Prejudice in naturalization decisions: Theory and evidence

Abstract

Immigrant groups that are marginalized in their host countries are disproportionately more likely to have their citizenship applications rejected. It is not readily obvious whether this disparity is due to prejudice on the part of decisionmakers or due to applicant di¤erences in meeting naturalization standards. To address this question, I develop a simple model of a council deciding whether to grant applicants citizenship. The model implies an empirical test for relative prejudice using average applicant group rejection rates. Using Switzerland as a case study, I apply the test to newly collected data from six large municipalities. In …ve municipalities, the test cannot reject the hypothesis of no relative prejudice with respect to country of origin. The rejection pattern of the sixth municipality is consistent with prejudice.The model illustrates that the underlying mechanism in the decisionmaking process has bearing on the inference of prejudice from empirical data.

The Price of Prejudice

“…where F is a CDF and X is a vector of variables consisting of the price of discrimination (see equation (4)) and gender and ethnicity as additional controls. We use maximum likelihood and assume a probit model to estimate equation (5). 18 The demand for discrimination then shows how likely decision makers on average are to discriminate given that a particular price is assigned to them.…”

“…3 Hence, we are able to put a price tag on discrimination choices or, borrowing Gary Becker's (1957) expression 4 , to estimate how discrimination responds to the "price of prejudice," rather than just observing that discrimination occurs when it is costless. 5 In our main treatment (called Info), we randomly assign a price of discrimination by giving decision makers the choice between two coworkers of different ethnicity and of known productivity. The decision maker and the chosen coworker then form a team and are paid according to team output.…”

“…4 "Price and Prejudice" is the title of part 2 in Becker (1976), which is a revised version of his Ph.D. thesis, published in 1957. 5 The reason why most tests involve a zero cost of discrimination to the discriminator is probably that they take the standard definition of discrimination as a point of departure. Altonji and Blank (1999, 3168) define discrimination in the labor market as "a situation in which persons who provide labor market services and who are equally productive in a physical or material sense are treated unequally in a way that is related to an observable characteristic such as race, ethnicity, or gender."…”