Attitudes toward Highly Skilled and Low-skilled Immigration: Evidence froma Survey Experiment

American Political Science Review
Vol. 104, No. 1
February 2010
doi:10.1017/S0003055409990372
Attitudes toward Highly Skilled and Low-skilled Immigration:
Evidence from a Survey Experiment
JENS HAINMUELLER Massachusetts Institute of Technology
MICHAEL J. HISCOX Harvard University
Pastresearchhasemphasizedtwocriticaleconomicconcernsthatappeartogenerateanti-immigrant
sentiment among native citizens: concerns about labor market competition and concerns about
the fiscal burden on public services. We provide direct tests of both models of attitude formation
using an original survey experiment embedded in a nationwide U.S. survey. The labor market competition
model predicts that natives will be most opposed to immigrants who have skill levels similar to their own.
We find instead that both low-skilled and highly skilled natives strongly prefer highly skilled immigrants
over low-skilled immigrants, and this preference is not decreasing in natives’ skill levels. The fiscal burden
model anticipates that rich natives oppose low-skilled immigration more than poor natives, and that this
gap is larger in states with greater fiscal exposure (in terms of immigrant access to public services). We
find instead that rich and poor natives are equally opposed to low-skilled immigration in general. In
states with high fiscal exposure, poor (rich) natives are more (less) opposed to low-skilled immigration
than they are elsewhere. This indicates that concerns among poor natives about constraints on welfare
benefits as a result of immigration are more relevant than concerns among the rich about increased taxes.
Overall the results suggest that economic self-interest, at least as currently theorized, does not explain
voter attitudes toward immigration. The results are consistent with alternative arguments emphasizing
noneconomic concerns associated with ethnocentrism or sociotropic considerations about how the local
economy as a whole may be affected by immigration.
Whydopeopleopposeorfavorimmigration? sentimentandthatindividualattitudestowardimmi-
Recent scholarly work examining survey data
gration are profoundly shaped by fears about labor
on individual attitudes toward immigration
market competition (Kessler 2001; Mayda 2006; Scheve
has generated inconsistent findings and no clear con-
and Slaughter 2001) and/or the fiscal burden on public
sensus view. Many studies suggest that opposition
services (Facchini and Mayda 2009; Hanson 2005;
to immigration is primarily driven by noneconomic
Hanson, Scheve, and Slaughter 2007). Borjas (1999)
concerns associated with cultural and ethnic tensions
identifies these as the two critical economic issues that
between native and immigrant populations (Bauer,
have dominated the debate over immigration policy
Lofstrom, and Zimmerman 2000; Burns and Gim-
in the United States. Simon (1989) has identified them
pel 2000; Chandler and Tsai 2001; Citrin et al. 1997;
as the two key concerns motivating anti-immigrant
Dustmann and Preston 2007; Espenshade and Hemp-
sentiment in Britain. But there is no agreement among
stead 1996; Fetzer 2000; Gang, Rivera-Batiz, and Yun
scholars about the relative impact of these different
2002; Lahav 2004; McLaren 2003). These studies em-
types of economic concerns or how they compare in
phasize noneconomic differences between individuals
importance with noneconomic considerations that
in terms of ethnocentrism and ideology in explaining
also motivate anti-immigrant sentiment. Resolving
attitudes toward immigrants and connect to an exten-
these questions is critical for understanding public
sive body of empirical research indicating that material
opposition to immigration and the growth of extremist,
self-interest rarely plays a role in shaping people’s opin-
often violent, anti-immigrant political movements.
ions about major policy issues (Kinder and Sears 1981;
One reason there is no consensus on why people
Sears and Funk 1990; Sears et al. 1980).
support or oppose immigration is that the data on indi-
A very different set of studies argue that material
vidual attitudes are ill-suited to testing the theoretical
economic concerns lie at the heart of anti-immigrant
relationships at issue. Studies examining economic
concerns about immigration typically begin with a
general equilibrium model and derive predictions
Jens Hainmueller is Assistant Professor, Department of Political
Science, Massachusetts Institute of Technology, 77 Massachusetts
about how native citizens who own different types
Avenue, Cambridge, MA 02139 (jhainm@mit.edu).
of productive factors, and who have different levels
Michael J. Hiscox is Clarence Dillon Professor of Interna-
of income, will differ in their views regarding highly
tional Affairs, Department of Government, Harvard Univer-
skilled and low-skilled immigration (Facchini and
sity, 1737 Cambridge Street, Cambridge, MA 02138 (hiscox@fas.
harvard.edu).
Mayda 2009; Hanson, Scheve, and Slaughter 2007;
Both authors are affiliated with Harvard’s Institute for Quan-
Mayda 2006; Scheve and Slaughter 2001). However,
titative Social Science (IQSS) which generously provided funding
due to data constraints, none of these studies have been
for the survey. We thank Alberto Abadie, George Borjas, Giovanni
able to test these specific predictions directly. They rely
Facchini, Gordon Hanson, Gary King, David Lynch, Anna Mayda,
instead upon indirect tests that leave the interpretation
Dani Rodrik, Ken Scheve, Matthew Slaughter, Dustin Tingley, the
co-editors, and five anonymous reviewers for very helpful comments.
of the results wide open. In particular, no study to
The usual disclaimer applies.
date has been able to distinguish between attitudes
1

---

Attitudes toward Immigration
February 2010
toward highly skilled immigrants and attitudes toward
Overall, the results indicate that existing political-
low-skilled immigrants, even though this distinction
economic models do not provide reliable guides to
is a critical feature of the theoretical story about
individual attitudes toward immigration. Material self-
how economic concerns affect attitude formation and
interest, at least as currently theorized, does not appear
policy preferences with respect to immigration.
to be a powerful determinant of anti-immigrant senti-
To test claims about how economic concerns shape
ment. The results are more consistent with alternative
attitudes toward immigration, we conducted a unique
arguments about attitude formation that emphasize
survey experiment that, for the first time, explicitly
noneconomic concerns among voters, associated with
and separately examines individuals’ attitudes toward
ethnocentrism or sociotropic considerations about how
highly skilled and low-skilled immigrants. In a nation-
the local economy as a whole may be affected by im-
wide U.S. survey, we randomly assigned respondents
migration.
to answer questions about immigrants with different
skill levels, thereby obtaining an unbiased comparison
between the distributions of attitudes toward highly
ECONOMIC CONCERNS AND
skilled and low-skilled immigrants. This comparison,
ATTITUDES TOWARD IMMIGRATION
and how it varies with respondent characteristics, al-
Although immigration may impact the native economy
lows us to directly test the predictions of the theoretical
in many ways, recent research has emphasized two
models about how economic concerns affect attitudes
critical economic concerns that could generate anti-
toward immigration.
immigrant sentiment among native citizens: concerns
The experiment yields results that present a major
about labor market competition and fears about the
challenge for existing political-economic models and
fiscal burden on public services. General equilibrium
the conclusions reached in many well-cited studies of
models of the native economy generate a variety of
attitudes toward immigration. The prominent labor
predictions about how natives with particular skill and
market competition model predicts that natives will
income characteristics should be affected by inflows of
be most opposed to immigrants who have skill levels
immigrants.
similar to their own. This is rejected by the data. We
find that both highly skilled and low-skilled respon-
dents strongly prefer highly skilled immigrants over
Labor Market Competition
low-skilled immigrants, and this preference is not de-
Analysis of the income effects of immigration typically
creasing in respondents’ skill levels. Support for both
begins with a closed-economy “factor-proportions”
highly skilled and low-skilled immigration is strongly
(FP) analysis (Borjas 1999; Borjas, Freeman, and Katz
increasing in respondents’ skill levels. In addition, these
1996, 1997). The FP model derives the distributional
relationships are similar for the subsamples of respon-
effects in the native economy from the impact that
dents that are currently in or currently out of the labor
immigration has on the relative supplies of factors
force. The results suggest that, among natives generally,
of production. If immigrants have low skill endow-
labor market competition is not a significant motivator
ments compared with natives, immigration will raise
of anti-immigrant sentiment.
the supply of low-skilled labor relative to other fac-
The fiscal burden model anticipates that rich (high-
tors (including highly skilled labor). These changes in
income) natives oppose low-skilled immigration and
relative factor supplies translate into changes in real
favor highly skilled immigration more than do poor
factor returns: wages of native low-skilled workers will
(low-income) natives, and that this difference should
fall as new (low-skilled) immigrants price themselves
be more pronounced in states with greater fiscal ex-
into employment; and, as more low-skilled labor is ap-
posure in terms of immigrant access to public ser-
plied to fixed amounts of the other factors, the real
vices. We find instead that rich and poor natives both
wages of highly skilled workers will rise. The reverse
equally prefer highly skilled over low-skilled immigra-
effects are expected in the case of inflows of highly
tion most of the time. In addition, the premium at-
skilled immigrants, which will drive up the real wages
tached to highly skilled versus low-skilled immigration
of low-skilled natives while reducing real returns for
is decreasing with the income levels of natives in states
highly skilled natives. Depending on what one assumes
with high fiscal exposure, where the welfare effects are
about wage flexibility, the impact of competition with
expected to be strongest. Rich natives are actually less
similarly skilled immigrants may also be manifested
opposed to low-skilled immigration in states with high
in higher rates of unemployment among natives.1
fiscal exposure than they are elsewhere. These results
The
are inconsistent with claims that rich natives are op-
FP model generates a clear prediction about attitudes
posed to low-skilled immigrants because they antici-
toward immigration: natives should oppose immigrants
pate a heavier tax burden associated with the provi-
with similar skill levels but favor immigrants with dif-
sion of public services. Moreover, we do find evidence
ferent skill levels.2
that poor natives are more opposed to low-skilled im-
migration in states with greater fiscal exposure than
1
they are elsewhere, suggesting that concerns about ac-
Alternative models also allow for geographic concentration of
cess to or overcrowding of public services contribute
wage and employment effects. See Card (1990) or Borjas (1999).
2 An online Appendix with formal derivations of these relationships
to anti-immigrant attitudes among poorer native
(as well as the relationships posited by the fiscal burden model) is
citizens.
available on the authors’ Web site. Notice that the predictions from
2

---

American Political Science Review
Vol. 104, No. 1
Empirical studies have found mixed results when
and Nelson 2000; Scheve and Slaughter 2001, 135–37).
testing this model (Burns and Gimpel 2000; Citrin et al.
In an open-economy Heckscher–Ohlin model, trade
1997; Dustmann and Preston 2006; Fetzer 2000; Gang,
can offset the impact of immigration as the output
Rivera-Batiz, and Yun 2002; Harwood 1986), although
mix of tradable goods changes in line with changes
two prominent articles have recently reported strong
in factor supplies. Assuming that the local economy
supporting evidence. Drawing upon data from the Na-
is not large relative to the rest of the world and/or
tional Election Studies (NES) surveys in the United
that inflows of immigrants are small relative to the lo-
States in the 1990s, Scheve and Slaughter (2001) find a
cal labor supply, local wages will not be affected—–the
strong positive correlation between respondents’ skill
“factor price insensitivity” result (Leamer and Levin-
levels, as measured by years of education, and stated
sohn 1995). In an amended open-economy model in
support for immigration. This correlation is interpreted
which skills of workers are highly specific to partic-
as evidence that low-skilled (less educated) natives fear
ular industries (Grossman and Helpman 1994; Jones
being forced to compete for jobs with low-skilled immi-
1971), the predictions match those from the FP analysis
grants. In a similar study Mayda (2006) examined cross-
only as long as all goods are traded (then natives will
national survey data from the 1995 National Identity
be disadvantaged by immigrants of similar skills lev-
Module of the International Social Survey Programme
els, regardless of industry specificity among the highly
(ISSP), as well as data collected between 1995 and 1997
skilled). But the real income effects are sensitive to the
by the World Values Survey (WVS) and finds that
inclusion of nontraded goods. Immigration can lead
the probability of voicing pro-immigration opinions
to a reduction in the price of nontraded goods (by
is positively associated with the skill levels of survey
raising the output of such goods more rapidly than it
respondents (measured by years of education). Again,
raises aggregate demand for them), and so it becomes
this correlation is presented as confirmation that con-
unclear whether native workers with skills similar to
cerns about labor market competition are a powerful
those of immigrants will be worse off in real terms (this
motivator of attitudes toward immigrants.
will depend in part on their consumption tastes).4 In
There are four main reasons to be wary of these
alternative types of open-economy models that allow
reported findings. First, it is unclear whether respon-
for economies of scale in production in the industries
dents can plausibly observe and correctly attribute the
employing immigrants, inflows of new workers can be
income effects of immigration that are anticipated in
shown to generate higher real wages for native workers
the FP model. A growing set of empirical studies has
with similar skills (Brezis and Krugman 1993). There
examined the effect of immigration on native wages
is, in short, a great deal of theoretical ambiguity about
and unemployment, but the evidence remains hotly
the labor market effects of immigration and the related
debated.3 Some studies claim large, adverse wage and
concerns we should expect to observe among native
employment effects of immigration on less educated
citizens.
workers (Borjas 1999, 2003, 2005; Borjas, Freeman,
Third, a variety of alternative explanations can ac-
and Katz 1996, 1997), whereas others conclude that
count for the positive correlation between education
the immigration effects are at most very small, and
and pro-immigration attitudes. Several studies have
possibly insignificant (Card 1990, 2001, 2007; Lewis
shown that more educated respondents tend to exhibit
2005). In a recent study Ottaviano and Peri (2008)
higher levels of ethnic and racial tolerance, stronger
find a net positive long-term effect of immigration
preferences for cultural diversity, and more economic
on average wages of natives. The inconclusiveness of
knowledge, all of which can lead them to favor im-
the empirical research on the labor market effects of
migration more than their less educated counterparts
immigration suggests the need for caution in using the
(Chandler and Tsai 2001; Citrin et al. 1997; Dustmann
simple FP model to make predictions about attitude
and Preston 2007; Fetzer 2000; Gang, Rivera-Batiz, and
formation and interpreting the evidence on attitudes.
Yun 2002; Hainmueller and Hiscox 2007). Existing tests
Second, in line with the mixed empirical evidence
are not equipped to discriminate between these claims
on the impact of immigration, many scholars have
and the argument that the association between edu-
pointed out that when we move away from the FP
cation and views about immigrants is due to concerns
analysis and consider more sophisticated economic
about labor market competition.5
models, it becomes very difficult to make clear pre-
Fourth and finally, all the above-mentioned tests that
dictions about the equilibrium effects of immigration
have examined attitudes toward immigration and tried
on wages and employment opportunities among na-
to link them to concerns about labor market com-
tive workers (see Friedberg and Hunt 1995; Gaston
petition have relied upon responses to survey ques-
tions that ask individuals about their attitudes toward
immigration in general and do not differentiate be-
the FP model, although widely taken as central in the literature, are
tween highly skilled and low-skilled immigrants.6 This
far from general (see discussion below).
3 For general reviews about the impact of immigration on wages
4
and employment see for example Bhagwati (2002), Borjas (1999),
See Hainmueller and Hiscox (2007). Specificity aside, a similar
Card (2005), Friedberg and Hunt (1995), and Longhi, Nijkamp, and
result is obtained in models in which factors outnumber traded goods.
Poot (2005). In a recent study, Borjas (2003, 1335) summarizes the
5 The same problem applies to a large body of studies that examine
evidence, observing that “the measured impact of immigration on
attitudes toward international trade and globalization more gener-
the wage of native workers fluctuates widely from study to study
ally (see Hainmueller and Hiscox 2006).
(and sometimes even within the same study) but seems to cluster
6 Scheve and Slaughter (2001) used responses to the NES immi-
around zero.”
gration question “Do you think the number of immigrants from
3

---

Attitudes toward Immigration
February 2010
is highly problematic because the key prediction of
The Fiscal Burden of Public Services
the simple FP model is that natives should oppose
immigrants with skill levels similar to their own but
The second critical economic concern associated with
support immigrants with different skill levels. Previous
immigration involves the immigrants’ use of public ser-
tests rely on the assumption that respondents have low-
vices (including public education and health services
skilled immigrants in mind when answering questions
and various types of welfare assistance, as well as basic
about immigration in general. This assumption is ques-
services such as police and fire protection, roads, parks,
tionable, given that respondents are likely to have sys-
and amenities) and their contribution to tax revenues.
tematically varying information about and perceptions
The standard approach to the analysis is to incorporate
of the skill attributes of immigrants. More educated
a simple model of public finance into the FP analysis of
respondents may be better informed about current im-
immigration (Facchini and Mayda 2009; Hanson 2005;
migration flows, for instance, and are likely to recog-
Hanson, Scheve, and Slaughter 2007). This approach al-
nize the considerable share of inflows accounted for by
lows immigration not only to affect the pretax incomes
skilled foreigners entering many Western nations (of-
of native individuals, but also to separately affect after-
ten because immigration policies are aimed explicitly
tax incomes via taxes and transfers. The predictions
at selecting immigrants based on their skill levels). It
depend on two key assumptions about (1) the net con-
is well known that such varying perceptions can lead
tributions of low-skilled and highly skilled immigrants
to biased estimates in survey research (Bertrand and
to the tax coffers and (2) the institutional mechanism
Mullainathan 2001; King et al. 2004). And of course,
in place to adjust taxes and transfers in response to
employing this questionable assumption still does not
fiscal imbalances. It is assumed that low-skilled im-
allow one to examine whether the skill levels of na-
migrants impose a substantial net burden on public
tives affect their attitudes toward highly skilled immi-
finance, whereas highly skilled immigrants are net con-
grants in the expected way. A complete and direct test
tributors in terms of taxes. There are two plausible
would ask respondents about their attitudes toward
institutional mechanisms that have been considered,
low-skilled immigrants and highly skilled immigrants
assuming the government must balance its budget: a
specifically and separately.
change in tax rates and a change in per capita transfers
The only previous study that comes close to such
(see Facchini and Mayda 2009).7 In the most commonly
a test actually reports results at odds with the recent
studied scenario, assuming the government adjusts tax
claims that labor market concerns are powerful shapers
rates while keeping per capita transfers constant, the
of attitudes. Hainmueller and Hiscox (2007) investi-
prediction is that rich (high-income) natives should
gate survey data for 22 European countries from the
prefer highly skilled over low-skilled immigrants more
European Social Survey, in which respondents were
than do poor (low-income) natives, because the skill
asked about their attitudes toward immigration from
levels of immigrants determine their fiscal impact, and
“richer” and “poorer” countries, a difference plausi-
progressivity in taxation implies that the rich benefit
bly associated with the expected average skill levels of
(lose) more from any associated reduction (increase)
immigrants. They find that in all 22 countries people
in taxes. In the alternative scenario, assuming the gov-
with higher education levels (and/or higher levels of
ernment adjusts per capita transfers but holds tax rates
occupational skills) are more likely to favor immigra-
constant, the prediction is the opposite: poor natives
tion regardless of where the immigrants come from and
prefer highly skilled over low-skilled immigrants more
their likely skill attributes. In addition, the positive link
than rich natives, because low-skilled immigrants tend
between education and support for (all types of) immi-
to crowd out poor natives in terms of access to public
gration is almost identical between those in the labor
services and erode their welfare benefits, whereas rich
force and those not in the labor force. Taken together,
natives are unaffected.
the existing theory and evidence on whether concerns
Two recent empirical studies have examined these
about labor market competition are a strong motivator
claims. Hanson, Scheve, and Slaughter (2007) use NES
of anti-immigrant sentiment remain ambiguous. At the
survey data to compare individual attitudes toward im-
very least, more complete and direct empirical tests are
migration in different U.S. states and find that rich indi-
necessary.
viduals are less likely to support immigration in states
that are highly exposed to fiscal costs as a result of
immigration (i.e., states with generous public services
and high rates of immigrant settlement) than in states
with lower exposure. This finding is interpreted as con-
foreign countries who are permitted to come to the United States
firmation that, as expected in the scenario in which the
to live should be increased a little, increased a lot, decreased a lit-
tle, decreased a lot, or left the same as it is now?” Mayda (2006)
government adjusts taxes to meet new spending obli-
examined answers to the ISSP question “Do you think the number
gations, rich natives fear being burdened with higher
of immigrants to (respondents country) nowadays should be: (a)
taxes as a consequence of low-skilled immigrants draw-
reduced a lot, (b) reduced a little, (c) remain the same as it is, (d)
ing on public services and draining government coffers.
increased a little, or (e) increased a lot.” The WVS asked the follow-
ing question: “How about people from other countries coming here
to work. Which one of the following do you think the government
7 Borrowing would be a third option, but as there are constitutional
should do (a) Let anyone come who wants to (b) Let people come as
constraints on borrowing by state governments in the United States,
long as there are jobs available (c) Place strict limits on the number
and the underlying model is static, standard analyses do not con-
of foreigners who can come here (d) Prohibit people coming here
sider this possibility (Facchini and Mayda 2009; Hanson, Scheve, and
from other countries? (e) Don’t know.”
Slaughter 2007).
4

---

American Political Science Review
Vol. 104, No. 1
Facchini and Mayda (2009) examine the cross-national
time period.9 It seems unlikely, then, that U.S. survey
survey data from the ISSP and find that respondent
respondents could be drawing on personal experience
income is negatively correlated with support for immi-
to attribute tax hikes to immigration.
gration in countries where low-skilled immigrants are
On the other hand, a recent study that looks at
a larger share of total immigration inflows. This finding
the link between immigration and U.S. state welfare
is also regarded as evidence that fears about higher
expenditures has found stronger support for the
taxes among rich natives, linked to use of public ser-
so-called “erosion hypothesis.” Hero and Preuhs
vices by low-skilled immigrants, lead to anti-immigrant
(2007) examine data on welfare spending for all
sentiments.
U.S. states in 1998 and find that states with larger
Again, there are reasons to treat these findings with
noncitizen populations tend to provide smaller cash
considerable caution. While there is some evidence
benefits in their welfare programs, and this effect is
that immigrants rely more on welfare programs than
larger the more accessible the welfare programs are
do native citizens (Borjas 1999; Fix and Passel 2002;
to immigrants. In the right panel of Figure 1 we plot
Hanson 2005; Zimmerman and Tumlin 1999), as im-
changes in state public welfare expenditures per capita
migrant households tend to be larger and poorer than
against changes in the immigrant population. There is
native households, there is more disagreement over
a negative correlation between the two. Although all
the extent to which immigrant inflows increase net tax
states have expanded per capita welfare expenditures
burdens on natives (Fix, Passel, and Enchautegui 1994;
over time, the increases have been smaller in states that
Smith and Edmonston 1997). A U.S. study conducted
experienced larger increases in the share of immigrants
by the National Research Council (NRC) reported
in their population.10 This pattern, taken together with
that the average immigrant to the United States could
the evidence on state taxes discussed above, suggests
be expected to impose a tax burden on natives in
that fears about the erosion of welfare benefits as a
the short term, but would be a net contributor to tax
result of immigration may actually be more relevant
coffers in the long term, to the tune of $80,000 (see
and plausible than worries about tax hikes.
Smith and Edmonston 1997).8 Estimating the long-
Finally, it should be noted that the survey-based
term fiscal consequences of immigration in a dynamic
tests summarized above are indirect and incomplete.
model of public finance is very difficult, of course,
Like the studies that examine concerns about labor
and requires taking into account fiscal contributions
market competition, existing tests of the fiscal burden
made by successive generations of immigrant and
model rely upon data on responses to NES and ISSP
native families. For countries with aging work forces,
survey questions that ask individuals about their
in particular, the long-term public finance gains from
attitudes toward immigration in general, not about
importing young workers likely outweigh the costs
their attitudes toward highly skilled or low-skilled
(Krugman and Obstfeld 2000). Perhaps short-term
immigrants specifically. They rest on the problematic
fiscal effects dominate longer-term effects in shaping
assumption that all respondents actually have low-
attitudes among native citizens, but the evidence is
skilled immigrants in mind when answering these
complicated enough to suggest caution in claiming
survey questions about immigration. And employing
that fears about the tax effects of immigration are a
this assumption still only allows a partial test of the
strong motivation for anti-immigrant sentiments.
theory: it does not allow one to test whether the
Quite separately, the finding that tax considerations
incomes of natives affects their attitudes toward highly
among natives play a strong role, and actually trump
skilled immigrants in the expected way.
concerns about cuts in per capita welfare benefits,
In sum, the existing research examining whether
seems especially surprising in the United States. Ev-
attitudes toward immigrants are strongly shaped by
idence on recent fiscal experiences of U.S. states seems
concerns about labor market competition and fears
inconsistent with this claim. Although states gained
about the fiscal burden on public services does not
broad discretion over welfare policies following the
provide convincing conclusions. Most importantly, as
welfare reform of 1996, they have not systematically
a result of data constraints, these studies have not been
raised taxes in recent years even though immigration
able to provide direct tests of the relevant theoretical
has increased. In fact, as shown in the left panel of Fig-
ure 1, looking across the states, there exists, if anything,
a negative correlation between changes in state income
9 For both tax rates and the percent foreign-born population, changes
tax rates and levels of immigration. States that expe-
are computed as the level in 2004 minus the level in 1990. Tax
rienced greater increases in their foreign-born popu-
rates are average marginal state tax rates on wages, taken from the
lations between 1990 and 2004 had smaller increases
NBER state tax database (Feenberg and Coutts 1993) available at
(or larger cuts) in the average marginal tax rates than
http://www.nber.org/ taxsim/state-marginal/. Income taxes are dollar-
states with smaller immigrant inflows over the same
weighted average marginal income tax rates as calculated by the
NBER TAXSIM model from micro data for a sample of U.S. tax-
payers. The results are very similar if tax rates on other sources of
income are used (i.e., taxes on interest received, dividends, pensions,
8 The study reports findings in 1996 dollars. The NRC study did re-
or property tax, etc.). Data on the percent foreign-born are taken
port that tax effects vary depending on the skill levels of immigrants:
from the U.S. Census 1990 and the American Community Survey
immigrants with an education beyond high school contribute an av-
2004.
erage of $105,000 to U.S. tax coffers over their lifetime, whereas the
10 Public welfare expenditures are taken from the U.S. Census of
least educated immigrants create a net deficit of $89,000 per person
Governments (see the following section for more details on the
(Smith and Edmonston 1997).
welfare spending measures).
5

---

Attitudes toward Immigration
February 2010
0
1
NV
opulation:
Loess Fit
Reg. Fit
P
AZ
NJ
Foreign-born
TX
GA
CO
cent
NY
FL CA
IL
er
NC
WA
P
VA MD
DE
OR
MA
NM
MEAN
HI
and
UT
MN
DC
NE
ID
CT
RI
TN
AR
SC
KS
OK
IN
MI
Capita,
AK
PA
WI IA
er
KY AL
MO
P
WY
OH
NH
Change in Percent Foreign Born: 2004 to 1990
ND
VT
LA
MS
SD
ME
02468
WV
MT
Spending
1500
1000
500
0
elfare
Change in Public Welfare Expenditure per capita: 2004 to 1990
W
lic
Pub
0
,
1
NV
Rate
Loess Fit
Reg. Fit
ax
T
AZ
Income
NJ
TX
GA
State
CO
CA
NY
IL
FL
NC
WA
ginal
VA
MD
OR
MA
DE
NM
MEAN
HI
Mar
UT
MN
e
DC
g
CT
NE
ID
RI
AR TNSC
OK
MI
vera
KS
IN
A
PA
IA
MO
AK
WI
in
KY
AL
WY
OH
NH
Change in Percent Foreign Born: 2004 to 1990
ND
es
LA
VT
MS
SD
ME
02468
WV
MT
Chang
1.
1990
4
2
0
–2
to
Change in Avg. Marginal State Income Tax Rate: 2004 to 1990
FIGURE
2004
6

---

American Political Science Review
Vol. 104, No. 1
propositions. No study to date has been able to
Version 1: Do you agree or disagree that the US
distinguish between attitudes toward highly skilled
should allow more highly skilled immi-
immigrants and attitudes toward low-skilled immi-
grants from other countries to come and
grants, even though this distinction is a critical fea-
live here? (emphasis added)
ture of the theoretical story. Below we describe a
Version 2: Do you agree or disagree that the US
survey experiment aimed at addressing this short-
should allow more low-skilled immi-
coming and providing an explicit test of arguments
grants from other countries to come and
about how economic concerns shape attitudes toward
live here? (emphasis added)
immigration.
Answer options (both versions):
THE SURVEY EXPERIMENT
Strongly
Somewhat
Neither agree
disagree
disagree
nor disagree
Design
1
2
3
Somewhat
Strongly
Our experiment was embedded in the Cognitive Styles
agree
agree
Survey (CSS), a survey instrument designed to study
4
5
opinions regarding trade and immigration. The CSS
was administered by the research firm Knowledge
The two question versions differed only in that they
Networks (KN) and fielded between December
described the immigrants’ skill level as either highly
2007 and January 2008 to some 2,285 panelists who
skilled or low-skilled.15 Accordingly, for half the re-
were randomly drawn from the KN panel. Of these,
spondents, referred to as the treatment group, we
1,601 responded to the invitation, yielding a final
measured preferences over highly skilled immigration,
stage completion rate of 70.1 %.11 The KN panel is
whereas for the other half, referred to as the control
a probability-based panel where all members have
group, we measured preferences over low-skilled immi-
a known probability of selection. It covers both the
gration. Randomization ensured that the two groups of
online and offline U.S. populations aged 18 years
respondents were (in expectation) identical in all other
and older. The sampling procedure for the CSS thus
observed and unobserved characteristics that may con-
constitutes a two-stage probability design.12 The
found a comparison across groups.16
recruitment rate for this study, reported by KN using
The general distribution of preferences over both
the AAPOR Response Rate 3 (RR3) guidelines, was
highly skilled and low-skilled immigrants is displayed
24.6%.13 The final respondent data were adjusted for
in Figure 2. For both types of immigration the barplots
the common sources of survey error (nonresponse,
show the fraction of respondents answering each of
coverage error, etc.) using poststratification weights.14
the five answer categories; the superimposed whiskers
The rate of item nonresponse was very low, below 1%
decode the upper .95 confidence interval derived from
for the questions we use in the analysis below.
the design-based variance estimator. Two features
For the core experiment, we randomly allocated re-
stand out in this graph. First, in line with previous
spondents to two groups of equal size and presented
studies, our survey once again confirms the profound
each group with one of two versions of the survey ques-
divide among the American public in opinions on
tion about immigration:
immigration. Pooling over both types of immigration,
about 50% of the respondents oppose an increase in
immigration, whereas about 25% favor it. Second and
more importantly, our findings for the first time docu-
11 All fielded sample cases had one e-mail reminder sent three days
ment the fact that preferences over immigration vary
after the initial email invitation. No monetary incentive was used in
rather dramatically depending on the immigrants’ skill
the CSS study. Of the invited respondents, 4.5 % did break off before
the interview was completed.
levels. Although more than 60% of the respondents
12 Panel members are randomly selected using random digit dialing
(in the control group) state that they strongly disagree
(RDD) sampling techniques on the sample frame consisting of the
or somewhat disagree with an increase in low-skilled
entire U.S. residential telephone population (both listed and un-
immigration, only 40% of the respondents (in the
listed phone numbers). Households are provided with access to the
treatment group) are opposed to an increase in highly
Internet and hardware if needed. In contrast to opt-in Web panels,
unselected volunteers are not allowed to join the KN panel. A de-
skilled immigration.17 Because of the randomization,
tailed report about the KN recruitment methodology and the survey
administration is available from the authors upon request.
13 Notice that an online panel such as KN is composed of people
15 Notice that we stratified the random assignment by four education
recruited at different times and committed to answer several surveys
levels (described below) so that an equal number of respondents
for a period of time. KN panelists must also complete profiling sur-
within each education level received the two different versions of
veys in order to become members of the panel. These differences
the question.
make directly comparing response rates between one-time surveys
16 We conducted extensive balance checks by comparing the distri-
(such as simple RDD telephone or mail sample) and panel surveys
butions of all our covariates in both groups. All tests confirmed that
difficult and perhaps not illuminating. See Callegaro and DiSogra
(as expected given the large sample size) randomization balanced
(2008) for an extended description of how to compute response
the distributions evenly. Results are available upon request.
metrics for online panels.
17 In the preimplementation pilot testing, we created a third,
14 Poststratification weights are raked to adjust to the demographic
“vanilla” version of the question that referred simply to “im-
and geographic distributions from the March Supplement of the 2007
migrants”, without mentioning skill levels, and we randomly as-
Current Population Survey.
signed respondents into a third group who answered this question.
7

---

Attitudes toward Immigration
February 2010
EMPIRICAL TEST I: THE LABOR MARKET
FIGURE 2.
Support for Highly Skilled and
COMPETITION MODEL
Low-skilled Immigration
Skill Levels of Natives
Strongly agree
Agree
Neither agree nor disagree
If concerns about labor market competition are im-
Disagree
Strongly disagree
portant in shaping attitudes toward immigration, we
Upper 95% confidence bound
expect, in line with the FP model of attitude formation,
that natives should oppose immigrants with similar skill
levels but favor immigrants with different skill levels.
That is, we expect that the skill levels of our survey re-
Allow more highly skilled immigration?
spondents should have a large and positive relationship
with support for low-skilled immigrants and a large and
negative effect on support for highly skilled immigrants.
To conduct an explicit test of this argument, we
follow previous studies and employ educational at-
tainment as our measure of respondent skill levels
(Facchini and Mayda 2009; Hanson, Scheve, and
Slaughter 2007, 2008; Mayda and Rodrik 2005; Scheve
Allow more low-skilled immigration?
and Slaughter 2001). This measure, which we label
EDUCATION, is a categorical indicator of the highest
0.0
0.1
0.2
0.3
0.4
0.5
0.6
level of education attained by the respondent. The cod-
Fraction
ing is: 1 = Not completed high school education, 2 =
High school graduate, 3 = Some college, 4 = Bache-
lor’s degree or higher. Alternatively, we also use a set
we know that this statistically significant difference
of binary indicator variables called HS DROPOUT,
between the two distributions is entirely driven by
HIGH SCHOOL, SOME COLLEGE, and BA DE-
the perceived differences in the skill attributes of the
GREE that are coded one if a respondent belongs
immigrants.18
to the respective category of EDUCATION and zero
In an additional experiment we replicated all our
otherwise. Summary statistics for all variables used in
tests based on within-group variation by using a cross-
the analysis are provided in Appendix B.
over design. For this follow-up test, we contacted a
random subset of the respondents two weeks after they
had completed the main survey. Half of these respon-
Attitudes toward Highly and Low-skilled
dents we randomly selected to receive the alternate
Immigrants and Natives’ Skill Levels
version of the question they had received in the original
survey two weeks prior. This approach allowed us to
Figure 3 plots the distributions of preferences condi-
compare the responses to both questions from the same
tional on respondents’ skill levels. The results suggest
individual while minimizing the danger of “consistency
two key findings. First, regardless of the respondents’
bias.”19 The results from the analysis of this follow-up
skill level, highly skilled immigrants are strongly pre-
experiment, which strongly confirm the results from
ferred over low-skilled immigrants. Second, in stark
the main experiment reported below, are described in
contrast to the predictions based on the theoretical
Appendix A.
model, we find that support for both types of immi-
gration is increasing (at a roughly similar rate) with re-
spondents’ skill level. For example, whereas only 7% of
the least skilled respondents (those who did not finish
high school) favor an increase in low-skilled immigra-
Opposition to immigration among this group was lower than op-
tion, 29% favor an increase in highly skilled immigra-
position to low-skilled immigration (in the pilot control group),
tion. However, we find a similar preference differential
and higher than opposition to highly skilled immigration (in the
among the most highly skilled respondents (those with
pilot treatment group). Because the results fell in the middle when
no skill levels were specified, we focused on just the two con-
at least a bachelor’s degree): only 31% prefer an in-
trasting versions of the question when we implemented the survey
crease in low-skilled immigration but more than 50%
experiment.
prefer an increase in highly skilled immigration.
18 As a robustness check we also replicated both the labor market
Taken together, these results are at odds with the
competition and the fiscal burden tests, excluding respondents who
claim that concerns about labor market competition
chose the neutral, middle category. The results, which are available
upon request, are virtually identical to the ones presented below
are a driving force in shaping attitude toward immigra-
where the middle category is included. Omitting the middle cate-
tion. Instead, the results are consistent with previous
gory leads, if anything, to an even stronger disconfirmation of the
findings indicating that people with levels of higher
conventional wisdom.
education are more likely to favor immigration (for a
19 It is well known that if asked questions about similar issues all at
variety of other economic and noneconomic reasons)
once, respondents tend to make their answers consistent even when
they would respond to the questions in substantially different ways
regardless of immigrants’ skill attributes (Hainmueller
were they asked separately.
and Hiscox 2007).
8

---

American Political Science Review
Vol. 104, No. 1
FIGURE 3.
Support for Highly Skilled and Low-skilled Immigration by Respondents’ Skill Level
Allow more low-skilled immigration?
Allow more highly skilled immigration?
Strongly agree
Strongly agree
Agree
Agree
Neither agree nor disagree
Neither agree nor disagree
Disagree
Disagree
Strongly disagree
Strongly disagree
BA, MA, PHD
BA, MA, PHD
Upper 95% confidence bound
Upper 95% confidence bound
SOME COLLEGE
SOME COLLEGE
HIGH SCHOOL
HIGH SCHOOL
OPOUT
OPOUT
HS DR
HS DR
0.0
0.2
0.4
0.6
0.8
0.0
0.2
0.4
0.6
0.8
Fraction
Fraction
Formal Tests of the Labor Market
who did not finish high school), we expect strong sup-
Competition Model
port for highly skilled over low-skilled immigration, so
that γ + δ · 1 > 0. For the most highly skilled respon-
We created a binary indicator variable, HSKFRAME,
dents with EDUCATION = 4 (those with a bache-
coded one if the respondent i received the question
lor’s degree or higher), we expect the exact opposite,
about highly skilled immigrants and zero if he or she
γ + δ · 4 < 0. In other words, low-skilled immigration
received the question about low-skilled immigrants.
is preferred over highly skilled immigration. Taken to-
The observed support for immigration is measured by
gether this implies that δ is negative, fairly large in
the categorical variable PROIMIG, which takes on the
magnitude (|γ| > δ/4), and statistically significant.
integer value associated with one of the five answer
In our second test specification we relax the assump-
categories j = (1, 2, . . . , 5) from “strongly disagree”
tion of linearity in the premium for highly skilled im-
to “strongly agree.” We model PROIMIG using an
migration and estimate
ordered probit model with poststratification weights.
For all uncertainty estimates we employ the robust lin-
µ
earized variance estimator that yields the valid design
i = α + γ HSKFRAMEi +
δk (HSKFRAMEi
based inferences.20
k∈{1,2,4}
To explicitly test the labor market competition ar-
·
gument, we estimate the systematic component of the
1 {EDUCATIONi = k}) +
ordered probit model with the specification.
k∈{1,2,4}
µ
θk 1 {EDUCATIONi = k} + Ziψ.
i = α + γ HSKFRAMEi + δ (HSKFRAMEi
· EDUCATIONi) + θ EDUCATIONi + Ziψ,
This specification allows a different premium con-
ditional on each of the four skill categories HS
where the parameter γ is the lower-order term on the
DROPOUT, HIGH SCHOOL, SOME COLLEGE,
treatment indicator that identifies the premium that
and BA DEGREE. Notice that we use SOME COL-
natives attach to highly skilled immigrants relative to
LEGE (respondents who have some college education
low-skilled immigrants. The parameter δ captures how
but did not graduate) as our reference category, so
the premium for highly skilled immigration varies con-
that γ identifies the premium estimated for this skill
ditional on the skill level of the respondent.
level. Accordingly, γ + δ1, γ + δ2, and γ + δ4 identify
The key predictions based on the standard model of
the premia estimated for those respondents in the cat-
labor market competition are as follows: For the least
egories HS DROPOUT, HIGH SCHOOL, and BA
skilled respondents with EDUCATIONi = 1 (those
DEGREE. The key prediction is that γ + δ1 is positive
and significant whereas γ + δ4 should be negative and
20 Let S(β) = ∂lnL be the score function where ˆβ is estimated by
significant.
∂β
solving ˆS(β) = 0. Following a first-order Taylor series expansion, the
We also enter a basic set of sociodemographic
linearized variance estimator is given by ˆ
V( ˆβ) = D V{ ˆS(β)}|
D ,
β= ˆβ
covariates Z including the respondent’s age (in seven
where D = { ∂ˆS(β) }−1.
age brackets), gender (female = 1, male = 0), and race
∂β
9

---

Attitudes toward Immigration
February 2010
(four dummies for white, Hispanic, black, and other)
skilled immigration among differently skilled respon-
in all specifications. The covariates are simply included
dents is not significant.
here to increase the comparability of some of the
Taken together, these results reveal several striking
coefficients with previous studies (Facchini and Mayda
features regarding the dynamic of respondents’ skill
2009; Hanson, Scheve, and Slaughter 2007; Mayda
levels and immigration preferences. To give some sense
2006; Scheve and Slaughter 2001). Notice that because
of the substantive magnitudes involved, we simulate
the randomization orthogonalizes HSKFRAME with
the predicted probability of supporting an increase in
respect to Z, the exact covariate choice does not affect
immigration (answers “somewhat agree” and “strongly
the results of the main coefficients of interest.21
agree” that the U.S. should allow more immigration)
for the median respondent (a white woman aged 45) for
all four skill levels and both immigration types based
Results for Tests of the Labor Market
on the least restrictive model (model five in Table 1).
Competition Model
Figure 4 shows the results and summarizes our key
Results for the tests are shown in Table 1. In the first
findings for the tests of the labor market competition
two columns we separately regress attitudes toward
argument.
highly skilled and low-skilled immigration on respon-
First, in contrast to the predictions from the labor
dents’ skill level (measured by educational attainment)
market competition model, support for both low- and
and the set of covariates. Following the labor market
highly skilled immigration is steeply increasing in re-
hypothesis we would expect that the support for low-
spondents’ skill levels. This increase in the probabil-
skilled (highly skilled) immigration should increase
ity of supporting immigration is very large in substan-
(decrease) in respondents’ skill level. In contrast, we
tive terms. For example, for highly skilled immigration
find that the correlation between respondents’ skill
it ranges from .23 [.18; .26] among respondents who
level and support for immigration is positive and sig-
did not finish high school to .40 [.35; .45] among col-
nificant for both types of immigration (columns one
lege graduates (the numbers in square brackets give
and two). In fact, we cannot reject the null hypothesis
the .95 percent confidence interval). Furthermore, the
that the effect of respondents’ skill on support for in-
increase is not linear, but instead is particularly pro-
creased immigration is identical for highly skilled and
nounced for the gap between respondents who have a
low-skilled immigrants (p-value .21).
college education and those who do not. This plateau
The next three models implement our main exper-
effect is in line with findings in some previous stud-
imental tests. To identify the premium attached to
ies (Chandler and Tsai 2001; Hainmueller and Hiscox
highly skilled relative to low-skilled immigrants, we
2007) showing that exposure to university education
use PROIMIG as our dependent variable and regress
seems to be the critical contributor to the generally
it on the indicator HSKFRAME that denotes whether
positive relationship between education and support
a respondent received the frame about highly skilled
for immigration.
immigrants rather than the question about low-skilled
Second, regardless of the respondents’ skill level,
immigrants. Results are shown in column three. The
highly skilled immigrants are much preferred over low-
high-skill frame indicator enters positive and highly sig-
skilled immigrants. This finding is at odds with the ex-
nificant, indicating that on average highly skilled immi-
pectation from the standard model of labor market
grants are strongly preferred to low-skilled immigrants.
competition that highly skilled natives should oppose
Column four includes the interaction of HSKFRAME
inflows of highly skilled immigrants and support in-
with respondents’ skill level, measured by EDUCA-
flows of low-skilled immigrants. On the average (i.e.,
TION. The interaction term enters with the expected
across the four skill levels), the predicted probability
negative sign, but it is statistically insignificant and the
of supporting highly skilled immigration is about 0.15
point estimate is very small in substantive terms. This
higher than the probability of supporting low-skilled
result suggests that, in contrast to expectations based
immigration and this difference is highly statistically
on the labor market competition model, the premium
significant.
attached to highly skilled immigration does not vary
Third, there seems to be no systematic variation
significantly with respondents’ skill level. In column
in the premium attached to highly skilled immigrants
five we also drop the linearity assumption regarding the
across respondents’ skill level. As clearly indicated by
effect of respondents’ skill level and replace EDUCA-
the dashed lines that connect the predicted probabil-
TION with our set of dummy variables that indicate the
ities for each type of immigration, the step function
highest level of educational attainment (SOME COL-
that describes increased support for immigration with
LEGE is the reference category) plus all interactions
rising skill levels among respondents is quite similar
with the high-skill question frame. We find that not one
for the two types of immigration. The relative differ-
of the interaction terms is significantly different from
ences in predicted probabilities of supporting highly
zero. A Wald test against the null that all interaction
skilled versus low-skilled immigration are .17 [.13; .20]
terms are jointly zero yields a p-value of .61, indicating
for respondents who did not complete high school, .12
that the variation in the premium attached to highly
[.10; .14] for high school graduates, .15 [.12; .18] for
those with some college education, and .17 [.13; .21] for
21
college graduates. The differences are not significantly
All results are substantively identical if additional (pretreatment)
covariates (suchas martial status or geographic indicators) or no
different and do not have opposite signs, as predicted
covariates at all are used. Results available upon request.
by the labor market competition model. The two dotted
10

---