain states from pursuing a grand strat-
egy wholly dominated by material interests (Rudolph
n.d.; cf. Albert et al. 2001; Lapid and Kratochwil 1996;
Lipschutz 1995).

Several important conclusions can be drawn from the
findings presented here. While grand strategy among
advanced industrial states over much of the post-WWII
period displays a neoliberal, trading state orientation,
the rise of a socially oriented counterglobalism has re-
vealed a significant political constraint on the state’s
ability to maximize material interests (Rosenau 1997).
Although the object of societal insecurities—national
identity—is both fluid and contested among liberal
democratic states, the evidence suggests that images
of control are dominant themes of societal security re-
gardless of whether ethnocultural or republican dimen-
sions of identity are emphasized. In a world where glob-
alization may be weakening elements of Westphalian
sovereignty, the politics of migration and border poli-
cies suggests the rise of “societal sovereignty,” one
based on an emphasis of authority in access to the polity
(Rudolph 2003a).

Policy developments during the 1990s represented
government attempts to address societal insecurities
through highly symbolic policies that present a strong
image of control; however, constraints on neoliberal
policies have become even more significant and in-
creasingly difficult for state’s to “finesse” as military
security interests have converged with societal interests
after 9/11. Indeed, the potential costs of the migration–
terrorism link call into question whether it is in the na-
tional interest to attempt to craft policy toward assuag-
ing fears rather than establishing stringent control over
flows. However, the current and opportunity costs—
both economic and political—of responding aggres-
sively to fears of global terrorism facilitated by interna-
tional migration will likely determine policy outcomes
and the balance between the three facets of security. For
states with large land borders, increasing the degree of
border security involves prohibitive costs. In economic
terms, these involve directs costs in the form of huge
increases in manpower, surveillance equipment, and
intelligence gathering. Making borders less porous also
presents even higher potential opportunity costs, as re-
ducing the movement of mobile factors represents a
significant negative externality on the efficiency of the
global economy. In political terms, internal control

618

ht
tp

s:
//

do
i.o

rg
/1

0.
10

17
/S

00
03

05
54

03
00

09
0X

P
ub

lis
he

d
on

lin
e

by
C

am
br

id
ge

U
ni

ve
rs

ity
P

re
ss

https://doi.org/10.1017/S000305540300090X

American Political Science Review Vol. 97, No. 4

necessarily entails a reduction in civil liberties in or-
der for governments to separate friend from foe and
to track the movement of foreigners present in the
country.

In terms of our understanding of policy development,
the evidence presented here suggests that policy is not
solely a bottom-up process of localized interests, collec-
tive action, and political mobilization. Rather, the state
is an important political agent in its own right. This is
not to say that local interests do not matter or that ef-
fective lobbying does not, at times, result in policy shifts
responding to those interests. However, the state can
serve to define issues, provide information, and shape
public discourse. It can both respond to latent interest
groups—the general public—and can also anticipate
changes in public opinion and craft policy in advanced
of demands for policy change. As a political agent, the
findings presented here also illustrate how processes
of “state learning” affect the calculation of interests
and the formulation of policy. Placed in a wider histor-
ical context, while migration has been a common phe-
nomenon throughout history, immigration and border
control are relatively new political imperatives for the
state (Torpey 2000). Because we are still learning about
the processes of migration and their economic and so-
cial impact, we might describe government responses to
these phenomena as being in a state of “policy infancy.”
As political learning develops, we see policy makers
attempting to maximize interests given imperfect in-
formation, political constraints, and limited resources.
Whether analysts typify such behavior as attempts to
“finesse” policy or simply reflect a government “mud-
dling through” a set of second-best options is simply
an issue of normative perspective. What is clear from
the standpoint of grand strategy is that the states that
can best negotiate the complex challenges posed by
migration will be those that garner the advantage in
our global age.

REFERENCES

Albert, Mathias, David Jacobson, and Yosef Lapid, eds. 2001. Iden-
tities, Borders, Orders. Minneapolis, MN: University of Minnesota
Press.

Andreas, Peter. 2000. Border Games. Ithaca, NY: Cornell University
Press.

Bade, Klaus J. 1983. Vom Auswanderungsland zum Einwanderungs-
land? Deutschland 1880–1980. Berlin: Colloquium Verlag.

Bade, Klaus J. 1987. “Transatlantic Emigration and Continental Im-
migration: The German Experience.” In Population, Labour and
Migration in the 19th- and 20th-Century Germany, ed. K. J. Bade.
New York: St. Martin’s Press.

Bean, Frank D., Robert G. Cushing, and Charles W. Haynes. 1997.
“The Changing Demography of U.S. lmmigration Flows: Pat-
terns, Projections, contexts.” In Migration Fast, Migration Future:
Germany and the United States, eds. K. Bade and M. Weiner. Prov-
idence, RI: Berghahn Books.

Bhagwati, Jagdish. 1998. A Stream of Windows. Cambridge, MA: MIT
Press.

Borstelmann, Thomas. 2002. The Cold War and the Color Line: Amer-
ican Race Relations in the Global Arena. Cambridge, MA: Harvard
University Press.

Brubaker, Rogers. 1992. Citizenship and Nationhood in France and
Germany. Cambridge, MA: Harvard University Press.

Castells, Manuel. 1975. Immigrant Workers and Class Struggles in
Advanced Capitalism. Politics and Society 5 (1): 33–66.

Castles, Stephen, and Godula Kosack. 1973. Immigrant Workers and
Class Structure in Western Europe. Oxford: Oxford University
Press.

Chang, Howard F. 1998. “Migration as International Trade.” In Re-
considering Immigration in an Integrating World, ed. C. Rudolph.
UCLA Journal of International Law & Foreign Affairs 3 (2): 371–
414.

Collins, William J., Kevin O’Rourke, and Jeffrey G. Williamson.
1999. “Were Trade and Factor Mobility Substitutes in History?”
In Migration: The Controversies and the Evidence, ed. R. Faini, J.
de Melo, and K. Zimmermann. Cambridge: Cambridge University
Press.

Cornelius, Wayne A. 1998a. The Role of Immigrant Labor in the U.S.
and Japanese Economies. La Jolla, CA: Center for U.S.–Mexican
Studies, UCSD.

Cornelius, Wayne A. 1998b. “Appearances and Realities: Controlling
Illegal Immigration in the United States.” In Temporary Work-
ers or Future Citizens? ed. M. Weiner and T. Hanami. London:
Macmillan.

Cornelius, Wayne A., Thomas Espenshade, and Idean Salehyan, eds.
2001. The International Migration of the Highly-Skilled. La Jolla,
CA: CCIS.

Cornelius, Wayne A., Philip L. Martin, and James F. Hollifield, eds.
1994. Controlling Immigration: A Global Perspective. Stanford,
CA: Stanford University Press.

Fetzer, Joel S. 2000. Public Attitudes Toward Immigration in the
United States, France, and Germany. Cambridge: Cambridge Uni-
versity Press.

Freeman, Gary P. 1995. “Modes of Immigration Politics in Liberal
Democratic States.” International Migration Review 26 (4): 881–
913.

Freeman, Gary P. 1979. Immigrant Labor and Racial Conflict in In-
dustrial Societies. Princeton, NJ: Princeton University Press.

Ghosh, Bimal, ed. 2000. Managing Migration. Oxford: Oxford Uni-
versity Press.

Gimpel, James G., and James R. Edwards, Jr. 1999. The Congressional
Politics of Immigration Reform. Boston: Allyn and Bacon.

Goldstein, Judith. 1986. “The Political Economy of Trade: Institu-
tions of Protection.” American Political Science Review 80 (1):
161–84.

Goulbourne, Harry. 1998. Race Relations in Britain Since 1945.
New York: St. Martin’s Press.

Hansen, Randall. 2000. Citizenship and Immigration in Post-War
Britain. Oxford: Oxford University Press.

Hargreaves, Alec G. 1995. Immigration, “Race” and Ethnicity in Con-
temporary France. New York: Routledge.

Hatton, Timothy J., and Jeffrey G. Williamson. 1998. The Age of Mass
Migration. New York: Oxford University Press.

Heckmann, Friedrich. 1985. “Temporary Labor Migration or Immi-
gration? “Guest Workers” in the Federal Republic of Germany.”
In The Effects of European Labor Migration on Sending and Re-
ceiving Countries, ed. R. Rogers. Boulder, CO: Westview.

Heinlein, Frank. 2002. British Government Policy and Decoloniza-
tion, 1945–1963. London: Frank Cass.

Heisler, Martin. 1998. “Contextualizing Global Migration: Sketching
the Socio-Political Landscape in Europe.” In Reconsidering Immi-
gration in an Integrating World, ed. C. Rudolph. UCLA Journal of
International Law & Foreign Affairs 3 (2): 557–593.

Herbert, Ulrich. 1990. A History of Foreign Labor in Germany,
1880–1980. Trans. William Templer. Ann Arbor, MI: University of
Michigan Press.

Hollifield, James F. 1992. Immigrants, Markets, and States. Cambridge,
MA: Harvard University Press.

Hollifield, James F. 1994. “Immigration and Republicanism in France:
The Hidden Consensus.” In Controlling Immigration, ed. W. A.
Cornelius et al. Stanford, CA: Stanford University Press.

Hollifield, James F. 1998. “Migration, Trade, and the Nation-
State.” In Reconsidering Immigration in an Integrating World, ed.
C. Rudolph. UCLA Journal of International Law & Foreign Affairs
3 (2): 595–636.

Hollifield, James F., and Gary Zuk. 1998. “Immigrants, Markets,
and Rights.” In Immigration, Citizenship, and the Welfare State in
Germany and the United States, ed. H. Kurthen, J. Fijalkowski, and
G. Wagner. Stanford, CT: JAI Press.

Holmes, C. 1988. John Bull’s Island: Immigration and British Society,
1871–1971. London: Macmillan.

619

ht
tp

s:
//

do
i.o

rg
/1

0.
10

17
/S

00
03

05
54

03
00

09
0X

P
ub

lis
he

d
on

lin
e

by
C

am
br

id
ge

U
ni

ve
rs

ity
P

re
ss

https://doi.org/10.1017/S000305540300090X

Security and International Migration November 2003

INSEE. 1985. Recesement general de la population de 1982: les
etrangers. Paris: INSEE.

INSEE. 1992. Recensement general de la population de 1990: nation-
alités, resultats du sontage au quart. Paris: INSEE.

Joppke, Christian. 1997. “Asylum and State Sovereignty: A Com-
parison of the United States, Germany, and Britain.” Comparative
Political Studies 30 (3): 259–98.

Joppke, Christian. 1999. Immigration and the Nation State. Oxford:
Oxford University Press.

Kennedy, Pau M. 1989. The Rise and Fall of the Great Powers. New
York: Vintage Books.

Kessler, Alan E. 1999. “Guarded Gates: Factor Mobility, Domestic
Coalitions, and the Political Economy of American Immigration
Control.” Ph.D. diss. UCLA.

Kindleberger, Charles P. 1967. Europe’s Postwar Growth. Cambridge,
MA: Harvard University Press.

King, Desmond. 2000. Making Americans. Oxford: Oxford Univer-
sity Press.

Korte, Hermann. 1985. “Labor Migration and the Employment of
Foreigners in the Federal Republic of Germany Since 1950.” In
Guests Come to Stay, ed. R. Rogers. Boulder, CO: Westview.

Koslowski, Rey. 2000. Migrants and Citizens. Ithaca, NY: Cornell
University Press.

Koslowski, Rey. 2002. “Information Technology, Migration and Bor-
der Control.” Presented at the Institute for Governmental Studies,
University of California, Berkeley.

Krozewski, Gerold. 2001. Money and the End of Empire. New York:
Palgrave.

Lapid, Yosef, and Friedrich Kratochwil, eds. 1996. The Return of
Culture and identity in IR Theory. Boulder, CO: Lynne Rienner.

Layton-Henry, Zig. 1984. The Politics of Race in Britain. London:
Allen & Unwin.

Layton-Henry, Zig. 1994. “Britain: The Would-Be Zero-Immigration
Country.” In Controlling Immigration, ed. W. A. Cornelius et al.
Stanford, CA: Stanford University Press.

Lipschutz, Ronnie D. 1995. On Security. New York: Columbia Uni-
versity Press.

Martin, Philip L. 1994. “Germany: Reluctant Land of Immigra-
tion.” In Controlling Immigration: A Global Perspective, ed. W. A.
Cornelius et al. Stanford, CA: Stanford University Press.

Massey, Douglas S. 1999. “International Migration at the Dawn of
the Twenty-First Century: The Role of the State.” Population and
Development Review 25 (2): 303–22.

Massey, Douglas S., and Felipe Garcia Espana. 1987. “The Social
Process of International Migration.” Science 237: 733–38.

Massey, Douglas S., et al. 1998. Worlds in Motion. Oxford: Oxford
University Press.

Migration News. Various dates.
Money, Jeannette. 1999. Fences and Neighbors: The Political Geogra-

phy of Immigration Control. Ithaca, NY: Cornell University Press.
M2 Presswire.
Newland, Kathleen, and Demetrios Papademetriou. 1998. “Manag-

ing International Migration: Tracking the Emergence of a New
International Regime.” In Reconsidering Immigration in an Inte-
grating World, ed. C. Rudolph. UCLA Journal of International Law
& Foreign Affairs 3 (2): 637–57.

Ogden, Philip E. 1989. “International Migration in the Nineteenth
and Twentieth Centuries.” In Migrants in Modern France, ed.
P. Ogden and P. White. London: Unwin–Hyman.

Oriol, Michel. 1992. “Islam and Catholicism in French Immigra-
tion.” In Immigration in Two Democracies: French and American
Experiences, ed. D. Horowitz and G. Noiriel. New York: NYU
Press.

O’Rourke, Kevin H., and Jeffrey G. Williamson. 1999. Globalization
and History. Cambridge, MA: MIT Press.

Papademetriou, Demetrios G. 1996. Coming Together or Pulling
Apart? Washington, DC: Carnegie Endowment for International
Peace.

Patterson, Shiela. 1969. Immigration and Race Relations in Britain,
1960–1967. London: Oxford University Press.

Paul, Kathleen. 1997. Whitewashing Britain. Ithaca, NY: Cornell Uni-
versity Press.

Potts, Lydia. 1990. The World Labour Market: A History of Migration.
Trans. Terry Bond. London: Zed Books.

Reimers, David M. 1985. Still the Golden Door: The Third World
Comes to America. New York: Columbia University Press.

Rolph, Elizabeth S. 1992. Immigration Policies: Legacy from the 1980s
and Issues for the 1990s. Santa Monica, CA: RAND.

Rosecrance, Richard. 1986. The Rise of the Trading State. New York:
Basic Books.

Rosenau, James N. 1997. Along the Domestic-Foreign Frontier: Ex-
ploring Governance in a Turbulent World. Cambridge: Cambridge
University Press.

Rosenblum, Marc R. 2003. “Congress, the President, and the INS:
Who’s in Charge of U.S. Immigration Policy?” Typescript. Univer-
sity of New Orleans.

Roy, O. 1994. “Islam in France: Religion, Ethnic Community or
Ethnic Ghetto?” In Muslims in Europe, ed. B. Lewis and Dom-
inque Schnapper. London: Pinter.

Rudolph, Christopher. 2003a. “Sovereignty and Territorial Bor-
ders in a Global Age.” Presented at the Annual Meeting of the
American Political Science Association, Philadelphia.

Rudolph, Christopher. 2003b. National Security and International Mi-
gration. Typescript. UCLA.

Rudolph, Christopher. N. D. “Globalization and Security: Migration
and Evolving Conceptions of Security in Statecraft and Scholar-
ship.” Security Studies. Forthcoming.

Shanks, Cheryl. 2001. Immigration and the Politics of American
Sovereignty, 1890–1990. Ann Arbor: University of Michigan Press.

Silverman, Maxim. 1992. Deconstructing the Nation. London:
Routledge.

Smith, Anthony D. 1986. The Ethnic Origins of Nations. Oxford: Basil
Blackwell.

Smith, Anthony D. 1991. National Identity. Reno: University of
Nevada Press.

Smith, Rogers M. 1997. Civic Ideals. New Haven, CT: Yale University
Press.

Spencer, Ian R. G. 1997. British Immigration Policy Since 1939.
London: Routledge.

Tapinos, Georges. 1975. “L’immigration étrangeré en France, 1946–
1973.” Institut National d’Etudes Demographiques No. 7. Paris:
Presses Universitaires de France.

Torpey, John. 2000. “States and the Regulation of Migration in the
Twentieth-Century North Atlantic World.” In The Wall Around
the West, ed. P. Andreas and T. Snyder. Lanham, Md.: Rowman &
Littlefield.

Ulrich, Ralf E. 1994. “Vertriebene and Aussiedler—The Immigration
of Ethnic Germans.” In The Economic Consequences of Immigra-
tion to Germany, ed. G. Steinmann and Ralf E. Ulrich. Heidelberg:
Physica-Verlag.

U.S. Immigration and Nationalization Service. 1959. Annual Report.
Washington, DC: U.S. Government Printing Office.

U.S. Immigration and Nationalization Service. 1998. Statistical Year-
book. Washington, DC: GPO.

Wæver, Ole. 1993. “Societal Security: The Concept.” In Identity, Mi-
gration, and the New Security Agenda in Europe, ed. O. Wæver
et al. New York: St. Martin’s Press.

Wæver, Ole, Barry Buzan, Morten Kelstrup, and Pierre Lemaitre,
eds. 1993. Identity, Migration, and the New Security Agenda in
Europe. New York: St. Martin’s.

Walt, Stephen M. 1987. The Origins of Alliances. Ithaca, NY: Cornell
University Press.

Weight, Richard. 2002. Patriots: National Identity in Britain, 1940–
2000. London: Macmillan.

Weil, Patrick. 1991. La France et ses étrangers. Paris: Calmann–Levy.
Weiner, Myron, and Sharon Stanton Russell, eds. 2001. Demography

and National Security. New York: Berghahn Books.
White, Paul. 1995. “Immigrants and the Social Geography of

European Cities.” In Mass Migration in Europe, ed. R. King. New
York: John Wiley & Sons.

World Bank. 1998–99. World Development Report 1998–99.
Washington, DC: World Bank.

World Trade Organization. 2000. International Trade Statistics 2000.
Geneva: WTO.

Zolberg, Aristide R. 1995. “From Invitation to Interdiction: U.S. For-
eign Policy and Immigration Since 1945.” In Threatened Peoples,
Threatened Borders, ed. M. Teitelbaum and M. Weiner. New York:
W. W. Norton.

620

ht
tp

s:
//

do
i.o

rg
/1

0.
10

17
/S

00
03

05
54

03
00

09
0X

P
ub

lis
he

d
on

lin
e

by
C

am
br

id
ge

U
ni

ve
rs

ity
P

re
ss

https://doi.org/10.1017/S000305540300090X

Working Papers

Paper 32, April 2011

The determinants of

international migration

Conceptualising policy, origin and
destination effects

Hein de Haas

DEMIG project paper 2

The research leading to these results is part of the

DEMIG project and has received funding from the

European Research Council under the European

Community’s Seventh Framework Programme

(FP7/2007-2013)/ERC Grant Agreement 240940.

www.migrationdeterminants.eu

This paper is published by the International Migration Institute (IMI), Oxford Department of International Development

(QEH), University of Oxford, 3 Mansfield Road, Oxford OX1 3TB, UK (www.imi.ox.ac.uk). IMI does not have an institutional

view and does not aim to present one. The views expressed in this document are those of its independent authors.

http://www.migrationdeterminants.eu/

2 IMI Working Papers Series 2011, No. 32

The IMI Working Papers Series

IMI has been publishing working papers since its foundation in 2006. The series presents
current research in the field of international

migration.

The papers in this series:

analyse migration as part of broader global change

contribute to new theoretical approaches

advance understanding of the multi-level forces driving migration

Abstract

The effectiveness of migration policies has been widely contested in the face of their
supposed failure to steer immigration and their hypothesized unintended, counter-
productive effects. However, due to fundamental methodological and conceptual
limitations, evidence has remained inconclusive. While the migration policy research is often
descriptive and receiving-country biased, migration determinants research tends to be
based on obsolete, theoretically void push-pull and gravity models which tend to omit
crucial non-economic, sending-country and policy factors. More fundamentally, this state-
of-the-art reveals a still limited understanding of the forces driving migration. Although
there is consensus that macro-contextual economic and political factors and meso-level
factors such as networks all play ‘some’ role, there is no agreement on their relative weight
and mutual interaction. To start filling that gap, this paper outlines the contours of a
conceptual framework for generating improved insights into the ways states and policies
shape migration processes in their interaction with structural migration determinants in
receiving and sending countries. First, it argues that the fragmented insights from different
disciplinary theories can be integrated in one framework through conceptualizing virtually
all forms of migration as a function of capabilities and aspirations. Second, to increase
conceptual clarity it distinguishes the preponderant role of states in migration processes
from the hypothetically more marginal role of specific immigration and emigration policies.
Subsequently, it hypothesizes four different (spatial, categorical, inter-temporal, reverse
flow) ‘substitution effects’ which can partly explain why polices fail to meet their objectives.
This framework will serve as a conceptual guide for the DEMIG (The Determinants of
International Migration) research project.

Keywords: migration, migration determinants, theory, structure, agency, migration policy,
states, effectiveness, substitution effects

Author: Hein de Haas, Senior Research Officer, International Migration Institute, University
of Oxford, hein.dehaas@qeh.ox.ac.uk.

IMI Working Papers Series 2011, No. 32 3

Contents

Introduction ………………………………………………………………………………………………………………. 4

1 Research questions and approach ……………………………………………………………………………… 6

2 Theories on the causes of migration ………………………………………………………………………….. 8

2.1 Functionalist migration theories ………………………………………………………………………….. 8

2.2 Conflict theory: bringing in structure …………………………………………………………………… 9

2.3 Filling the sending-country gap ………………………………………………………………………….. 13

2.4 Bridging non-economic and economic migration determinants …………………………….. 14

3 Towards theoretical synthesis: an aspirations-capabilities framework …………………………. 15

4 Linking micro and macro levels ………………………………………………………………………………… 22

5 Theorizing the role of states and policies in migration processes ………………………………… 23

6 Hypothesizing substitution effects of migration policies …………………………………………….. 25

Conclusion ……………………………………………………………………………………………………………….. 28

References ………………………………………………………………………………………………………………. 31

4 IMI Working Papers Series 2011, No. 32

Introduction1

How do the migration policies of receiving and sending states affect the size, direction and
nature of international migration? This is the central question addressed by the DEMIG
research project (The Determinants of International Migration: A theoretical and empirical
assessment of policy, origin, and destination effects). This project aims to generate new
theoretical and empirical insights into the way states and policies shape migration processes
in their interaction with other migration determinants in receiving and sending countries.
First, this paper outlines the scientific rationale of this project by analysing the main gaps in
migration policy and migration determinants research. Second, it outlines the contours of a
conceptual framework and a set of hypotheses for generating improved insights into the
ways states and policies shape migration processes in their interaction with structural
migration determinants in receiving and sending countries, which will guide the DEMIG
project.

In wealthy countries, immigration, in particular of low-skilled and culturally distinct
people from poorer countries, is increasingly perceived as a problem in need of control. The
common – but not unproblematic – perception is that policy-makers have reacted to this
pressure by implementing restrictive immigration policies and increasing border controls
(Castles and Miller 2009; Massey et al. 1998). However, the effectiveness of such policies
has been often contested in the face of their oft-supposed failure to significantly affect the
level of immigration and their hypothesized unintended, perverse and often counter-
productive effects such as pushing migrants into permanent settlement, discouraging return
and encouraging irregular movements and migration through alternative legal or
geographical channels (Castles 2004b; de Haas 2007; Grütters 2003). However, other
scholars have argued that, on the whole, state policies have been largely effective
(Brochmann and Hammar 1999; Collyer 2006; Strikwerda 1999), which also seems to be
partly confirmed by a limited number of quantitative studies indicating that specific policy
interventions can have a significant effect on migration flows (Czaika and de Haas 2011a).

Despite apparently increasing immigration restrictions, the volume of South–North
migration has only increased over the past few decades. But does this mean that migration
policies have failed and that states are generally unable to control migration? Not
necessarily. First of all, we should not confuse statistical association with causality, which is
particularly difficult to establish because we generally lack counterfactual cases. For
instance, one might argue that the migration-reducing effects of immigration restrictions
are counterbalanced by the migration-increasing effects of growing economic gaps between
sending and receiving countries or economic growth in receiving countries, or the lifting of
exit restrictions by origin countries (cf. de Haas and Vezzoli 2011). Hence, sustained or
increasing migration does not necessarily prove policy ineffectiveness – as migration
volumes might have been higher without migration controls. The other way around, a
decrease in migration does not prove the policy successful – although politicians are
generally eager to make such claims – as such a decrease might for instance also be the
result of economic growth or an end of conflict in origin countries, or an economic recession
in destination countries. So, finding better methodological approaches to establish
(multiple) causality constitutes the first challenge facing research on this issue.

1
I would like to thank Valentin Danchev and Simona Vezzoli for their valuable feedback on an earlier version of

this paper.

IMI Working Papers Series 2011, No. 32 5

Besides the huge difficulties involved in ‘proving’ causality as such, a second challenge
is to bring more precision in research by assessing the relative importance of immigration
policies compared to the effects of other migration determinants. After all, it can hardly be
surprising that most policies discouraging or encouraging particular manifestations of
migration will have ‘some’ effect. The real question is about the relative magnitude of this
effect compared to macro-contextual migration determinants, which will eventually also
determine the effectiveness and efficiency of policies. Although some studies assert a
statistical relation between certain policy measures and particular migration flows, the
relative importance of policy effects compared to the effects of other migration
determinants remains largely unclear. It is one thing to find that restrictions on, say, low-
skilled labour migration have a significant effect on decreasing inflows, but the real question
is how large this effect is compared to the effect of other factors such as economic growth,
employment, violent political conflict and personal freedoms. If the latter factors explain
most variance in migration, one might for instance conclude that policies have a certain, but
also limited effect on overall volumes and trends of migration. In other words, if most
variance in migration is explained by structural migration determinants or other (e.g. labour
market or macro-economic) policies, the margin of manoeuvre for migration policies is
fundamentally limited.

In addition to finding better ways to measure the existence and relative magnitude of
policy effects, a third, related, challenge is to improve insights into the very nature and
evolution of migration policies. There seems to be reason to question the general assertion
that migration policies have become more restrictive over the past decades. Although this
idea is often taken for granted, the diverse and multiple nature of migration policies raises
questions about our ability and utility to measure ‘overall’ levels of restrictiveness, and even
about the overall assumption that policies have become more restrictive. While several
countries have raised barriers for particular categories of migrants (for instance, low-skilled
workers and asylum seekers), not all countries have done so, and immigration of other
categories (for instance, family migrants and high-skilled workers) has often been facilitated.
Changes in migration policy typically facilitate the entry of particular origin groups while
simultaneously restricting the entry of other groups. For instance, ‘Fortress Europe’ may be
an adequate metaphor to characterize policies towards asylum seekers and refugees
(Hatton 2004), but seems inappropriate to characterize the immigration policies of EU or
OECD countries as a whole (cf. Czaika and de Haas 2011a).

Another example is the US Immigration and Nationality Act of 1965, which ‘equalized’
immigration policies by ending positive discrimination of European immigrants and
contributing to increasing non-European migration. This also reveals the strong Eurocentric
bias underlying common views that migration to the USA, Canada, Australia and New
Zealand was largely ‘free’ until at least the mid twentieth century (cf. Hatton and Williamson
1998) – it may have been relatively free for Europeans, but this was certainly not the case
for Asians or Africans, for whom recent reforms have meant a liberalization. Also countries’
membership and accession to regional blocks such as the European Union typically coincides
with liberalization of migration of citizens of member states, while immigration restrictions
for ‘third-country’ nationals are sustained or further tightened. Because migration policies
typically consist of a ‘mixed bag’ of various measures targeting particular groups of
immigrants, there is a considerable risk of over-generalizing. While migration policies are
likely to affect patterns of migration selectivity, the impact on the overall magnitude of

6 IMI Working Papers Series 2011, No. 32

migration flows is more uncertain as these are strongly affected by other macro-structural
factors, while migrants’ agency and strategies tend to create meso-level structures (such as
migrant networks) which facilitate migration over formally closed borders. Since state
policies simultaneously constrain or enable immigration and emigration of particular (age,
gender, skill, ethnic, regional) groups along particular geographical pathways, states perhaps
play a more significant role in structuring emigration through influencing the (initial)
composition and spatial patterns of migration, rather than in affecting overall volumes and
long-term trends, which, particularly in liberal democracies, appear to be primarily affected
by other, economic, social and cultural migration determinants.

These examples show that any serious inquiry into the effect of migration policies not
only needs to define the concept, but also to ‘unpack’ or disaggregate ‘migration policies’
into the multitude of laws, measures and regulations states deploy in their attempts to
regulate immigration and emigration along categories that are based on national origin and
further characteristics such as gender, age, education, occupation and officially defined
main migration motives (e.g. labour, refugee, family, student). As migration policies are
typically affected and shaped by different, often opposed, interests, policies are typically
internally incoherent, which further emphasizes the need to break down policies into the
specific measures and regulations they comprise.

In addition, conventional views of increasing migration policy restrictiveness typically
ignore emigration policies pursued by origin states, which are as diverse and multiple as
immigration policies, but which seem to have become less restrictive overall. Only a
declining number of strong, authoritarian states with closed economies are willing and
capable of imposing blanket exit restrictions. Paradoxically, while an increasing number of,
particularly developing, countries seem to aspire to regulate emigration, their capability to
do so is fundamentally and increasingly limited by legal (human rights), economic and
political constraints (de Haas and Vezzoli 2011). The ability of governments to affect overall
immigration and emigration levels seems to decrease as the level of authoritarianism goes
down. This also reveals the need to look beyond the role of migration policies per se and to
explore the ways in which states affect the migration process more generally.

1 Research questions and approach

So, the crucial question remains: how do states and policies shape migration processes
independently of and in their interaction with other migration determinants in receiving and
sending countries? This question can be disaggregated into three fundamental, interrelated
sub-questions:

1. What are the main determinants of international migration?
2. What has been the nature and evolution of migration policies?
3. How do states and policies affect migration flows independently of and in

interaction with other migration determinants?

Due to serious methodological and theoretical flaws, scholarly research has so far hardly
been able to produce convincing answers to these questions, and the second and third
questions in particular. The inconclusive nature of this debate reveals an overall lack of
conceptual, analytical and empirical rigour in the study of migration policy effects. Most
existing evidence is descriptive, biased and partial, which is related to the weak embedding
of migration policies research into general theories on the causes of migration.

IMI Working Papers Series 2011, No. 32 7

In this context, it is important to emphasize that the limited of capacity of research to
answer these key questions is not exclusively linked to limitations of data and statistical
models (which usually receive the blame), but also to the rather weak theoretical
foundations of ‘push-pull’ or gravity models which are routinely, but uncritically, used for
studying migration determinants. For the very reason that they are often not grounded in
migration theory, they tend to ignore or fail to properly specify several theoretically
important migration determinants in receiving and, particularly, sending countries. Even
with ideal data, statistical analyses will not lead to compelling evidence if theoretically
relevant migration determinants are omitted in empirical models, or if models are based on
the short term or only focus on one particular migration flow (e.g. asylum seekers). This
makes it impossible to study possible knock-on effects or what I have dubbed ‘substitution
effects’ (see below) of one particular measure through the diversion of migration flows to
other geographical, legal or illegal channels.

In order to improve insights into the role of states and policies in migration processes,
there is a need to embed the systematic analysis of policy effects into a comprehensive
analytical framework of the sending- and receiving-country factors driving international
migration. Although there is consensus that macro-contextual economic and political factors
and meso-level factors such as networks all play ‘some’ role, there is no agreement on their
relative weight and mutual interaction. How do migration policies precisely affect migration
if we control for the many other factors that drive international migration? Or, to turn the
question around: how do macro-level processes such as ‘development’, economic growth,
demographic change, education, democratization and conflict in origin and destination
countries affect migration independently from policy interventions? In other words, what
are the constraints and relative margins within which migration policies can have an effect?

Why has research on this issue hardly advanced over the past decades? A first
problem is the rather weak connection between studies on migration policies and migration
determinants on the one hand and fundamental research and theories on the causes of
migration on the other. A second problem is that fundamental theoretical research on the
nature and causes of migration processes has made relatively little progress over the last
few decades (cf. Arango 2000; Bakewell 2010; Massey et al. 1998). There is a plethora of
research on the social, cultural and economic impacts of migration on sending and,
particularly, receiving societies. In comparison, and with the possible exception of research
on migration networks, there has been much less theoretically driven research on the
nature and causes of migration processes themselves. This particularly applies to the study
of the precise role of policies and states in migration processes. Other factors obstructing
advances in research on migration determinants are data problems and unproductive
divisions between, particularly economic and non-economic, social science disciplines (cf.
Boswell 2008) as well as qualitative and quantitative approaches.2

2
The lack of theoretical progress might also be due to the fact that the very term ‘cause’ in social sciences

refers to explanations that are time-space independent; and often independent from people’s perception and
interpretation. Hence, conventional understandings of causality might preclude migration studies from
developing new theories. Research has shown that migration is a patterned process, not a random one, and is
also (historically and structurally) contingent. For example, functionalism may try to see how migration realize
some social functions. This also raises the question whether the available theories on ‘causes’ can be readily
used or if we need new to elaborate new understandings of causality and behaviour that can accommodate
agency and contingency. Therefore, strictly speaking, it may seem more appropriate to refer to ‘theories that

8 IMI Working Papers Series 2011, No. 32

To start filling these research gaps, in this paper I aim to outline the contours of a
theoretical and empirical research agenda for generating improved insights into the ways
states and policies shape migration processes independently from and in their interaction
with other migration determinants in receiving and sending countries. First, I will review
existing, often disciplinary, theories on migration and I will argue how their fragmented
insights can be integrated in one framework through conceptualizing virtually all
manifestations of (internal and international) migration as a function of capabilities and
aspirations to migrate. Second, I will argue that considerable conceptual confusion can be
removed if we distinguish the preponderant role of states in migration processes from the
hypothetically more marginal role of specific immigration and emigration policies.
Subsequently, based on a brief theorization of the role of states and policies in migration I
will hypothesize four different (spatial, categorical, inter-temporal, reverse flow)
‘substitution effects’ explaining migration policy failure, which can guide further research on
migration determinants within and outside the context of the DEMIG project.

2 Theories on the causes of migration

2.1 Functionalist migration theories

The preceding analysis has indicated that a robust analysis of the role of states and policies
in migration processes is conditional on its sound embedding within a more general
theoretical framework on the determinants of migration processes. Although there is a
quantitative, generally econometrically oriented literature on migration determinants
including some studies on the effect of policies (for a review, see Czaika and de Haas 2011a),
the literature is generally characterized by a conspicuous ignorance of insights from recent
migration theories. Hence, migration determinants research is generally based on obsolete,
theoretically void ‘push-pull’ and gravity models.

Implicitly or explicitly, gravity and push-pull models are rooted into functionalist social
theory. Functionalist social theory tends to see society as a system – or an aggregate of
interdependent parts, with a tendency towards equilibrium. This perspective, in which
people are expected to move from low-income to high-income areas, has remained
dominant in migration studies since Ravenstein (1885; 1889) formulated his laws of
migration. The idea that migration is a function of spatial disequilibria constitutes the
cornerstone assumption of so-called ‘push-pull’ models which still dominate much gravity-
based migration modelling as well as common-sensical and non-specialist academic thinking
about migration. Push-pull models usually identify various economic, environmental, and
demographic factors which are assumed to push migrants out of places of origin and lure
them into destination places. While deeply rooted in functionalist, equilibrium thinking, it is
difficult to classify push-pull models a theory because they tend merely to specify a rather
ambiguous list of factors that play ‘a’ role in migration. Push-pull models tend to be static
and tend to portray migrants as ‘passive pawns’ lacking any agency (which can perhaps be
defined as the ability of people to make independent choices – to act or not act in specific

attempt to explain migration’ than to ‘theories on the causes of migration’, because there are different forms
of explanations and only some of them refer to causes. Furthermore, ‘causes’ convey a rather universalistic,
static and ‘sterile’ view which largely rules out agency and contingency. I am grateful to Valentin Danchev for
drawing my attention to some of these issues.

IMI Working Papers Series 2011, No. 32 9

ways – and, crucially, to alter structure) and fail to conceptualize migration as a process (de
Haas 2010a).3

Neo-classical migration theory is the best known and most sophisticated application of
the functionalist social scientific paradigm in migration studies. At the macro-level, neo-
classical economic theory explains migration by geographical differences in the supply and
demand for labour. At the micro-level, neo-classical migration theory views migrants as
individual, rational and income-maximizing actors, who decide to move on the basis of a
cost-benefit calculation. Assuming free choice and full access to information, they are
expected to go where they can be the most productive, that is, where they are able to earn
the highest wages. Todaro (1969) and Harris and Todaro (1970) elaborated the basic two-
sector model of rural-to-urban migration, explaining migration on the basis of ‘expected
income’ (adjusted for the likelihood of unemployment) differentials. The initial Harris-
Todaro model for internal migration has, with some modifications, also been applied to
international migration (Borjas 1989; Borjas 1990). Later modifications of the neo-classical
model included the costs and risks of migration, and conceptualized migration as an
investment in human capital in order to explain migration selectivity (Bauer and
Zimmermann 1998; Sjaastad 1962).

Neo-classical and other equilibrium migration models largely explain migration by
geographical differences in (expected) incomes and wage levels (Harris and Todaro 1970;
Lee 1966; Todaro 1969). Although it would be hard to deny that economic differentials play
a major role in driving migration processes, this almost sounds more like a truism or
assumption than a theory. Furthermore, this basic insight alone is insufficient to explain the
strongly patterned, non-random nature of real-life migration processes. For instance, these
models have difficulties explaining return migration, migration in the absence of wage
differentials and, particularly, adequately grasping the role of states, networks and other
institutions in structuring migration. They also largely ignore non-economic migration
drivers and typically fail to explain development-driven increases in migration.

2.2 Conflict theory: bringing in structure

Other theories of migration reject the underlying functionalist assumption of conventional
neo-classical models that migration decisions are based on the rational cost-benefit
calculation of income-maximizing individuals operating in well-functioning markets. The new
economics of labour migration (NELM) hypothesizes that migration, particularly under
conditions of poverty and risk, is difficult to explain within a neo-classical framework. NELM
conceptualizes migration as a collective household strategy to overcome market failures and
spread income risks rather than a mere response of income-maximizing individuals to
expected wage differentials (Stark 1991; Stark and Bloom 1985; Taylor 1999). This gives
considerable theoretical room to explain migration in the absence of significant wage
differentials. NELM also argues that income inequality and relative deprivation within
sending societies are major drivers of migration (Skeldon 2002; Stark and Taylor 1989).
Through remittances, migration can also be a livelihood strategy used by families and

3
Agency can simultaneously reproduce and transform structure. In their seminal review of the concept,

Emirbayer and Mische (1998) provide a more precise definition of agency as ‘the temporally constructed
engagement by actors of different structural environments—the temporal-relational contexts of action—
which, through the interplay of habit, imagination, and judgment, both reproduces and transforms those
structures in interactive response to the problems posed by changing historical situations’.

10 IMI Working Papers Series 2011, No. 32

households to raise investment capital if credit markets fail. Within a broader social
scientific perspective, it is possible to reinterpret NELM as a theory that explains migration
as an active attempt – an act of agency – by social groups to overcome structural
constraints. An important methodological inference of these ‘new’ theories is that market
access, income inequality, relative deprivation, and social security are important migration
determinants, and need to be included in empirical models if possible.

NELM-inspired migration theory seems particularly relevant for explaining migration in
developing countries and other situations in which migrants face considerable constraint
and risk, and therefore also seems applicable to ‘non-labour’ forms of migration, such as
refugee migration (de Haas 2010a; Lindley 2007). This points to a more general weakness of
conventional ways of classifying migration into distinct types (labour/economic, refugee,
family; or voluntary vs. non-voluntary) and the concomitant tendency to develop separate
theories for them. This is deeply problematic, as these migration types reflect legal rather
than sociological categories. These categorizations ignore empirical evidence that migration
is typically driven by a range of contextual factors and that individual motivations to migrate
are often mixed. This makes strict distinctions such as between voluntary and forced
migration, or between family and labour migration, often deeply problematic. This seems
certainly to be the case in the context of restrictive immigration policies, in which
prospective migrants perceived policies as opportunity structures within which the choice of
migration channel is likely to be based on relative ease and costs rather than on a
consideration of which category best matches their ‘genuine’ migration motives.

While some would still classify NELM as an amended form of neo-classical theory,4 a
more profound critique of neo-classical and push-pull migration theories would stress their
a-historical nature and their failure to conceptualize how macro-structural factors such as
states, policies, labour markets, status hierarchies, power inequalities and social group
formation (including migrant networks) strongly constrain individual choice and explain why
most migration tends to occur in socially selective and geographically strongly patterned
ways; that is, along well-defined pathways or corridors between particular origins and
destinations. Conventional economic models usually incorporate structural factors as
additional costs and risks individuals face. It certainly does make sense to assume that
structural constraints affect the cost-benefit calculus and destination choice. However, the
reduction of such factors to individual costs and benefits makes such models inherently
blind to the very structural features of such factors, which can only be analysed on the
group (family, community, society) level as they are embedded in and reproduced by
patterns of relations between people. Despite the considerable merits of neo-classical
approaches, their methodological individualism largely inhibits them from capturing
structural factors.

At a more fundamental level, functionalist social theory can been criticized for being
unable to explain growing disequilibria, structural power inequalities, social contradictions
and the role of conflict in social transformation; as well as for its inability to conceptualize
structure and agency. In contrast, ‘conflict theory’, the social scientific opposite of
functionalist/equilibrium theory, postulates that social and economic systems tend to

4
While different in several respects from neo-classical theory, NELM does not question its rationality and

utility-maximizing assumptions. However, this partly semantic debate is not necessarily a very useful one, as
the outcome also depends on how exactly to define ‘neo-classical’.

IMI Working Papers Series 2011, No. 32 11

reproduce and reinforce structural inequalities and serve the interests of the powers that
be, and that they can only be altered through a radical change in power structures through
the organized (structured) resistance of oppressed groups (e.g. revolution, unions, strikes).
In other words, social transformation does not often come smoothly, and often requires
collective action enabled by rising consciousness about one’s perceived oppression and
one’s ability to overcome such oppression by peaceful or violent resistance (on conflict
theory, see Collins (1994)).

Within the general social-scientific paradigm of ‘conflict theory’, Marxist, dependency,
and world systems theory tend to see migration as the direct outflow of the spread of global
capitalism and the related marginalization and uprooting of rural populations around the
world (de Haas 2010a) who have no choice other than to migrate to cities (within and across
borders!5) to join the urban proletariat. Migration is therefore seen as a process that serves
the interests of large corporations and specific economic interest groups (such as farmers)
and states that are strongly lobbied by these interests. These approaches can be criticized
for being overly deductive and deterministic, with their concomitant portrayal of individuals
as passive victims of economic macro-forces. In other words, individual migrants are hardly
attributed any agency and, as far as they act, they are supposed to make irrational choices.
In order to explain why people behave (migrate) in ways that go against their own objective,
material interests, Marxist theory uses the concept of false consciousness, which can be
defined as the ‘failure to recognize the instruments of one’s oppression or exploitation as
one’s own creation, as when members of an oppressed class unwittingly adopt views of the
oppressor class’ (Wiener 2005). The assumption that all or most migrants behave irrationally
seems equally unrealistic as the full rationality and income-maximizing assumptions of
orthodox neo-classical models. For instance, it would be difficult to reason that the choices
of refugees or unemployed graduates to emigrate are not rational to a considerable extent.

Although few would still agree with the more orthodox versions of neo-Marxist theory
in the face of ample empirical evidence pointing to the fact that poor people also exert a
considerable amount of agency, it would also be naïve to deny that migration processes are
to a significant extent determined by contextual factors, and that while individual choice is
certainly not absent, it is considerably constrained by structural factors –facilitating
migration of specific social groups along specific geographical and legal pathways while
simultaneously impeding it for many others groups and along many other pathways. This
seems particularly important for poor people with limited access to resources and markets
and living in politically repressive environments.

A powerful example of ‘structure’ – among several others – that appears to be
particularly crucial as a migration determinant is the segmentation of labour markets. Dual
labour market theory (Piore 1979) argued that international migration is mainly driven by
pull (employment) factors, since the segmentation of labour markets creates a permanent
demand for cheap immigrant labour at the bottom, ‘secondary’ end of the labour market to
occupy jobs that ‘primary’ workers typically shun, primarily because of social status and
relative deprivation motives. The latter exemplifies the deep socio-cultural roots of what

5
In this respect, there is no essential difference between internal and international migration, which can both

be seen as part and parcel of the same process of rural transformation and urban growth. Although
international migration is more often subject of state regulation, this is not necessarily the case, and several
states have attempted to regulate internal migration. This is another example of how conventional migration
categories often impede rather than facilitate our understanding of migration as a social process.

12 IMI Working Papers Series 2011, No. 32

superficially appears to be ‘just’ an economic phenomenon. Although this is a partial theory,
that ignores sending-side explanations altogether and implicitly (and problematically)
assumes a quasi-unlimited supply of migrant workers, its core argument is very powerful to
explain the remarkable persistence of low-skilled migration to wealthy countries over the
past half century as well as the coexistence of domestic unemployment and immigration:
the demand for low-skilled migrants is sector-specific and has become structurally
embedded in labour market structures and socio-cultural hierarchies.

In this context, Stephen Castles (2002: 1152) has argued that ‘it is one of the great
fictions of our age that the “new economy” does not need “3-D workers” any more’. He
argued that industrialized counties continue to import unskilled labor, and that – in the
absence of sufficient legal channels for low-skilled labour migration – this often takes the
form of systematic use of irregular migrants or asylum seekers (see Section 6 on
‘substitution effects’), whose very lack of rights makes them easy to exploit. Although the
industries and mines in which low-skilled migrants worked have declined since the early
1970s, Saskia Sassen (1988) has argued that new internal and international divisions of
labour have arisen, particularly in ‘global cities’, where the luxury consumption needs of the
high-skilled have created new labour market demand, particularly in the lower skilled
services, such as cleaning, childcare, restaurant work, gardening, but also in garment
manufacture, construction, garment manufacture and food processing (see also Castles
2002).

Further elaborating upon the work by Piore, Castles, Sassen and others, it is possible
to theorize that, over development processes, labour markets have grown increasingly
complex and multi-segmented (and not just ‘dual’) while the general level and degree of
specialization in education has increased. As the geographical expanse of labour markets
typically increases as education goes up, increasing levels and complexity of education and
labour markets seems to drive people to migrate in order to match supply and demand. This
seems to be one of the main reasons why relatively wealthy and developed societies are
inherently more mobile and migratory than relatively poor societies.

Studying and comparing the structure of labour markets as well as concomitant
differences in (group-level, domestic and international) income inequalities and relative
deprivation (cf. Czaika 2011) can also help us to further understand the occurrence of
significant migration between regions or countries with similar average income levels.
However, these hypotheses have remained largely untested. The methodological inference
of these theoretical insights is that, in order to advance our understanding of the structural
drivers of migration processes, there is a need to develop empirical approaches to assess
the interrelated roles of labour market structure, education and skill structure, social
fractionalization (for the latter, see Czaika 2011) and relative deprivation in affecting the
volume and, particularly, the social composition and the geographical patterning of
migration flows.

This example of labour markets exemplifies that, in order to explain real-world
migration patterns, there is a need to go beyond gravity or push-pull approaches by looking
beyond the level of ‘national averages’ such as GDP per capita and exploring the internal
structure of societies and economies. This can partly be achieved through quantitative
approaches, particularly through developing new indicators that capture key structural
features such as inequality (e.g. Gini index), relative deprivation (Czaika and de Haas 2011b;
Stark and Taylor 1991), social security (e.g., governments’ welfare spending), and labour

IMI Working Papers Series 2011, No. 32 13

market structure (for instance, sectoral composition and the relative share of the informal
sector). It goes without saying that all these factors are deeply affected by (non-migration!)
policies pursued by states.

2.3 Filling the sending-country gap

The weakness of labour market-based migration theories is that they focus on receiving-
country demand factors, and generally ignore how origin-country factors such as labour
market structure, income levels and inequalities, social security, conflict, states and public
policies, affect migration. At best, labour market-focused migration theories assume a quasi-
unlimited supply of (particularly low-skilled) migrant labour, which seems to be implicitly
based on the naïve notion that high population growth, poverty and warfare in developing
countries ‘push’ migrants to leave, thereby virtually reducing their agency to zero. This
notion clearly conflicts with empirical and theoretical insights on the intrinsic relationship
between migration and broader processes of development and social transformation (de
Haas 2009; Hatton and Williamson 1998; Massey 1991; Skeldon 1997; Zelinsky 1971). The
latter insights question the ‘unlimited supply hypothesis’ and reveal a much more complex
picture of how development processes affect migration and crucially undermine the
assumptions underpinning conventional migration theories.

For instance, conventional ideas that development in origin countries will reduce
international migration are ultimately based on the assumption of ‘push-pull’ and ‘gravity’
models that there is an inversely proportional relationship between absolute levels and
relative differences of wealth on the one hand and migration on the other. By contrast,
another group of theories postulate that development leads to generally increased levels of
immigration and emigration. ‘Migration transition theory’ (de Haas 2010a) hypothesizes
that constraints-loosening and aspirations-increasing economic and human development
and parallel demographic transitions tend to have an inverted J-curve or U-curve effect on
emigration rates. This hypothesized non-linearity and the complexity of development-
migration linkages contrast with conventional theories and also compel us to design
different, theoretically informed empirical approaches away from standard ‘push-pull’ and
gravity models.

More in general, the receiving-country bias of migration research points to the
importance of advancing our theoretical understanding of the origin-country determinants
of migration processes at different levels of aggregation. For instance, debates on migration
policies are almost automatically focused on immigration control, revealing a general
receiving-country bias in migration research, which completely ignores the important role of
emigration policies pursued by sending states (de Haas and Vezzoli 2011). Social security
and welfare spending is another example of a potentially crucial origin-country migration
determinant. While there are several studies on the contested and questionable existence
of a ‘welfare magnet’ effect on migration, this discussion is conspicuously biased towards
destination states or countries, while there is reason to believe that factors such as social
security matter equally if not more from an origin-society perspective (Kureková 2011). For
instance, a recent study on labour migration from Central and Eastern Europe (CEE) during
the post-communist transition and after EU accession found that countries with less
generous welfare states had significantly higher shares of their workers leaving to work
abroad, also after controlling for the effect of wage differentials (Kureková 2011). More

14 IMI Working Papers Series 2011, No. 32

generally, this example also shows the need to fully take into account the role of structural
and institutional factors in origin societies in shaping migration processes.

2.4 Bridging non-economic and economic migration determinants

Conventional migration theories focus on how income and wage levels and, to a lesser
extent, income inequalities affect migration processes. Although they might differ in their
specification, they share a focus on economic differentials as the main driver of migration.
This coincides with a research focus on labour migration and near-total separation from
research on ‘forced’ or refugee migration. The implicit suggestion is that these different
migration categories represent fundamentally different processes. There are many reasons
to contest this view. After all, labels such as ‘labour’, ‘refugee’, ‘family’ or ‘student’
migration primarily reflect legal categories, which are useful for administrative procedures,
but are not very meaningful categories to help understand migration as a social process. For
instance, the ‘voluntary’/‘forced’ migration dichotomy is simplistic because it assumes that
one category of migrants enjoys total freedom and the other category has no choice or
agency at all.

The legal-bureaucratic categories frequently used in social scientific research conceal
the fact that, on a macro-level, migration processes are driven by a multitude of economic
and non-economic factors and that, on a micro-level, migrants are motivated by a
combination of multiple, interconnected but analytically distinct social, cultural, economic
and political factors. For instance, economic development is positively associated with
democratization processes (cf. Burkhart and Lewis-Beck 1994), and economic development
and democratization are likely to affect migration processes simultaneously. It would be
naïve to assume that refugees are also affected by economic and social considerations,
certainly where destination choice is concerned. Likewise, ‘labour migrants’ are likely also to
weigh personal freedoms in their migration decision-making. And ‘family migrants’ are
potential workers too.

However, the ways in which non-economic factors such as political and personal
freedoms affect migration is much more complex than it seems at first sight. From transition
theory, we already know that the relation between development, economic growth and
migration is fundamentally non-linear. Also democratization processes and related increases
in personal freedoms seem to have rather ambiguous, and potentially non-linear effects on
migration. For instance, while a lack of freedoms is likely to fuel migration aspirations, the
same lack of freedoms may simultaneously decrease people’s capabilities to migrate (de
Haas 2010a). After all, authoritarian, non-democratic states also tend to curtail the
freedoms of people to migrate, either through blocking exit or by creating bureaucratic
obstacles such as excessive costs to acquire passports (de Haas and Vezzoli 2011; McKenzie
2005). This might explain the negative association between a lack of political freedoms and
emigration (de Haas 2010a).

The same study suggested that there might be an unexpected positive relation
between the lack of freedoms and levels of immigration (de Haas 2010a). It is as yet unclear
whether this provides evidence for any variant of ‘numbers vs. rights hypotheses’ (Ruhs and
Martin 2008), according to which there would be a trade-off between the rights that states
attribute to migrants and the number of migrants who are allowed in. One possible
explanation is that authoritarian states which give fewer rights to their citizens and even
fewer to migrants also have a higher ability to impose segmented, inherently discriminatory

IMI Working Papers Series 2011, No. 32 15

labour markets, to organize and recruit labour, to segregate immigrants from native
populations, and are less sensitive to domestic political pressure for immigration reduction.
More generally, this shows the need to radically improve our understanding of the role of
states and migration policies in shaping migration processes.

These few examples also show the need to look beyond specific policies, and to
consider the nature of states. For instance, the position of states both on the
authoritarianism-democracy and on the strong-weak central power continuums seems to be
an important macro-structural determinant of migration processes, as both positions affect
aspirations and capabilities to migrate and the extent to which states will desire and be able
to ‘steer’ migration. There is also a clear need to differentiate between different types of
freedoms (e.g. freedom of expression, movement, and physical integrity, freedom from
cruel and inhuman punishment, freedom from violence) as they are likely to affect
migration in different ways.

3 Towards theoretical synthesis: an aspirations-capabilities
framework

The main challenge for advancing migration theory is how to synthesize the different
migration theories developed across a range of social science disciplines – ranging from
economics to anthropology. Faced with the daunting complexity and diversity of migration
processes, migration scholars have often – and perhaps wearingly – argued that an all-
encompassing and all-explaining theory of migration will never arise (Castles and Miller
2009; Salt 1987; Van Amersfoort 1998). Unfortunately, this probably sensible observation
has coincided with a strong tendency to abandon theorizing migration altogether. Although
migration is certainly a complex and apparently ‘messy’ process, this goes for virtually all
social processes. Moreover, migration may be complex, but it is certainly not a random
process. Instead, it is a strongly socially structured and spatially patterned process, in which
strong regularities can be discerned.

More generally, ‘all-comprehensiveness’ is not what social theory should be about in
the first place. Social theory formation is precisely about striking a delicate balance between
the desire to acknowledge the intricate complexities and the richness of social life on the
one hand and the scientific need to discern underlying regularities, patterns and trends on
the other. Theory formation is exactly about generalizing, which is a reductionist process by
definition, where the exception may well prove the rule. Although it is indeed naïve to
assume that a one-size-fits-all theory explaining migration at all places and at all times will
ever arise, there is undoubtedly more room for theorizing on migration processes and how
they reciprocally connect to broader processes of social and economic change.

Much can already be gained from developing a more unified social-scientific
perspective on migration, in which unproductive disciplinary boundaries are broken down.
In their seminal review of migration theories, Massey et al. (1993) rightly argued that the
different theories on migration are not necessarily mutually exclusive. Because different
disciplines use different jargons and methodological tools, they often seem irreconcilable,
but below the surface they often study similar processes and causal links. Once conceptual
confusion is resolved by debate, and sufficient openness is created to learn from other
methodological approaches, a lot of the apparent contradictions turn out to be rather
spurious, and cross-fertilization can enrich theoretical thinking. For instance, the new

16 IMI Working Papers Series 2011, No. 32

economics of labour migration (NELM), which is one of the major past advances in economic
migration theory, was apparently inspired by research on household composition and
livelihood strategies conducted by anthropologists and sociologists (Lucas and Stark 1985:
901). Although there are marked differences between different theories, disciplines and
associated research traditions, they are not necessarily incompatible, and there is
considerable room to identify more common grounds and to create conceptual bridges.

However, an eclectic ‘combining of insights’ cannot solve some more fundamental
problems, particularly when theories have different paradigmatic roots. For instance, it
seems impossible to merge neo-classical and neo-Marxist migration theories, because they
differ in their most fundamental assumptions. For similar reasons, theoretical problems
cannot be solved by simply ‘plugging in’ variables ‘representing’ the different theories in the
same regression equation, as is often the tendency. What is really lacking, and what is
hindering theoretical synthesis, is a more comprehensive and convincing ‘behavioural’
framework of migration than the current theories offer. The only systematically elaborated
micro-behavioural model of migration is neo-classical. Although neo-classical migration
theory has been much reviled for a number of more and less convincing reasons, no credible
alternative has been proposed so far.

Despite the enormous value of macro-level theories developed by sociologists,
geographers and demographers, because of their very macro-level nature they often lack a
‘behavioural link’6 to the micro-level. In other words, they do not make explicit the
behavioural assumptions underpinning the macro-level correlations they assume or
describe. It would be to commit a classical ‘ecological fallacy’ to confound macro-level
migration determinants (e.g. population growth, environmental degradation, climate
change or variability) with individual migration motives – which is exactly what the push-pull
and non-expert literature on environmental change and migration typically does. After all,
people do not migrate ‘because of’ abstract concepts such as demographic transitions,
declining fertility, ageing, population density, environmental degradation or factor
productivity. For instance, there may often be a correlation between demographic and
migration transitions, but this does not make clear why people should necessarily migrate
under conditions of high population growth. People will only migrate if they perceive better
opportunities elsewhere and have the capabilities to move. Although this assertion implies
choice and agency, it also shows that this agency is constrained by (historically determined)
conditions which create concrete opportunity structures.

Ultimately, in the social world, ‘causality’ therefore runs through people’s agency,
producing outcomes on the aggregate level which can perhaps be measured through macro-
indicators. But any convincing macro-model should be underpinned by a credible micro-
behavioural link. The lack of micro-behavioural foundation makes most macro-theories
deterministic. In fact, the problem with the very term ‘determinants’ is that it conveys a
somehow deterministic picture of ‘causation from outside’, independent from migrants’
agency and internal migration dynamics. It seems therefore desirable to (re)define the

6
It is important to mention that in many social scientific approaches, behaviour has been conceived as a

response to the environmental or contextual stimuli without taking into account cognitive processes, social
hierarchies and agency. This actually comes close to push-pull and neo-classical models, whose notions of
agency are rather mechanistic, if they exist at all. So, there is a need to define the concept of ‘behaviour’ more
broadly so as to include agency.

IMI Working Papers Series 2011, No. 32 17

concept of ‘determinants’ so as to include human agency, which has independent power to
change social structures.

Crucially, most macro-theories ignore agency. At the same time, neo-classical
migration theory has a reductionist, mechanistic concept of agency. Hence, what we need is
a new and more realistic micro-level model or framework of migration. Such a framework
should take into account empirical insights of decades of migration research from across a
range of disciplines, but at the same time it should remain basic and parsimonious enough
so as to fulfil its generalizing ambitions. Such a framework should specify the basic
assumptions about the factors that make people decide to migrate (or not). Two further
conditions need to be met: first, such a model should incorporate a sense of agency, and
should not conceive migration as an almost ‘mechanistic’ response to a range of ‘pushes’ or
‘pulls’, or wage differentials. Ultimately, this is also the reason why gravity models normally
used for trade cannot be assumed to be valid to model human migration. People are not
goods. Goods are passive. People are humans, who make active decisions based on their
subjective aspirations and preferences, so their behaviour is not just a function of macro-
level disequilibria, neither does their behaviour necessarily decrease these disequilibria.
Second, such a micro-model should incorporate a sense of structure, in the sense that
migration behaviour is constrained by structurally determined resource and information
limitations.

This above analysis leads to the proposition that, in order to improve our insights into
the factors driving migration, and to synthesize prior theories, an improved theoretical
model of migration should:

• conceive migration aspirations as a function of spatial opportunity (instead of only
income or wage) differentials and people’s life aspirations

• conceive migration propensities as a function of their aspirations and capabilities to
migrate.

These two basic assumptions about migration behaviour can serve as basic building blocks
to build a theory of migration which synthesizes many existing theoretical and empirical
insights. Although this still needs considerable theoretical elaboration in future work, such a
conceptualization would allow us to:

• integrate economic and non-economic theories on migration and overcome
‘migration category’-based theorizing (particularly through the notion of
opportunity instead of income differentials)

• integrate theories on so-called ‘voluntary’ and ‘forced’ migration (particularly
through the constraints dimension embedded in the capabilities concept)

• link micro- to macro-theories (as macro-level factors shape opportunity structures
which condition – that is, simultaneously enable and constrain – people’s migration
decisions as far as their own capabilities allow)

• open new avenues for integrating agency and culture into migration theory (such
as through the concept of aspirations).

The conceptualization of migration as a function of opportunity rather than income or wage
differentials compels us to study how social, economic and political conditions affect
migration processes simultaneously. Improved empirical models should reflect this and
would allow for the study of the relative importance of each of such factors as well as their
mutual interaction. In an attempt to move beyond the artificial separation between

18 IMI Working Papers Series 2011, No. 32

economic and non-economic explanations, it seems useful to apply Amartya Sen’s (1999)
capabilities approach to migration theory. In his book Development as Freedom, Sen (1999)
defined development as the process of expanding the substantive freedoms that people
enjoy. In order to operationalize these ‘freedoms’, he used the concept of human capability,
which refers to the ability of human beings to lead lives they have reason to value, and to
enhance the substantive choices they have (Sen 1997: 1959). Sen stressed that freedom is
central to the process of development primarily for its intrinsic, wellbeing-enhancing power,
which has to be clearly distinguished from the instrumental effectiveness of freedoms of in
contributing to economic progress, which have been the usual benchmark to ‘measure’
development.

Within this capabilities perspective, I conceive human mobility as an integral part of
human development for both intrinsic and instrumental reasons. First, people can only
move if they have the capabilities to do so. Human mobility can be defined as the capability
to decide where to live – and migration is the associated functioning. Expansions in this
capability (through accessing social, human and/or economic resources) are an expansion of
the choices open to an individual and therefore of their freedom. This is the intrinsic
argument why mobility can be an integral part of human development. At the same time,
movement can enable people to improve other dimensions relevant to their capabilities
such as their income, their health, the education of themselves and of their children, and
their self-respect. This is the instrumental value of mobility for development. This is why it is
important to distinguish between the capability to move and the act of movement. In fact,
some manifestations of migration (e.g. refugee migration) are a result of the choices and
freedoms of individuals becoming more restricted. So, enhanced mobility is not only the
freedom to move – it is also the freedom to stay in one’s preferred location. Having choice
to stay or to go, and where to go, captures the very essence of agency.

The application of a capabilities-focused conceptualization of development (Sen 1999)
also creates conceptual room to fully include factors such as education, health, social
security, various (income, gender, ethnic) inequalities, and personal and political freedoms
as migration determinants. It also creates room to broaden our view of freedom- and
wellbeing-generating resources to include not only economic, but also human and social
resources or ‘capitals’.

Another conceptual advantage of Sen’s perspective is that the notion of capabilities
creates analytical room to start incorporating notions of agency in migration theory. The
concept of agency is intrinsically linked to the power of social actors to affect processes of
structural change. It is important to emphasize that agency can both sustain as well as alter
processes and structural conditions (cf. Emirbayer and Mische 1998). From this, migration
itself can be conceptualized as a form, or expression of, agency (see also de Haas 2009), and
not only a ‘functionalist’ response to spatial differentials in economic opportunity. However,
the extent to which social actors can exert agency (that is, their capability) is dependent on
structural conditions which determine the space of manoeuvre within which individuals can
make independent choices. Within the capabilities framework, the act of migration itself can
be wellbeing-enhancing for the intrinsic value of the migration experience. Crucially, this
enables us to incorporate manifestations of migration and mobility (e.g. ‘migration as
adventure’, ‘gap years’, ‘lifestyle migration’, au pair migration), where the experience itself
is an important motive for moving, and the improvement of material circumstances plays a
relatively minor (or no) role. As with tourism, through discovering new horizons and

IMI Working Papers Series 2011, No. 32 19

acquainting oneself with other cultures, in particular for young people, migration can have
an intrinsic wellbeing-enhancing dimension.

As a next conceptual step, and drawing on Isaiah Berlin’s (1969) concepts of negative
and positive liberty, we can conceptualize capabilities as a function of positive and negative
freedoms. Within Berlin’s perspective, negative liberty means the absence of obstacles,
barriers or constraints. This comes close to classical ways of conceiving freedom, which are
particularly focused on the role of states and politics (including war and violent oppression)
in imposing constraints on people’s freedom or even being an outright threat to people’s
lives. This concept of liberty is also the basis for the United Nations Refugee Protection
regime,7 and international human rights organizations. Within this perspective, democracy,
conflict prevention and promoting the rule of law are typically seen as ways to promote
people’s freedoms and to prevent ‘forced’ migration.

Berlin’s (1969) concept of positive liberty refers to the possibility or the fact of acting
in such a way as to take control of one’s life and realize one’s fundamental purposes. This
concept pertains to the agency of individuals and groups to change their life circumstances
and to escape from disadvantaged positions. It is enshrined in international human rights8
and notions of ‘empowerment’ in development theory. Positive liberty embodies the notion
that the absence of external constraint (negative liberty) is not a sufficient condition for
people to improve their wellbeing. This is a point that Amartya Sen has particularly stressed
in his development theory. For instance, a given state might be formally democratic and
there might be an absence of political persecutions, but illiterate and poor people generally
lack the capabilities and resources to actually make use of such liberties. In other words,
people need access to resources in the forms of social, human and material (including
financial) capital in order to exert their agency, such as the freedom to migrate or not to
migrate. This reveals a fundamental paradox: although relative deprivation of freedoms and
an awareness of better opportunities elsewhere may make people aspire to migrate,
absolute deprivation of either negative or positive freedoms, or both, will prevent them
from exerting such migratory agency.

So, from a capabilities point of view, the very term ‘forced migration’ is somehow an
oxymoron, as people still need capabilities to be able to migrate. While deprivation of
negative freedoms is likely to motivate people to migrate, they need a certain level of
empowerment or access to positive freedoms (social, human and/or material capital) in
order to actually be capable of fleeing towards a particular destination. When people are
deprived of both freedoms, they are generally forced to stay where they are. In conflict

7
Under the United Nations Convention Relating to the Status of Refugees from 1951, a refugee is a person

who (according to the formal definition in article 1A of this Convention), owing to a well-founded fear of being
persecuted on account of race, religion, nationality, membership of a particular social group, or political
opinion, is outside the country of their nationality, and is unable to or, owing to such fear, is unwilling to avail
him/herself of the protection of that country.
8
The Universal Declaration of Human Rights states that all human beings – regardless of cultural,

organizational, religious or ethnic associations – should be entitled to certain social, political and legal rights of
toleration. Although this definition, and the term ‘toleration’ in particular, still largely resonates ‘negative
freedoms’, the way this is enshrined into international law goes beyond this, and includes ‘positive freedoms’
such as people’s right to a ‘standard of living adequate for the health and well-being of himself and of his
family, including food, clothing, housing and medical care and necessary social services, and the right to
security in the event of unemployment, sickness, disability, widowhood, old age or other lack of livelihood in
circumstances beyond his control’ (article 25.1) as well as the right to education and employment.

20 IMI Working Papers Series 2011, No. 32

situations, the most deprived are typically the ones who are ‘forced to stay’. The concept of
negative freedom is also useful for theorizing the role of immigration and emigration
policies. Restrictive immigration policies can decreases people’s ‘negative freedoms’ to
migrate, and can create situations of ‘involuntary immobility’, a term aptly coined by Carling
(2002). Such involuntary immobility can also occur under restrictive emigration policies. For
instance, repressive countries often experience lower emigration, because authoritarian
states have a larger capability to deprive their citizens of the right to exit (de Haas and
Vezzoli 2011).

However, even under liberal migration policies where people may enjoy abundant
negative freedoms, if they are deprived of the basic positive freedoms and access to social,
human and economic resources, they will still be unable to migrate, particularly over larger
distances. All of this helps to explain the paradox of why development often coincides with
increasing levels of migration. From this, I hypothesize that most emigration is likely to
occur when people enjoy a maximum of negative freedoms and a moderate level of positive
freedoms, as very high levels of positive freedoms and declining spatial opportunity
differentials would somehow decrease their aspirations to migrate. This also shows why so-
called push-pull theories are fundamentally flawed: with the exception of extreme
situations like slavery, people are not goods that can be passively moved: they need to
move by themselves, and a fundamental precondition for that to happen is that they have
the willingness and capabilities (i.e. economic, social and human resources) to do so.

This brings in the concept of aspirations, which is a crucial element of this attempt at
theoretical synthesis and, particularly, the attempt to better incorporate agency in
migration theory. Conventional migration theories either totally disregard (such as Marxist
or push-pull models) or have very reductionist notions of agency. Although within neo-
classical and other functionalist migration theories, there is room for individual decision-
making, there is no genuine room for agency, because individual behaviour is a totally
predictable, mechanistic outcome of wage and other opportunity differentials. The
underlying assumptions are that people are free from constraints, enjoy full access to
information, and make migration decisions with the aim of maximizing their utility. These
are clearly unrealistic assumptions. Although mainstream economics and, to a certain
extent, migration economics have come a long way to acknowledge information and market
imperfections in their theories and models, the utility-maximizing notion underlying
decision-making has not been fundamentally challenged.

Here, it is important to observe that push-pull and gravity models as well as neo-
classical and other functionalist migration theories implicitly assume that people’s
preferences and, hence, aspirations are constant across societies and over time, and
basically boil down to individual income (or ‘utility’) maximization. In other words, people
living in different societies, despite the huge variations in the amount and type of
information and social, cultural and economic resources they can access, are somehow
assumed to react in similar fashions to similar external stimuli or exogenously defined ‘push’
and ‘pull’ factors. This is what makes functionalist theory inherently mechanistic and their
micro-models totally devoid of any real sense of agency, as individual choices are entirely
predictable and human beings are, indeed, conceptualized to be ‘pulled’ and ‘pushed’ in
space like atoms without their own will and ability to make independent choices and,
herewith, affect structural change. Functionalist theory conceptualizes migration as an
equilibrium- and system-reinforcing process. It therefore leaves no analytical room for

IMI Working Papers Series 2011, No. 32 21

either structural inequalities embedded in social hierarchies or migrants exercising agency.
Such agency can either reproduce the existing structural inequalities and social relations
(the normal pattern), or it can contribute to social transformation, which can be defined as a
fundamental shift in the way society is organized (Castles 2010) that goes beyond the
incremental and smooth change embodied in functionalist theory.

The crucial problem is that functionalist migration theory assumes that overall
(migration) preferences are more or less constant across societies and over time. This
ignores the fact that culture, education and access and exposure to particular forms of
information are likely to have a huge impact on (1) people’s notions of the good life and,
hence, personal life aspirations; and (2) their awareness and perception of opportunities
elsewhere. If people do not aspire to other lifestyles ‘elsewhere’, even if they seem
‘objectively’ or ‘materially’ better, they will not translate this awareness into a desire to
migrate. In fact, cultural ‘home preference’ seems to be a major explanation for why most
people do not migrate. On the other hand, if migration-as-an-experience is intrinsically seen
as wellbeing-enhancing, people might even voluntary opt for ‘objectively’ less favourable
circumstances. Nevertheless, it seems reasonable to hypothesize that, in general, people’s
personal life aspirations and awareness of opportunities elsewhere increase when levels of
education and access to information improve in processes that are usually conceived as
‘human development’. If this coincides with the occurrence of significant differences in
structurally determined spatial opportunity differentials, this is more likely to generate
aspirations to migrate in an attempt to fulfil these life aspirations.

In another paper, I have therefore stressed the importance of considering the role of
factors such as access to education and information in migration processes, not only
because ‘human capital’ endows people with greater resources and incentives to migrate
(which is the focus of economic migration theories), but also because these factors have
arguably an aspirations-increasing effect and can paradoxically lead to increasing aggregate
emigration propensities also under conditions of high economic growth and human
development (de Haas 2010a).9 This shows the benefit of conceptualizing migration as a
function of aspirations and capabilities within a given set of opportunity structures.

Altogether, this yields a more comprehensive picture of behavioural causes of
migration beyond the basic model of income-maximizing individuals reacting to wage
differentials. Such an amended theoretical framework also helps us to re-conceptualize
migration as an intrinsic part of processes of human development rather than the (causal)
‘outcome’ of development failure10 or a function of income and wage differentials or other
externally given ‘push’ and ‘pull’ factors. Conceiving migration as a function of capabilities
and aspirations to migrate also gives us better, albeit certainly not perfect, conceptual tools
to start incorporating meaningful notions of agency in theoretical models and empirical
approaches. More in general, the simultaneous incorporation of agency and structure in
migration theories remains one of the main challenges for advancing migration theory and,
hence, the specification of more realistic empirical approaches. As I already mentioned
above, even purely at the discursive level, one of the most fundamental problems is that the
very term ‘determinants’ linguistically ‘externalizes’ the factors affecting migration, moving

9
This is additional to the structural mobility-fuelling role of increasing labour market complexity.

10
This idea is particularly prevalent in popular ideas that a ‘misery push’ is driving much South–North

migration.

22 IMI Working Papers Series 2011, No. 32

it implicitly outside the realm of agency and implying a physical, mechanistic notion of
causality in which migrants are assumed to (passively and entirely predictably) react to
those factors.

A key condition for incorporating structure and agency is to connect both concepts
and to understand their dialectics. In this respect, ‘structure’ is often erroneously seen as a
set of constraints, whereas in reality structures simultaneously constrain and facilitate
agency. As we have seen, factors such as states and policies, economic and social
inequalities as well as networks have a strong structuring effect on migration, which means
that they are inclusive for some (age, gender, skill, ethnic, regional) groups and exclusive for
others, and that they strongly favour migration along certain geographical pathways while
discouraging it along others. This typically leads to a rather neat social and geographical
structuring and clustering of migration.

So, the ensemble of structural conditions creates complex opportunity structures,
endowing different individuals and social groups with different sets of negative and positive
freedoms, which, depending on how these constellations affect their capabilities and
aspirations, may or may not make them decide to migrate. In its turn, such agency will
reciprocally affect these initial conditions through feedback effects, exemplifying the
dialectics of structure and agency in migration processes.

4 Linking micro and macro levels

The challenge to link agency and structure is also related to the difficulties of linking micro-
level explanations of migration, which focus on how individual characteristics, access to
resources, perceptions and preferences shape migration behaviour, to macro-level level
theories which, ultimately, see migrants’ behaviour as a rather passive, and therefore rather
predictable, outcome of given opportunity structures. In the literature it has been argued
that meso-level theories on the formation of networks and migration systems provide this
vital link (Faist 1997). The migration literature has identified various feedback mechanisms
which explain why, once started, migration processes tend to become partly self-
perpetuating, leading to the formation of migrant networks and migration systems (Castles
and Miller 2009; de Haas 2010b; Mabogunje 1970; Massey 1990; Massey et al. 1998). Such
feedback loops provide a powerful, concrete example of the dialectics between agency and
structure, as they show how migrants create meso-level structures such as networks and
the ‘migration industry’ (Castles 2004a)11 that have a knock-on (feedback) effect in
reinforcing migration between particular places and countries through counter-flows of
ideas and information (Mabogunje 1970), as well as decreasing the costs and risks of
migration (Massey et al. 1998), thereby actively defying structural constraints such as high
travel costs and restrictive immigration policies. This is a prime example of how migrants
exert agency and are able to change initial structural conditions in such a way that they
further facilitate migration along particular pathways. It is also a prime explanation of why
states often find it difficult to control once-started migration processes. These notions are
crucial for theorizing the role of states and policies in migration processes.

According to Castles (2004a) the ‘migration industry’ consists of clusters and networks of travel agents,
lawyers, bankers, labour recruiters, brokers, interpreters, housing agents as well as human smugglers and
traffickers, which have an interest in the continuation of migration. Besides networks, these are other
examples of meso-level structures created by the migration process itself.

IMI Working Papers Series 2011, No. 32 23

However, existing theories on these ‘internal dynamics’ of migration processes are
characterized by some fundamental weaknesses which I reviewed in another paper (de Haas
2010b). First, the usual focus on migrant networks coincides with a neglect of other
feedback dynamics that operate through the impact of migration on the sending and
receiving contexts. Migration inevitably changes the initial structural conditions under which
migration takes place in sending and receiving communities and societies, which, in their
turn, reciprocally affect people’s aspirations and capabilities to migrate. Examples of such
structural impacts include the impact of migration on income inequality and relative
deprivation in origin societies, the migration-facilitating role of remittances, and the rise of
immigrant-dominated entrepreneurial sectors in destination countries, as well as the
segmentation of labour markets along ethnic lines. Such processes contribute to the
formation of migration systems – a set of places or countries linked by flows and counter-
flows of people, goods, services, and information, which tend to facilitate further exchange,
including migration, between the places (de Haas 2010b; Fawcett 1989; Kritz et al. 1992;
Mabogunje 1970; Massey et al. 1998).

Second, the largely circular logic of these theories reveals an inability to conceptualize
which migration-undermining feedback mechanisms may counteract migration-facilitating
feedback dynamics and which may explain the endogenous decline of established migration
systems. Theoretically, this can be explained by applying insights from the critical social
capital literature pioneered by Portes (1998) and, in particular, the notion of negative social
capital, to migration theories (de Haas 2010b). Migrants do not necessarily help each other,
and strong social ties and networks can also exclude non-group members. One of the
methodological lessons is that empirical models should not just assume that the strength of
network effects is a function of the size of migrant communities, as recent quantitative work
tends to do. The relative importance of networks in facilitating migration crucially depends
on the relative dependence on social (as compared to economic and human) capital among
migrant communities. Moreover, positive network effects tend to decline over time.

5 Theorizing the role of states and policies in migration processes

If anything, the above analysis points to the preponderance of structural factors such as
economic and human development, labour market structure, social stratification, income
inequalities, relative deprivation and social security, and the role of negative freedoms as
well as positive freedoms in the form of access to material, social and human capital in
shaping people’s capabilities and aspirations to migrate. This compels us to ask the
following crucial question: within this broader whole of big forces and structural factors, and
migrants’ considerable agency to shape and consolidate migration pathways and networks,
what role is still left for migration policies pursued by states? Is that a comparatively
marginal one, or do policies still play a key role?

There is no simple answer to that question, first of all because the role of states and
policies seems to vary according to the nature of the states, and is also dependent on the
phase of migration system formation. The answer also crucially depends on whether we
refer to the role of states in general or the role of specific migration policies. However,
based on this theoretical framework it is possible to elaborate a few hypotheses. These are
based on the notion that migration policies primarily affect negative freedoms in the form
of the right to leave or enter a national territory, but that, primarily through non-migration

24 IMI Working Papers Series 2011, No. 32

policies (such as economic and education), states also affect people’s positive freedoms.
While these factors affect people’s capabilities to migrate, factors such as repression and
poverty affect people’s aspirations to migrate.

First, the power of states to influence immigration and, particularly, emigration is
much higher for repressive, authoritarian and centralized states than for liberal, democratic
and decentralized states, which need to take more account of democratic processes and
fundamental human rights. Second, we can hypothesize that states and policies often play
an important role in the initiation of international migration, whether in the form of
recruitment (for instance, of ‘guest workers’), visa requirements, colonialism, military
occupation, or political repression (Castles and Miller 2009; de Haas 2008; Entzinger 1985;
Massey et al. 1998; Penninx 1982; Skeldon 1997). On the other hand, it is important to
emphasize that this is not always the case and that certain policies, such as recruitment, can
also be an attempt to formalize already existing flows.

However, once a certain number of migrants have settled at the destination,
migration can become partly self-perpetuating (Castles 2004b; Massey 1990; Massey et al.
1998). The ‘internal dynamics’ of migration processes make additional movements more
likely through various social, cultural and economic feedback mechanisms (de Haas 2010b).
According to migration systems theory (Mabogunje 1970), such mechanisms lead to almost
organized migratory flows between particular regions and countries (Kritz et al. 1992; Portes
and Böröcz 1987). In particular, migrant networks are believed to play a crucial role in
facilitating continued migration over formally closed borders (Böcker 1994), which is a key
example of how migrants’ agency and counter-stategies can actively undermine states’
attempts to control migration.

Many migration scholars are therefore sceptical about the abilities of liberal
democratic states to control migration. They argue that fluctuations in migration primarily
respond to structural demand factors determined by human development, economic cycles,
employment and changes in the structure of (segmented) labour markets; factors which
largely lie beyond the reach of policy-makers (Castles and Miller 2009; Thielemann 2006). At
the same time, migrant networks further facilitate migration along established pathways.
Hence the assertion that ‘borders are largely beyond control and little can be done to really
cut down on immigration’ (Bhagwati 2003:99; cf. Düvell 2005). Other scholars have
countered such scepticism by arguing that, on the whole, immigration policies have been
largely effective (Brochmann and Hammar 1999; Carling 2002; Collyer 2006; Strikwerda
1999).

However, this is partly a spurious disagreement. Considerable conceptual confusion
can be reduced by clearly distinguishing the preponderant role of states in migration
processes from the comparatively more marginal role of specific immigration and
emigration policies. There can be no doubt that states can play an absolutely crucial role in
shaping and transforming migration patterns (Brochmann and Hammar 1999; Castles and
Miller 2009; Skeldon 1997; Strikwerda 1999). Over the course of modern history, trends and
patterns of migration have been intrinsically linked to processes of state formation and
decline, economic and territorial imperialism and warfare. The very notion of international
migration presumes the existence of national states and clearly defined territorial and
institutional borders. The importance of factors such as economic and human development,
labour markets, education and income inequalities points to the importance of non-
migration policies, such as labour market, taxation, social welfare and foreign policies in

IMI Working Papers Series 2011, No. 32 25

indirectly affecting migration processes. From this, it is possible to hypothesize that state
influence is primarily felt through general policies rather than migration policies per se, as
the latter have a limited influence on the main determinants of migration.

In the face of the dispute in migration research about the effectiveness of migration
policy, it is important minimize conceptual confusion by clearly defining what constitutes
migration policy and by distinguishing policy effectiveness from policy effects (Czaika and de
Haas 2011a). Migration policies can be defined as laws, rules, measures, and practices
implemented by national states with the stated objective to influence the volume, origin
and internal composition of migration flows. The term ‘effectiveness’ refers to the extent to
which policy objectives have been met, while the ‘effect’ just refers to the actual impact of a
particular law, measure or regulation. This gives effectiveness a strong evaluative (and
subjective) dimension.

Czaika and De Haas (2011a) identified three policy gaps which can explain perceived or
real policy failure. First, the ‘discourse gap’ is the considerable discrepancy between the
stated objectives of politicians’ migration discourses and rhetoric on the one hand and
concrete policies on the other. Second, the ‘implementation gap’ is the frequent disparity
between policies on paper and their actual implementation. Third, the ‘efficacy’ gap is the
extent to which an implemented policy has the capacity to affect migration flows (see Figure
1). While questioning the common assumption that migration policies have become more
restrictive overall, they argued that studies should use concrete policies rather than public
discourses as an evaluative benchmark; since the policy will almost always be construed as a
failure if official, ‘tough’ immigration discourses are taken as a reference.

6 Hypothesizing substitution effects of migration policies

The migration policy literature has argued that immigration policies frequently fail because
they have several unintended, often counter-productive effects. Within the framework
developed in this paper, migrants’ agency – in particular their creative ability to defy
immigration rules by adopting new migration strategies and pathways – plays a key role in
explaining such unintended effects. However, the existence and strength of such ‘perverse’
effects is highly contested, and therefore requires better empirical testing. It is reasonable
to assume that migration policies, if implemented, must have some effect on migration. The
crucial questions are: which effects, and what is the relative importance of these effects
compared to other migration determinants. In order to create conceptual clarity, it is useful
to distinguish the effect of migration policies on the following:

1. the volume of migration
2. the spatial orientation of migration
3. the composition (legal channel and migrant characteristics) of migration
4. the timing of migration
5. reverse (return) migration

26 IMI Working Papers Series 2011, No. 32

Figure 1: Conceptual framework of migration policy effects and effectiveness12

I am grateful to Mathias Czaika and Simona Vezzoli for their contributions to Figure 1.

IMI Working Papers Series 2011, No. 32 27

Recent reviews of immigration policies (Czaika and de Haas 2011a) and emigration policies
(de Haas and Vezzoli 2011) lead to the hypothesis that policies are more effective in
determining the selection and composition of migration rather than the overall volume and
long-term trends of migration. However, the impact of policies on migration volumes of the
particularly targeted category (e.g. asylum seekers or family migrants) receive most if not all
attention, which is unfortunate as the effects on other flows are crucial in understanding the
structural and long-term effects of migration policies on overall migration flows.

Based on the above analysis, I hypothesize that immigration restrictions can
potentially lead to four main types of substitution effects which can reduce the effect of
restrictions on inflows in the particular, targeted category:

1. Spatial substitution effects may occur through the diversion of migration to
countries with less restrictive regulations for similar categories of migrants. There is
some largely descriptive evidence observing such spatial substitution effects for
asylum, family and irregular migration to Europe and North America. In the Dutch
language, such spatial substitution effects have also been dubbed as the ‘waterbed
effect’ (Grütters 2003; van der Erf 2003).

2. Categorical substitution effects may occur due to a reorientation towards other
legal or illegal channels when entry through one particular channel becomes more
difficult. For instance, it has frequently been argued that the lack of immigration
channels for low-skilled labour migrants has compelled migration through family,
asylum or student migration channels by people who basically migrated to work (de
Haas 2007; Harris 2002; Massey 2004) and that it has increased irregular migration
(Castles 2004b; de Haas 2008; van Liempt and Doomernik 2006).

3. Inter-temporal substitution effects or ‘now or never migration’ may occur if
migration surges in the expectation of a future tightening of migration regulations.
For instance, it has been argued that when the Federal Republic of Germany tried
to discourage family reunification in the late 1970s, family migration to the Federal
Republic increased, since many migrants feared that, eventually, family
reunification might be forbidden completely (Entzinger 1985: 267). There was a
surge in Surinamese migration to the Netherlands in the 1970s around
independence (Van Amersfoort 2010), and a surge in West Indian migration before
1962, when restrictions were introduced with the Commonwealth Immigrants Act
(Peach 1968). Such effects have also been described for asylum migration (Grütters
2003). After the introduction of more restrictions, immigration typically shows a
sharp fall. The long-term effect of such restrictions may thus be limited by the pre-
measure surge in inflows.

4. Reverse flow substitution effects occur when immigration restrictions decrease
return migration flows. Several studies have argued that restrictive immigration
policies discourage return migration and therefore push migrants into permanent
settlement. This phenomenon has been described for Turkish and Moroccan ‘guest
worker’ migration to north-west Europe, where many temporary workers ended up
settling after the post 1973 recruitment ban (Böcker 1994; Castles and Miller 2009;
Entzinger 1985). If migration restrictions decrease inflows but simultaneously also
decrease return flows, their effect on net inflows becomes much more ambiguous.
However, such hypotheses have not been subjected to empirical tests.

28 IMI Working Papers Series 2011, No. 32

These four hypotheses about the unintended effects of policy restrictions need to be taken
into account when measuring the effect of particular policies on migration flows. Decreases
in restrictiveness are likely to have the opposite effect, and restrictive emigration policies
can also have more or less similar spatial, categorical, inter-temporal and reverse flow
substitution effects. As has been argued above, the danger of exclusively focusing on the
particular inflow targeted by the policy is to over-estimate its effect. It is only by focusing on
the effects of policy on overall migration flows through other spatial and legal channels and
over a longer time period that a more comprehensive and methodologically valid picture
can be obtained.

Additional hypotheses can be elaborated on the policy effects of frequently used non-
restrictive policy instruments. Examples may include the oft-assumed ‘pull effect’ of
legalizations of irregular migrants, which have made such policies politically controversial.
However, the existence of such pull effects has been contested based on descriptive
quantitative analyses (Sandell 2006), indicating that this hypothesis needs proper testing.
Another example is the effect of labour recruitment agreements. It has been argued that
their effect is much lower than often hypothesized (Reniers 1999; Shadid 1979), but here
also there is an absence of adequate, empirical tests.

Besides measuring the direct effects of migration policies on the volume of flows
within the migration category (for instance, by origin, skill, or gender group) targeted by
specific policies, empirical analyses within the DEMIG project will focus on testing for these
various substitution effects in order to acquire a more comprehensive empirical insight into
the effects of migration policies. It goes without saying that empirical analyses will control
for other theoretically relevant sending- and receiving-country migration determinants
derived from the conceptual framework developed in this paper.

Conclusion

In this paper, I have argued that although the effectiveness of migration policies has been
widely contested in the face of their supposed failure to steer immigration and their
hypothesized counter-productive effects, empirical evidence has remained inconclusive as a
consequence of fundamental methodological and conceptual limitations. Although the
general migration policy literature has yielded a rich set of hypotheses on possible policy
effects, empirical evidence is mostly descriptive or anecdotal. At the same time, the
migration determinants literature suffers from methodological problems and is largely
based on obsolete and theoretically uninformed push-pull and gravity models, and is biased
by omitting crucial sending-country, non-economic and policy factors. The scholarly analysis
of policy effects has remained under-theorized, and poorly connected to general migration
theory.

Because of this lack of precision and specification, it remains unclear how migration
policies affect migration flows when other forces driving international migration are taken
into account. Most empirical models miss out the ‘big picture’ by focusing on short-term
fluctuations on particular migration flows and do not take into account the impact of
policies on overall and long-term migration patterns and trends.

More fundamentally, the contested nature of this debate reveals a still limited
understanding of the forces driving international migration and the lack of theoretically
driven research. Although there is consensus that macro-contextual economic and political

IMI Working Papers Series 2011, No. 32 29

factors and meso-level factors such as networks all play ‘some’ role, there is no agreement
on their relative weight and mutual interaction. To start filling this gap, this paper outlined
the contours of a conceptual framework for generating improved insights into the ways
states and policies shape migration processes in their interaction with structural migration
determinants in receiving and sending countries.

In this paper, I tried to argue that the current research impasse can only be overcome
by firmly embedding the multi-method (quantitative and qualitative), longitudinal empirical
analysis of policy effects into a more comprehensive theoretical framework of the macro-
and meso-level forces driving international migration. I have argued that the fragmented
insights from different disciplinary theories can be integrated in one model through
conceptualizing virtually all manifestations of migration (internal and international) as a
function of capabilities and aspirations to migrate. I also proposed a set of hypotheses on
perverse ‘substitution effects’ of migration policies which can guide future empirical
research.

However, the limited ability of prior research to assess the role of states and policies
and migration processes is not only linked to theoretical problems, but also to concomitant
methodological problems and important limitations. The DEMIG project aims to address
these three problems simultaneously. In a future paper, I intend to further work out
methodological and empirical approaches for testing policy, origin and destination effects.
Nevertheless, from this paper it may already be clear that, in order to be tested, the key
hypotheses about potential substitution effects require particular data and methodological
approaches.

First of all, spatial substitution effects can only be studied through ‘double
comparative’ approaches which simultaneously study the migration of multiple origin
groups to and from multiple destination countries.13 Such double comparative analyses
require the availability of bilateral (country-to-country) flow data. Also for studying inter-
temporal substitution effects, a key requirement is the availability of bilateral flow data
which preferably spans several decades. The theoretical relevance of reverse flow
substitution effects reveals the need to consider immigration and emigration as separate
social phenomena (‘net migration’ largely being a statistical artefact with little sociological
meaning) which require aggregate and, preferably, bilateral migration data that
differentiate between outflows and inflows. The study of categorical substitution effects
requires migration flow data which differentiate between the different migrant categories.

Therefore, besides the development of a conceptual framework the contours of which
have been presented in this paper, the second aim of the DEMIG project is to create a
longitudinal database compiling bilateral (country-to-country) migration flow data over the
1950–2010 period, based on flow data collected from national statistical offices and other
sources – particularly, but not exclusively, from OECD countries. A third aim and project
activity is to systematically review receiving- and sending-country migration policies (cf.
Czaika and de Haas 2011a; de Haas and Vezzoli 2011), which will result in methodological
choices with regards to the operationalization of particular migration policy instruments
(e.g. visas, bilateral agreements, recruitment) on which to focus the empirical analyses. In
fact, the parsimonious operationalization of migration policies is the key challenge facing

A similar approach has been used by Van Tubergen et al. (2004) in studying the economic incorporation of
immigrants in 18 Western countries.

30 IMI Working Papers Series 2011, No. 32

this project and migration determinants research more generally. Subsequently, this
framework will be subjected to empirical tests drawing on gross and bilateral (country-to-
country) migration flow data.

However, it is important to emphasize that not all problems can be ‘fixed’ just by
collecting better data and specifying better quantitative models. Ultimately, empirical
research should be theory- and not data-driven, and the point is that many theoretically
relevant structural factors are indeed difficult to quantify. There are serious limitations in
the availability of reliable indicators and it would also be naïve to assume that such
indicators can capture all relevant dimensions (e.g. the concepts of ‘power’, ‘exclusion’,
‘racism’, ‘discrimination’ or ‘migration policy’) of such structural features. Empirical-
quantitative models should be improved as much as possible. However, this cannot solve all
problems, and the ‘non-quantifiability’ of certain factors should not be a reason to ignore
them.

To combine the different strengths of quantitative and qualitative approaches,
methodological triangulation seems to be a more promising avenue. Such an approach
systematically combines formal quantitative tests of key indicators using panel datasets with
detailed case studies studying the relation between transformations of economic structures
and labour markets and migration patterns for particular countries or regional blocks. Such
case studies should provide an empirically ‘thick’, informed description, supplemented,
whenever possible, with exploratory quantitative analysis. This can serve to develop new
ideas and hypotheses as well as a ‘plausibility-check’ of results generated by formal tests.
Policy reviews should also include a qualitative assessment of the effects and effectiveness
of these policies, from which hypotheses can be derived. Because much information on
policies will be lost through quantification, the qualitative review and categorization of
migration policies has a value in itself, and contributes to the improvement of the
conceptual framework.

Methodological heterodoxy and true interdisciplinary openness are therefore central
conditions for advancing research on migration determinants. Through creatively integrating
qualitative and quantitative approaches, it is possible to increase insights into the nature
and evolution of migration policies and their effects on the size, direction, timing and
composition of migration flows. Eventually, such an open, creative and flexible approach will
enhance our ability to create a generalized theoretical understanding of the determinants of
international migration.

IMI Working Papers Series 2011, No. 32 31

References

Arango, J. (2000) ‘Explaining migration: a critical view’, International Social Science Journal
52(165): 283–96.

Bakewell, O. (2010) ‘Some reflections on structure and agency in migration theory’, Journal
of Ethnic and Migration Studies 36(10)1689–708.

Bauer, Th. and K. Zimmermann (1998) ‘Causes of international migration: a survey’, in P.
Gorter, P. Nijkamp and J. Poot (eds.), Crossing Borders: Regional and Urban Perspectives
on International Migration, Aldershot: Ashgate.

Berlin, I. (1969) Four Essays On Liberty, Oxford: Oxford University Press.

Bhagwati, J. (2003) ‘Borders beyond control’, Foreign Affairs, January/February: 98–104.

Böcker, A. (1994) ‘Chain migration over legally closed borders: settled migrants as
bridgeheads and gatekeepers’, Netherlands’ Journal of Social Sciences 30(2): 87–106.

Borjas, G. J. (1989) ‘Economic theory and international migration’, International Migration
Review 23: 457–85.

Borjas, G. J. (1990) Friend or Strangers: The Impact of Immigrants on the US Economy, New
York: Basic Books.

Boswell, C. (2008) ‘Combining economics and sociology in migration theory’, Journal of
Ethnic and Migration Studies 34(4): 549–66.

Brochmann, G. and T. Hammar (eds.) (1999) Mechanisms of Immigration Control, Oxford
and New York: Berg.

Burkhart, R. E. and M. S. Lewis-Beck (1994) ‘Comparative democracy: the economic
development thesis’, The American Political Science Review 88(4): 903–10.

Carling, J. (2002) ‘Migration in the age of involuntary immobility: theoretical reflections and
Cape Verdean experiences’, Journal of Ethnic and Migration Studies 28(1): 5–42.

Castles, S. (2002) ‘Migration and community formation under conditions of globalization’,
International Migration Review 36(4): 1143–68.

Castles, S. (2004a) ‘The factors that make and unmake migration policies’, International
Migration Review 38(3): 852–84.

Castles, S. (2004b) ‘Why migration policies fail’, Ethnic and Racial Studies 27(2): 205–27.

Castles, S. (2010) ‘Understanding global migration: a social transformation perspective’,
Journal of Ethnic and Migration Studies 36(10): 1565–86.

Castles, S. and M. J. Miller (2009) The Age of Migration, Basingstoke and London: Macmillan.

Collins, R. (1994) Four Sociological Traditions, New York and Oxford: Oxford University Press.

Collyer, M. (2006) ‘States of Insecurity: Consequences of Saharan Transit Migration’,
Working Paper 31, University of Oxford: Centre on Migration, Policy and Society
(COMPAS).

32 IMI Working Papers Series 2011, No. 32

Czaika, M. (2011) ‘Migration and Social Fractionalization: Double Deprivation as a
Behavioural Link’, IMI Working Paper 36 (DEMIG Project Paper 6), University of Oxford:
International Migration Institute.

Czaika, M. and H. de Haas (2011a) ‘The Effectiveness of Immigration Policies: A Conceptual
Review of Empirical Evidence’, IMI Working Paper 33 (DEMIG Project Paper 3), University
of Oxford: International Migration Institute.

Czaika, M. and H. de Haas (2011b) ‘The Role of Internal and International Relative
Deprivation in Global Migration’, IMI Working Paper 35 (DEMIG Project Paper 5),
University of Oxford: International Migration Institute.

de Haas, H. (2007) ‘Turning the tide? Why development will not stop migration’,
Development and Change 38(5): 819–41.

de Haas, H. (2008) ‘North-African migration systems: evolution, transformations and
development linkages’, in S. Castles and R. Delgado Wise (eds.), Migration and
Development: Perspectives from the South, Geneva: International Organization for
Migration.

de Haas, H. (2009) Mobility and Human Development, New York: UNDP.

de Haas, H. (2010a) ‘Migration Transitions: a Theoretical and Empirical Inquiry into the
Developmental Drivers of International Migration’, IMI Working Paper 24 (DEMIG Project
Paper 1), University of Oxford: International Migration Institute.

de Haas, H. (2010b) ‘The internal dynamics of migration processes: a theoretical inquiry’,
Journal of Ethnic and Migration Studies 36(10): 1587–617.

de Haas, H. and S. Vezzoli (2011) ‘Leaving Matters: The Nature, Evolution and Effects of
Emigration Policies’, IMI Working Paper 34 (DEMIG Project Paper 4), University of Oxford:
International Migration Institute.

Düvell, F. (2005) Illegal Immigration in Europe: Beyond Control, Basingstoke: Palgrave
Macmillan.

Emirbayer, M. and A. Mische (1998) ‘What is agency?’, American Journal of Sociology
103(4): 962–1023.

Entzinger, H. (1985) ‘Return migration in Western Europe: current policy trends and their
implications, in particular for the second generation’, International Migration XXIII(2):
263–90.

Faist, T. (1997) ‘The crucial meso-level’, in T. Hammar, G. Brochmann, K. Tamas and T. Faist
(eds.), International Migration, Immobility and Development, Oxford: Berg.

Fawcett, J. T. (1989) ‘Networks, linkages, and migration systems’, International Migration
Review 23: 671–80.

Grütters, C. A. F. M. (2003) Asieldynamiek, een systeemdynamische analyse van de
Nederlandse asielprocedure in de periode 1980 – 2002, Tilburg: Wolf Legal Publishers.

Harris, J. R. and M. P. Todaro (1970) ‘Migration, unemployment and development: a two-
sector analysis’, American Economic Review 60: 126–42.

Harris, N. (2002) Thinking the Unthinkable: The Immigration Myth Exposed, London: I. B.
Tauris.

IMI Working Papers Series 2011, No. 32 33

Hatton, T. J. (2004) ‘Seeking asylum in Europe’, Economic Policy 19: 5–62.

Hatton, T. J. and J. G. Williamson (1998) The Age of Mass Migration: Causes and Economic
Impact, New York: Oxford University Press.

Kritz, M. M., L. L. Lim, and H. Zlotnik (eds.) (1992) International Migration System: A Global
Approach, Oxford: Clarendon Press.

Kureková, L. (2011) ‘From job search to skill search: political economy of labour migration in
Central and Eastern Europe’, PhD thesis, Department of International Relations and
European Studies, Budapest: Central European University.

Lee, E. S. (1966) ‘A theory of migration’, Demography 3(1): 47–57.

Lindley, A. (2007) ‘The Early Morning Phonecall: Remittances from a Refugee Diaspora
Perspective’, Working Paper 47, University of Oxford: Centre on Migration, Policy and
Society (COMPAS).

Lucas, R. E. B. and O. Stark (1985) ‘Motivations to remit: evidence from Botswana’, Journal
of Political Economy 93: 901–18.

Mabogunje, A. L. (1970) ‘Systems approach to a theory of rural–urban migration’,
Geographical Analysis 2(1): 1–18.

Massey, D.S. (1991) ‘Economic development and international migration: comparitive
perspective’, in S. Díaz-Briquets and S. Weintraub (eds.), Determinants of Migration from
Mexico, Central America, and the Carribean, Boulder, CO: Westview Press.

Massey, D. S. (1990) ‘Social structure, household strategies, and the cumulative causation of
migration’, Population Index 56: 3–26.

Massey, D. S. (2004) ‘Social and economic aspects of immigration’, Annals of the New York
Academy of Sciences 1038(1): 206–12.

Massey, D. S., J. Arango, G. Hugo, A. Kouaouci, A. Pellegrino and J. E. Taylor (1993) ‘Theories
of international migration: a review and appraisal’, Population and Development Review
19(3): 431–66.

Massey, D. S., J. Arango, G. Hugo, A. Kouaouci, A. Pellegrino and J. E. Taylor (1998) Worlds in
Motion: Understanding International Migration at the End of the Millennium, Oxford:
Clarendon Press.

McKenzie, D. J. (2005) Paper Walls are Easier to Tear Down: Passport Costs and Legal
Barriers to Emigration, Washington DC: World Bank.

Peach, C. (1968) West Indian Migration to Britain: A Social Geography, Oxford: Oxford
University Press, for the Institute of Race Relations.

Penninx, R. (1982) ‘A critical review of theory and practice: the case of Turkey’, International
Migration Review 16(4): 781–818.

Piore, M. J. (1979) Birds of Passage: Migrant Labor and Industrial Societies, Cambridge:
Cambridge University Press.

Portes, A. (1998) ‘Social capital: its origins and applications in modern sociology’, Annual
Review of Sociology 24: 1–24.

34 IMI Working Papers Series 2011, No. 32

Portes, A. and J. Böröcz (1987) ‘Contemporary immigration: theoretical perspectives on its
determinants and modes of incorporation’, International Migration Review 23: 606–30.

Ravenstein, E. G. (1885) ‘The laws of migration’, Journal of the Royal Statistical Society 48:
167–227.

Ravenstein, E. G. (1889) ‘The laws of migration’, Journal of the Royal Statistical Society 52:
214–301.

Reniers, G. (1999) ‘On the history and selectivity of Turkish and Moroccan migration to
Belgium’, International Migration 37(4): 679–713.

Ruhs, M. and P. Martin (2008) ‘Numbers vs. rights: trade-offs and guest worker programs’,
International Migration Review 42(1): 249–65.

Salt, J. (1987) ‘Contemporary trends in international migration study’, International
Migration 25(3): 241–51.

Sandell, R. (2006) ‘Spain’s Immigration Experience: Lessons to be Learned from Looking at
the Statistics’, Working Paper 30/2006: Real Institute Elcano.

Sassen, S. (1988) The Mobility of Labor and Capital: A Study in International Investment and
Labor Flow, Cambridge: Cambridge University Press.

Sen, A. (1997) ‘Editorial: human capital and human capability’, World Development 25(12):
1959–61.

Sen, A. (1999) Development as Freedom, New York: Anchor Books.

Shadid, W. A. (1979) ‘Moroccan Workers in the Netherlands’, Leiden: University of Leiden.

Sjaastad, A. H. (1962) ‘The costs and returns of human migration’, Journal of Political
Economy 70(5): 80–93.

Skeldon, R. (1997) Migration and Development: A Global Perspective, Essex: Longman.

Skeldon, R. (2002) ‘Migration and poverty’, Asia–Pacific Population Journal 17(4): 67–82.

Stark, O. and J. Taylor (1989) ‘Relative deprivation and international migration’,
Demography 26: 1–14.

Stark, O. and J. E. Taylor (1991) ‘Migration incentives, migration types: the role of relative
deprivation’, Economic Journal 101(408): 1163–78.

Stark, O. (1991) The Migration of Labor, Cambridge and Oxford: Blackwell.

Stark, O. and D. E. Bloom (1985) ‘The New Economics of Labor Migration.’ American
Economic Review 75:173-78.

Strikwerda, C. (1999) ‘Tides of migration, currents of history: the state, economy, and the
transatlantic movement of labor in the nineteenth and twentieth centuries’,
International Review of Social History 44: 367–94.

Taylor, J. E. (1999) ‘The new economics of labour migration and the role of remittances in
the migration process’, International Migration 37(1): 63–88.

Thielemann, E. R. (2006) ‘The effectiveness of governments’ attempts to control unwanted
migration’, in C. Parsons and T. Smeeding (eds.), Immigration and the Transformation of
Europe, Cambridge: Cambridge University Press.

IMI Working Papers Series 2011, No. 32 35

Todaro, M. P. (1969) ‘A model of labor migration and urban unemployment in less-
developed countries’, American Economic Review 59: 138–48.

Van Amersfoort, J. M. M. (1998) ‘Migratie en Migratietheorieën’, in Münstermann and
Entzinger Penninx (eds.), Etnische Minderheden en de Multiculturele Samenleving,
Groningen: Wolters-Noordhoff.

Van Amersfoort (2010) Een som van misverstanden: Het kabinet Den Uyl en de immigratie
van Surinamers, Justitiële Verkenningen 36(2): 109-121.

van der Erf, R. (2003) ‘Asieldruk op West-Europa onverminderd hoog: Waterbed blijft maar
in beweging’, Demos Jaargang 19.

van Liempt, I. and J. Doomernik (2006) ‘Migrants’ agency in the smuggling process: the
perspectives of smuggled migrants in the Netherlands’, International Migration 44(4):
165–90.

van Tubergen, F., I. Maas, and H. Flap (2004) ‘The economic incorporation of immigrants in
18 western societies: origin, destination, and community effects’, American Sociological
Review 69(5): 704–27.

Wiener, J. (2005) ‘Working-Class Republicans and “False Consciousness”’, Dissent, Spring
2005.

Zelinsky, Z. (1971) ‘The hypothesis of the mobility transition’, Geographical Review 61(2):
219–49.

Turn in your highest-quality paper
Get a qualified writer to help you with

“ review ”

Get high-quality paper

Guarantee! All work is written by expert writers!

Still stressed from student homework?

Get quality assistance from academic writers!

Order now