Discrimination in marketized welfare services: a field experiment on Swedish schools

Providers' tendency to cream skim clients according to certain sociodemographic traits is widely believed to increase under marketization, and therefore also discrimination. However, due to a lack of experimental research, little is known about the presence of discrimination in marketized welfare services and of the potential drivers of such biased treatment. The lack of research is particularly evident in regard to socioeconomic status (SES) discrimination and publicly financed for-profit providers. Moreover, competition, an important aspect of marketization, has not been investigated. Focusing on the interesting case of the Swedish school sector, we aim to improve knowledge on these matters. In a field experiment, 3,430 elementary school principals were randomly contacted though e-mail by parents with Arabic- or Swedish-sounding names and in low- or high-socioeconomic professions. The fictional parents were interested in placing their children at the school. The Swedish school sector resembles marketized public services in several Western countries. The results show clear signs of ethnic as well as SES discrimination, particularly in regard to more qualitative aspects of the replies. However, we find no significant differences in discrimination between public and private/for-profit schools and depending on the degree of competition in the school market.

Automated summary

The tendency of providers to selectively serve clients based on sociodemographic traits, known as cream skimming, is believed to increase discrimination under marketization. However, little is known about discrimination in marketized welfare services and the factors driving such biased treatment, due to a lack of experimental research. This study focuses on the Swedish school sector and finds clear evidence of ethnic and socioeconomic discrimination, but no significant differences based on school type or market competition.