Is the Immigration Consensus Purely Race Neutral?

• In Event: Experimental Studies of Racial and Ethnic Prejudice

Abstract

One of the most important and consistent findings to emerge from the academic study of immigration politics over the past decade is the seemingly uniform preference for high-skilled immigrants among citizens in the United States (e.g. Hainmueller and Hopkins 2015) and many other industrialized nations (e.g. Iyengar et al. 2013). Employing sophisticated experimental designs, these works demonstrate that citizens’ judgments about whether or not to admit an immigrant are significantly impacted by an immigrant’s skill level independent of her nationality, skin tone, or religious practices---and irrespective of citizens’ own economic positions, ethnic prejudice, or partisanship. The principle conclusion of this recent work is that citizens’ attitudes toward immigration generally reflect sociotropic considerations as opposed to self-interest: high-skilled immigrants are preferred because people are making economic determinations about the net benefits of their entry vis-à-vis low-skill immigrants.
In this article, we raise important normative, theoretical, and empirical concerns about how to interpret the findings from this recent body of research. At a normative level, we attempt to come to terms with the fact that Americans seem to prefer a type of immigrant that is rare given that most of the foreign-born population in the U.S. is low-skilled (Camarota 2012).
With respect to theory, we focus on how we should interpret what we term the “skill premium” and question race-neutral interpretations. Prevailing sociotropic interpretations of the skill premium presume that it derives from a determination among citizens that high-skill immigrants offer more benefits and impose fewer costs vis-à-vis low-skill immigrants. Here, we offer an alternative interpretation, which is that the skill premium partly reflects a prejudice-based aversion to the modal immigrant most prevalent within one’s country.
In terms of the empirical analysis of public opinion on immigration, this alternative account subsumes the contention that citizens make cultural judgments about immigrants based on information about their skill level, and that these cultural judgments vary based on the race of the immigrant. Sophisticated experimental designs, such as conjoint analysis, attempt to address this concern by controlling for cultural attributes. However, as Dafoe et al. (2015) have shown, omitted variable bias can still creep into these carefully controlled designs, creating problematic confounds. On top of this, none of the leading studies demonstrating the skill premium provide compelling, direct tests of the presumed sociotropic mechanism underlying the observed premium. In sum, our contention is that cultural and economic considerations are deeply intertwined and likely not independent.
The purpose of this article is to subject the ostensibly race-neutral interpretation of the skill premium to a series of “critical tests” designed to potentially offer confirmatory evidence for a purely economic mechanism. Where such tests completely (or partially) fail, we conduct further tests capable of offering supportive evidence for our thesis that prejudice partly underlies Americans’ preference for high-skilled immigrants. Drawing upon original experimental data collected via Survey Sampling International and Amazon Mechanical Turk, we find numerous empirical patterns that cast doubt on a race-neutral account of the skill premium: (1) Respondents evince a stronger preference for a highly-skilled immigrant when evaluating a Latino/a applicant but not a white European applicant; (2) It is only among the most prejudiced that we observe racial differences in the conferring of a skill premium, as those low in prejudice equally prefer high-skill white and Latino/a immigrants to each of their respective low-skill counterparts; (3) Although the skill premium does decrease in response to manipulated information about the sociotropic value of low-skilled immigrants, these effects are concentrated among low-prejudice individuals; (4) Although we find that pre-existing attitudes about the value of low- vs. high-skilled workers moderates the skill premium, even those who claim that low-skilled workers are the most valuable still prefer high-skilled immigrants, suggesting that skill is potentially tapping cultural traits; (5) We find that manipulating skill---despite holding cultural attributes constant---affects people’s perceptions of the cultural characteristics of immigrants, such as perceived linguistic and cultural assimilation, suggesting that conjoint designs may not be able to perfectly control for cultural attributes.
Taken as a whole, the results from this article, rather than supporting the notion of an immigration politics driven by an unprejudiced and uniform public desire for a more skillful workforce, yield a picture of mass preferences over immigration best described as economic reasoning with a racial hue.

Authors

• Neil Malhotra, Stanford University

• Benjamin J. Newman, University of California, Riverside