A study from the University of Warwick provides the first large-scale evidence that wearing the Islamic veil in France is linked to a dramatic 21.6 percentage-point drop in economic participation — and that this gap comes from labour market discrimination, not religious preference.
Summary
New research published as a working paper by the University of Warwick (No. 1618, July 2026) by Antoine Jacquet and Sébastien Montpetit offers the first empirical evidence on the economic costs of wearing the Islamic veil in a secular Western country. The study uses data from the Trajectoires et Origines (TeO) survey, conducted by France’s national statistical institutes in 2008 — the largest sample of its kind, covering 3,003 Muslim women.
The findings show that veiled Muslim women are 21.6 percentage points less likely to be active in the labour market or education than non-veiled Muslim women, even after accounting for education, work experience, fertility, religiosity, and religious environment. This gap is economically significant — equivalent to having 1.2 additional children under age four.
Using a structural model adapted from Carvalho (2013), the researchers separate religious motives from economic motives for veiling. Their analysis shows that lower economic returns to integration — driven by labour market discrimination and legal restrictions — rather than religious preferences explain most of the negative association. Furthermore, individual religious motives account for roughly one-third of the variation in veiling behaviour, more than twice the explanatory power of the religious environment. This challenges narratives that Muslim women are coerced into veiling by their communities.
The study also finds that discreet religious symbols (such as jewellery) are linked to higher economic participation and are worn by younger, educated, moderately religious women — suggesting they serve as a tool for integration in the French context.
The research has significant policy implications. The authors argue that tightening restrictions on veiling risks worsening integration for women who veil out of personal conviction, while policies targeting discrimination would more directly address the constraint highlighted by the data.
Who Is Affected? A Simple Comparison of Muslim Women in France
This table compares three groups of Muslim women: those who wear no religious symbol, those who wear only discreet symbols (like jewellery), and those who always wear a conspicuous symbol (mostly the veil).
| What We Measured | Women With No Symbol | Women With Discreet Symbols Only | Women Who Always Wear a Conspicuous Symbol (Veil) | What This Means |
|---|---|---|---|---|
| Have a job | 54 out of 100 | 39 out of 100 | — | Veiled women are much less likely to be employed |
| Not working, not looking, not studying | 19 out of 100 | 20 out of 100 | 63 out of 100 | Nearly two-thirds of always-veiled women are outside the workforce |
| Have a university degree | 22 out of 100 | 15 out of 100 | — | Veiled women tend to be less educated on average |
| Years of work experience | About 7 years | About 4 years | — | Veiled women have far less work experience |
| Say religion is “very important” in their life | 47 out of 100 | 57 out of 100 | 77 out of 100 | Veiled women are more religious, as expected |
| Grew up in a family where religion was “very important” | 47 out of 100 | 56 out of 100 | 70 out of 100 | Veiled women also come from more religious families |
Source: TeO Survey (INED/INSEE, 2008), 3,003 Muslim women aged 18–60. “—” means the survey did not report a comparable figure for that group. Percentages are rounded to the nearest whole number.
What Really Explains the Veiling Gap? Religious Faith vs. Economic Barriers
This table shows how strongly different factors are linked to veiling and to economic participation. The last column shows how much of the difference in veiling behaviour each factor explains.
| Factor | Does It Make Veiling More Likely? | Does It Make Economic Participation More Likely? | How Much of the Veiling Difference Does It Explain? | Simple Takeaway |
|---|---|---|---|---|
| A woman’s personal religiosity | Yes — strongly | No clear effect | About 35% | Personal faith is the biggest reason women veil |
| Family and community religious pressure | Yes — moderately | No clear effect | About 15% | Community pressure matters, but far less than personal faith |
| Education | No — more education means less veiling | Yes — strongly | About 7% | Educated women are more likely to work and less likely to veil |
| Work experience | No — more experience means less veiling | Yes — strongly | About 11% | Experience helps women work; it does not encourage veiling |
| Feeling French | No — a stronger French identity means less veiling | — | About 5% | Identity plays a small role |
| Country of birth | — | — | About 14% | Where a woman was born matters for veiling patterns |
Source: Authors’ calculations from the TeO Survey (2008). The table shows that the “religious motives” explanation is not supported by the data, while the “economic discrimination” explanation is — especially for work experience.
The French Context: A Unique Laboratory for Studying Veiling
France provides a distinctive setting for examining the economic consequences of veiling. Since 1989, when three girls were expelled from middle school for refusing to remove their headscarves, the Islamic veil has remained a flashpoint in French politics and society. The controversy culminated in a 2004 law banning conspicuous religious symbols in public schools, followed by a 2010 ban on full-face veils in public spaces and various local bans on burkinis.
French law imposes strict neutrality requirements on civil servants, prohibiting the expression of religious beliefs while on duty — including the wearing of conspicuous religious symbols. Since August 2016, private firms can also introduce neutrality requirements in their internal rules under certain conditions. Beyond legal restrictions, research has documented significant discrimination against veiled women in hiring: a correspondence test found that veiled applicants receive call-back rates up to 80% lower than non-veiled applicants for apprenticeship positions.
This context creates a unique natural experiment: Muslim women in France face substantial economic penalties for veiling, while their religious convictions may encourage the practice. Understanding how women navigate this tension — and what it reveals about the relative weight of religious versus economic motives — is central to the study.
The Data: Unprecedented Insight into Muslim Women’s Lives
The study leverages the Trajectoires et Origines (TeO) survey, conducted in 2008 by France’s National Institute for Demographic Studies (INED) and National Institute for Statistics and Economic Studies (INSEE). This comprehensive survey targeted adults aged 18–60 residing in metropolitan France, purposefully oversampling immigrants and minorities.
The TeO dataset includes 3,003 Muslim women with non-missing data on veiling behaviour — the largest sample of its kind in France. Crucially, the survey contains detailed information on religious practices, including the frequency of wearing religious symbols, self-reported importance of religion, attendance at religious ceremonies, dietary restrictions, and religious marriage.
The researchers supplemented this data with a novel source: the Annuaire des mosquées de France (La Boussole, 2004), a comprehensive directory of mosques and Muslim praying rooms in France, which they digitized manually. This allowed them to measure the local religious environment through the capacity of Muslim worship facilities per thousand inhabitants.
For economic outcomes, the study used both cross-sectional data on current activity status and a retrospective panel constructed from respondents’ year-by-year accounts of work status, enabling the researchers to control for time-varying observables and time fixed effects.
The Findings: A Stark Economic Penalty
The descriptive statistics reveal striking differences between veiled and non-veiled Muslim women. Among women who always wear a conspicuous religious symbol (overwhelmingly the veil), 63% are inactive — neither working, looking for work, nor studying — compared to just 19% of non-veiled women. Veiled women are also significantly less educated, have fewer years of work experience, and are more likely to be second-generation immigrants.
These raw differences persist after controlling for a rich set of covariates. In the preferred specification — which includes controls for education, work experience, fertility, partner characteristics, individual religiosity, religious environment, and survey conditions — always wearing a conspicuous symbol is associated with a 21.6 percentage-point reduction in the probability of being economically active. This effect is economically large: equivalent to the impact of having 1.2 additional children under age four.
By contrast, women who always wear discreet religious symbols (such as jewellery) are approximately 7 percentage points more likely to be active. This pattern suggests that discreet symbols may allow women to preserve some religious benefits while imposing little labour-market cost, as they are not subject to the same legal restrictions or discrimination as conspicuous symbols.
Robustness checks confirm these findings. The results hold when excluding students, restricting the sample to second-generation immigrants, using logit or probit specifications, and analysing the retrospective panel data. Placebo tests show no significant association between religious symbol wearing and economic participation for Muslim men or for women of other religious affiliations — confirming the unique place of the Islamic veil in the French context.
Selected Summary Statistics by Veiling Status
| Variable | All Muslim Women | No Symbol | Discreet Symbol | Conspicuous Symbol |
|---|---|---|---|---|
| Age | 34.6 | 35.6 | 26.4 | — |
| Second-Generation Immigrant | 38% | 39% | 56% | — |
| Employed | 46% | 54% | 39% | — |
| Inactive | 28% | 19% | 20% | 63% |
| Higher Education Degree | 21% | 22% | 15% | — |
| Years of Work Experience | 5.9 | 7.1 | 3.6 | — |
| Religion “Very Important” | 55% | 47% | 57% | 77% |
| Religion “Very Important” in Education | 51% | 47% | 56% | 70% |
| Maghrebi Immigrants in Neighbourhood (>40%) | 46% | 44% | 41% | 51% |
| Observations | 3,003 | 2,021 | 318 | 664 |
Source: TeO Survey (2008). Discreet symbols include jewellery; conspicuous symbols include headcoverings and clothing.
Determinants of Joint Employment and Veiling Decision
| Variable | Effect on Being Active & Not Veiling | Effect on Being Active & Veiling Conspicuously | Effect on Being Inactive & Veiling Conspicuously |
|---|---|---|---|
| Individual Religiosity | −19 pp | +12 pp | +13 pp |
| Vertical Religious Transmission | ns | ns | +19 pp |
| Share of Maghrebi Immigrants | ns | ns | ns |
| Mosque Capacity | ns | ns | +9 pp per seat/100 inh. |
| Schooling (per year) | +1.8 pp | ns | −0.5 pp |
| Work Experience (per year) | +2.0 pp | ns | ns |
Source: Authors’ estimates from multinomial logit model. “ns” indicates not statistically significant at conventional levels. pp = percentage points.
Religious vs. Economic Motives: A Structural Decomposition
To disentangle the mechanisms behind the veiling-participation gap, the researchers estimated a discrete-choice model based on Carvalho’s (2013) theory of veiling. This model posits that women choose whether to work and how much to veil to maximize utility, trading off religious returns (commitment to faith and signalling to community) against economic returns (wages and employment opportunities).
The theoretical framework yields testable predictions about how religiosity and human capital should affect utility across different combinations of activity and veiling. The religious motives channel predicts that the religious returns to veiling should be larger for women who participate economically — because the work environment is “riskier” in terms of opportunities for religiously prohibited behaviour. The economic discrimination channel predicts that the economic returns to participation should be smaller for women who veil — because veiling reduces wages and employment prospects.
The results are striking. Tests of the religious motives channel find no support: neither individual religiosity nor religious environment is associated with economic participation conditional on veiling. By contrast, tests of the economic discrimination channel find support, particularly for work experience: women with more work experience are less likely to participate economically if they veil, even holding religiosity constant. In other words, the returns to experience are lower for veiled women.
The researchers illustrate these findings with a counterfactual exercise. For an “average” woman in the sample, veiling reduces utility for active women (because economic motives dominate) but increases utility for inactive women (because they are less affected by economic penalties). In a hypothetical Muslim-majority context without labour market discrimination, active and inactive women would have roughly equivalent incentives to veil — a pattern consistent with evidence from Indonesia (Shofia, 2020) but sharply different from France.
Decomposing Religious Motives: Individual vs. Signalling
A key contribution of the study is its ability to distinguish between private religious motives and signalling motives for veiling. The TeO survey includes multiple measures of religiosity: frequency of attendance at religious ceremonies, self-reported importance of religion, use of religion for self-identification, respect for dietary restrictions, and religious marriage. These are combined into a latent index of individual religiosity using a measurement system approach.
For signalling motives, the researchers use two proxies: vertical transmission (importance of religion in the respondent’s education and whether she has a religious first name) and horizontal transmission (share of Maghrebi immigrants in the neighbourhood and local mosque capacity).
The decomposition analysis reveals that individual religiosity explains roughly one-third (34.7%) of the variation in veiling behaviour, more than twice the share explained by the religious environment (14.6%). This finding challenges the narrative that a “silent majority” of Muslim women are coerced into veiling by their families or communities. Instead, the results suggest that veiling is primarily an individual choice rooted in personal religious conviction.
This finding has important policy implications. If veiling primarily reflects private religious motives rather than coercion, then policies premised on “liberating” women by restricting veiling are poorly aligned with women’s actual motivations — and may produce unintended consequences.
The Discreet Symbol Phenomenon: A Tool for Integration?
An intriguing finding concerns discreet religious symbols, such as jewellery. The TeO survey reveals that these symbols are worn by women who are younger, educated, economically active, and moderately religious — a profile that closely matches the “new veiling” movement described by Carvalho (2013) in Muslim-majority countries.
In the French context, however, discreet symbols serve a different function. Because they are not subject to the legal restrictions or labour market discrimination faced by conspicuous symbols, they allow women to preserve some religious benefits while imposing little economic cost. The researchers develop a model extension showing that when discrimination against conspicuous symbols is disproportionately large (as in France), moderately religious but educated women may optimally choose to work and wear discreet symbols.
This pattern suggests that in a secular context with strong economic penalties for conspicuous religious expression, discreet symbols may function as a tool for economic integration — allowing women to reconcile their religious convictions with the demands of the labour market.
Policy Implications: Rethinking Restrictions on Veiling
The study’s findings have significant implications for debates on French secularism (laïcité) and policies restricting religious expression in public spaces.
First, the results suggest that restrictions on religious symbols are unlikely to improve the economic participation of veiled Muslim women. Since religiosity is primarily associated with economic participation through veiling — more religious women are more likely to veil, but conditional on veiling, religiosity has little direct effect — policies that reduce veiling may simply reduce the welfare of women who veil out of personal conviction without improving their economic outcomes.
Second, the findings highlight the importance of labour market discrimination as a barrier to integration. The economic penalties associated with veiling — reduced call-back rates, legal restrictions in certain professions, and workplace discrimination — appear to be the primary driver of the participation gap. Policies that reduce these penalties, such as stronger enforcement of anti-discrimination law and clearer limits on “neutrality clauses” in private firms, would more directly address the constraint highlighted by the data.
Third, the study questions the narrative that Muslim women are coerced into veiling by their communities. Since individual religious motives explain more of the variation in veiling than the religious environment, policies premised on “liberating” women from communitarian pressures may be based on a misunderstanding of women’s actual motivations.
The authors conclude that shifting the policy focus away from bans and toward removing institutional barriers to education and employment for Muslim women — including those who wear conspicuous symbols — would better serve the goal of integration.
Conclusion: Economic Barriers, Not Religious Preferences
The study’s central conclusion is that the lower economic participation of veiled Muslim women in France is more consistent with economic penalties associated with veiling than with religious preferences discouraging work. Veiling is costly in the French labour market — not because it reflects a rejection of integration, but because it triggers discrimination, legal restrictions, and reduced economic opportunities.
At the same time, the research shows that individual religious motives, rather than social signalling, are the primary driver of veiling behaviour. Muslim women who veil do so primarily for personal reasons linked to their own beliefs, not because they are forced by their families or communities.
These findings challenge both the narrative that veiling is a symptom of failed integration and the policy approach of restricting religious symbols as a means of promoting integration. Instead, they suggest that removing barriers to education and employment for veiled women — and combating discrimination in the labour market — would more effectively promote both gender equality and social cohesion.
As Europe grapples with questions of religious diversity, integration, and secularism, this research provides a timely reminder that the economic fate of Muslim women is shaped less by their religious convictions than by the barriers they face in the societies they inhabit.


















