In this post, Aala Abdelgadir and Vasili Fouka discuss their recent APSR article exploring the consequences of the headscarf ban in France.

In response to rising immigration flows and the fear of Islamic radicalization, several Western countries have enacted policies to restrict religious expression and emphasize secularism and Western values. Eight European countries have banned headscarves in a government setting and four prohibit pupils from covering their hair in schools. Most recently, Quebec banned public employees from wearing religious symbols in some public sector jobs. Despite the increasing ubiquity of headscarf bans, there has been little systematic evidence of their impact.

De facto discrimination

In an article recently published in the American Political Science Review, we investigate the effects of a landmark headscarf ban – the 2004 French law prohibiting conspicuous religious symbols in schools – on the educational and socioeconomic outcomes of Muslim women. Our paper is the first attempt to identify the effects of veiling bans on the economic and cultural integration of the population that they target.

Such effects are not obvious ex ante. Restrictions of religious dress are intended to reduce the visibility of religion in the public sphere. This could, in theory, increase assimilation among immigrant minorities, though research suggests reasons to doubt that conclusion.

Veiling bans are de facto discriminatory, as they target specific groups, and surveys suggest they are perceived as such by Muslims. Discrimination has been robustly connected to diminished psychological well-being, which can impair individuals’ behaviors in many domains, including educational performance and broader participation in the economy and society. 

In contrast, the link between discrimination and minorities’ social and psychological integration is not well understood. Discrimination could weaken attachments to the targeted group and increase assimilation, but it could also provoke resentment against the majority and increase minority identification. 

Overcoming the endogeneity problem

Empirical evidence on discrimination’s impact on social and cultural integration remains inconclusive. In part, the study of discrimination and identity is hampered by an endogeneity problem. Namely, discrimination and its perceptions are themselves shaped by minority members’ integration and identity. As a result, prior studies have struggled to link changes in minority integration solely to (actual or perceived) discrimination. 

To overcome the endogeneity problem, we use a difference-in-differences strategy that allows us to isolate the differential effect of the law on the socioeconomic and identity outcomes of Muslim women.

Using data from the French Labor Force Survey, we compare the difference in outcomes between Muslim and non-Muslim women for cohorts in school during the 2004 law versus cohorts who completed school before the ban. Because the educational attainment of Muslim and non-Muslim women in older cohorts had been running in parallel, we would expect this to continue for younger cohorts in a world without a ban. Since the ban only targeted Muslim women, there is no reason why the trajectory of young non-Muslim women would be affected by the implementation of the 2004 law. If the law had any effect, we would instead observe it in changes in the trajectory of younger cohorts of Muslim women. Any change in the relationship of Muslim and non-Muslim women’s educational attainment for cohorts in school in 2004 and later can be traced to the 2004 law.

Educational, social, economic, and health effects

We find that the gap in secondary school attainment between Muslim and non-Muslim women more than doubles for younger cohorts in school during the ban and remains large for subsequent cohorts. After the ban, younger cohorts of Muslim women increasingly dropped out of secondary school or took longer to complete it. These direct effects on educational attainment bled into long-term economic outcomes. The estimated magnitudes indicate that the veiling law widened the employment gap between Muslim and non-Muslim women by more than a third and the labor force participation gap by more than half.

Social and cultural integration was also affected by the law. Younger cohorts of Muslim women were more likely to identify as French. At the same time, Muslim women’s religious identity was also strengthened for younger cohorts affected by the ban. These results indicate that bans raise the salience of multiple identities within the minority group. They also suggest that Muslim women may respond to a law that casts national and religious identity as incompatible by reaffirming their belonging to both France and their ethnic and religious communities.

Our data suggests that the ban generated these observed effects through discrimination at school. Using data from a social attitudes survey (known as Trajectories and Origins), we find that affected cohorts are more likely to report experiencing racism at school, but not in any other setting (e.g. public transport, store, police). Following the literature in public health and psychology, we evaluate whether these increased perceptions of discrimination in school affected educational performance by depressing psychological well-being. We find tentative evidence that the affected cohorts of Muslim women experienced worse health outcomes. Though the Trajectories and Origins survey lacks questions about mental health and well-being, treated cohorts are more likely to report experiencing health problems related to life conditions or difficulties in private life at school age.

How can we be sure that these effects are directly linked to the 2004 law and not resulting from discrimination against Muslims more broadly? First, our evidence shows a sharp drop in secondary education specifically among cohorts of Muslim women who were in school during the ban. Second, if the law spurred broader anti-Muslim sentiment, school-age Muslim men should have also sustained a negative educational shock. We find limited evidence for this. Affected cohorts of Muslim men experienced a small, temporary dip in secondary school completion rates but no systematic effect on their education.

Our findings highlight the potential pitfalls of policies targeting vulnerable minority groups. While the objective of states may be to increase social cohesion by homogenizing behavior and reaffirming common values, the implementation of such policies implies a de facto discriminatory treatment of minorities. Such differential treatment, especially against adolescents, may have pernicious effects on individual outcomes of minority group members. By broadening the social and economic gap between majority and minority groups, it may also undermine, instead of promote, the goal of social cohesion that states wish to achieve.

Aala Abdelgadir is a Ph.D. candidate in Political Science at Stanford University, studying comparative politics and international relations. Her dissertation focuses on the intersection of how identity is communicated and perceived by voters within an electoral context.

Vasiliki Fouka is an Assistant Professor of Political Science at Stanford University. She studies questions of identity, including ingroup-outgroup relations, immigrant assimilation, and the conditions that encourage or discourage prejudice against ethnic and racial minorities.

The authors’ recent APSR article is available free of charge until the end of February 2021.