A Sometimes Hidden Economic Dimension to Individual Immigration Preferences: Cross-National Evidence in Support of the Labor Competition Hypothesis

• DOI: 10.1177/1065912918819863
• Published date: 01 December 2019
• Author: Megan Roosevelt,David H. Bearce
• Date: 01 December 2019
• Subject Matter: Articles

https://doi.org/10.1177/1065912918819863

Political Research Quarterly

2019, Vol. 72(4) 894 –909

© 2018 University of Utah

Article reuse guidelines:

sagepub.com/journals-permissions

DOI: 10.1177/1065912918819863

journals.sagepub.com/home/prq

Article

Across countries, survey data consistently show that a

large majority of citizens oppose greater immigration

(Facchini and Mayda 2009; Rosenblum and Cornelius

2012). While there is no puzzle about what most indi-

viduals prefer, scholars have debated the determinants of

these anti-immigration attitudes. One possible explana-

tion is economic self-interest. When migrants enter the

national economy, they may compete for jobs especially

against lower-skilled natives, leading the latter to oppose

the former on this basis. Another explanation is cultural

bias. Even if they pose no particular economic threat,

many citizens may oppose immigrants simply because

they are different in terms of language, religion, and/or

skin color. From the outset, it should be noted that eco-

nomic self-interest and cultural bias are not competing

explanations; the truth of one does not demonstrate that

the other is false.

An early wave of research (e.g., Mayda 2006;

O’Rourke and Sinnott 2006; Scheve and Slaughter 2001)

offered some support for an economic explanation from

survey evidence showing that individuals with more edu-

cation were less opposed to immigration based on the

understanding that most migrants are lower-skilled and

would, thus, compete more directly with less educated

natives. But Hainmueller and Hiscox (2007) challenged

this conclusion, arguing that the economic pressure

related to immigration is often very light and that educa-

tion also captures pro-immigrant socialization, which

means that the positive correlation between education

and pro-immigration attitudes is just as consistent with

the cultural explanation as it is with the economic expla-

nation. As Hainmueller and Hopkins (2014, 241,

819863PRQXXX10.1177/1065912918819863Political Research QuarterlyBearce and Roosevelt

research-article2018

1University of Colorado Boulder, USA

Corresponding Author:

David H. Bearce, University of Colorado Boulder, 333 UCB, Boulder,

CO 80309, USA.

Email: david.bearce@colorado.edu

A Sometimes Hidden Economic

Dimension to Individual Immigration

Preferences: Cross-National Evidence

in Support of the Labor Competition

Hypothesis

David H. Bearce1 and Megan Roosevelt1

Abstract

This paper seeks to restore labor competition as an explanation for anti-immigration attitudes, recognizing that

education may proxy both individual-level skill and cultural socialization. We thus need new tests to distinguish

the effect of education based on skill from that due to socialization. If the education effect is consistent with these

relationships, then we can have greater confidence that it is capturing the former and not simply the latter. This paper

thus develops and conducts a new test, using data from the International Social Survey Program’s National Identity

Survey fielded in 2013 across thirty-two countries. From a factoral framework, our test identifies three national-

level factors that should influence how much labor market pressure lower skilled citizens feel from immigration: the

quantity of immigrants, the direction of capital/investment flows, and the amount of trade protection. These national-

level factors are interacted with individual-level education, showing that the attitudinal differences based on education

increase with more immigrants but decrease with greater investment inflows and increased trade protection. These

results demonstrate why this economic dimension may sometimes be hidden: in national contexts where there are few

immigrants, capital follows labor, and/or there is trade protection, labor competition as a driver of anti-immigration

preferences should lessen.

Keywords

immigration attitudes, labor skill, immigrants, capital flows, trade openness

Bearce and Roosevelt 895

emphasis added) later concluded, as “an explanation of

mass attitudes toward immigration, the labor market

competition hypothesis has repeatedly failed to find

empirical support, making it something of a zombie

theory.”

But this debate is not dead. Indeed, it is hard to look at

the populist political wave sweeping across the advanced

industrial democracies with many less educated citizens

expressing strong opposition to immigration without see-

ing a potential link between these attitudes and their labor

market vulnerability. This possibility does not mean that

the former lacks a cultural, even racist, component. But

identifying whether there is an economic component to

these attitudes has some important policy implications. If

anti-immigration attitudes were only based on cultural

bias, then programs to address the economic vulnerability

of lower-skilled citizens need not be part of the policy

package. Indeed, one might even argue that it would

make little sense to appease bigots with greater economic

support if their anti-immigration attitudes have no basis

in terms of labor market pressure. But if such attitudes do

have an economic basis, then worker training programs

and unemployment compensation, which could be justi-

fied for a number of other reasons (Hays, Ehrlich, and

Peinhardt 2005; Scheve and Slaughter 2007), acquire an

additional rationale in that they might help some citizens

be more tolerant of foreigners and less hostile to

immigrants.

This debate, however, cannot be settled simply by

developing a measure of labor market skill that captures

only skill and not cultural socialization. Since it takes

time to develop, skill—even when based on narrow occu-

pational characteristics—is positively correlated with

educational socialization. Instead, we need to look for

new tests that distinguish the effect of education based on

labor market skill from that due to socialization. Stated

differently, we need to identify potential relationships

between immigration attitudes and education that fit with

labor market skill but not with socialized attitudes. If the

education effect is consistent with these relationships,

then we can have greater confidence that it is capturing

the former and not simply the latter, helping to revive

labor competition as an explanation for anti-immigration

attitudes.

We first develop and then conduct a new test in this

paper, using data from the International Social Survey

Program’s (ISSP) 2013 National Identity Survey fielded

across thirty-two countries, which represents the largest

study (to date) conducted on immigration attitudes. Using

a factoral framework, our test identifies three national

factors that influence how much labor market pressure

lower skilled citizens should feel from immigration: (1)

the quantity of immigrants, (2) the direction of capital/

investment flows, and (3) the amount of trade protection.

Thus, the labor market pressure associated with immigra-

tion should depend not only on foreign labor inflows but

also on the other two primary features of economic glo-

balization, namely, international capital mobility and

international trade.

In this cross-national survey sample with more than

twenty-five thousand respondents, we interact these three

national-level factors with individual-level education to

show that the attitudinal differences based on education

not only increase with more immigrants but also decrease

with greater investment inflows and increased trade pro-

tection. These conditional results offer new evidence sup-

porting the labor competition hypothesis given that they

are both predicted by its factoral foundations and hard to

derive from any cultural theory. Thus, we demonstrate

why this economic dimension underlying immigration

policy attitudes may sometimes be hidden. In national

contexts where there are few immigrants, capital follows

labor, and/or there is trade protection, labor competition

as a driver of anti-immigration preferences should lessen.

But this economic dimension should also be expected to

grow in different national contexts (e.g., with more immi-

grants, investment outflows, and/or reduced tariffs).

These results also demonstrate why we must be care-

ful about drawing any strong conclusions about the labor

competition hypothesis based on evidence from a single

country: there is significant variation across countries in

the labor market effect of immigration and the corre-

sponding power of labor market skill to predict immigra-

tion policy attitudes. Take, for example, the United States:

the third wave of the National Identity Survey shows the

education result to be weak in this country subsample,

unlike for the full sample. And the argument presented

here explains why this should be the case: strong capital

inflows in the U.S. economy offset the wage effect of

immigration.

A New Test for the Labor

Competition Hypothesis

Scholars have objected to labor competition as an expla-

nation for stronger anti-immigration attitudes among the

less skilled, arguing that the economic impact of immi-

gration is weak; consequently, citizens should not feel

any real threat to their job and income from foreign work-

ers regardless of their skill level. As Hainmueller and

Hiscox (2007, 437) concluded, “a growing body of

empirical research shows that the actual effects of immi-

gration flows on income, employment, and unemploy-

ment are quite small.” On this basis, any scholarly belief

that immigration preferences should be strongly based on

economic considerations “might not be grounded in eco-

nomic reality” (Hainmueller and Hopkins 2014, 228). It

should be noted that this argument against the labor