Is ethnic discrimination due to distaste or statistics?

Abstract

provide direct evidence for hiring discrimination based on ethnicity: job applications with native names receive between 14 and 50 percent more positive callbacks than applications with non-native names in the US, Germany and Belgium. However, identifying discrimination is one thing; tackling it is another.To combat labour market discrimination effectively, we need to understand its underlying mechanisms. As reviewed by Guryan & Charles (2013), the leading explanations for labour market discrimination still go back to the theoretical models of taste-based discrimination, as introduced by Becker (1957), and statistical discrimination, as introduced by Phelps (1972) and Arrow (1973). In the model of taste-based discrimination, members of the majority experience a disutility from interacting with minority workers and are willing to pay a financial price to avoid such interactions. Becker (1957) describes three sources of discriminatory tastes: employers, co-workers and customers. Statistical discrimination occurs when employers examine statistics about a group's average performance to predict a particular applicant's productivity as a time-efficient and profit-maximising response to imperfect information about the actual productivity of the individual job candidate.As reviewed by Guryan & Charles (2013), most papers attempting to answer the question whether taste-based or statistical discrimination is a more appropriate explanation for unequal treatment in the labour market have conducted indirect assessments: they have measured whether particular patterns in economic data square predictions of the model being tested. The problem with this literature is that testing between the two models is only convincing to the degree that a particular pattern is explicable exclusively by one model, a challenge that is, as shown by Guryan & Charles (2013), rarely met. Recent work, however, has attempted to test more essential arguments of the taste-based model or the statistical discrimination model in explaining labour market discrimination (see Bertrand

Immigrant volunteering: A way out of labour market discrimination?

“…On the one hand, following Becker's (1957) model of taste-based discrimination, employers as well as their employees and customers may have less distaste for interacting with well-integrated migrants compared with their less integrated counterparts (Adida et al, 2014;Handy and Greenspan, 2009). On the other hand, following Arrow's (1973) model of statistical discrimination, the volunteering signal may weaken stereotypes with respect to foreign job candidates' productivity, values and motivation (Baert and De Pauw, 2014;Derous et al, 2009). …”

Measuring Hiring Discrimination A History of Field Experiments in Discrimination Research

“…(emphasis in the original, Midtbøen & Rogstad, 2012, p. 203), and closely linked to this last question, "Is ethnic discrimination due to distaste or statistics?" (Baert & de Pauw, 2014). Different academic disciplines, in particular economists, legal scholars and sociologists, have addressed the issue of discrimination using a variety of methods.…”

Are Survey Experiments as Valid as Field Experiments in Management Research? An Empirical Comparison Using the Case of Ethnic Employment Discrimination

“…Recent empirical studies suggest that researchers believe that survey experiments constitute a capable design for studying organizational topics as discrimination and gender differences in management (Baekgaard and George, ; Baert and De Pauw, ; Brescoll, ). However, using survey experiment designs for topics such as discrimination implies that respondents' behavioural intentions are the same as their real behaviour.…”