Skip to main content

Muslims in France: identifying a discriminatory equilibrium


We analyze the assimilation patterns of Muslim immigrants in Western countries with a unique identification strategy. Survey and experimental data collected in France in 2009 suggest that Muslims and rooted French are locked in a suboptimal equilibrium whereby (i) rooted French exhibit taste-based discrimination against those they are able to identify as Muslims and (ii) Muslims perceive French institutions as systematically discriminatory against them. This equilibrium is sustained because Muslims, perceiving discrimination as institutionalized, are reluctant to assimilate and rooted French, who are able to identify Muslims as such due to their lower assimilation, reveal their distaste for Muslims.

This is a preview of subscription content, access via your institution.

Fig. 1
Fig. 2


  1. Following Constant et al. (2009) or Gorinas (2014), we measure the degree of assimilation by the level of identification with ancestral homeland and with the host country.

  2. For three of the 18 items measuring assimilation, we even find that these patterns diverge: SM lower assimilation decreases, while SX higher assimilation increases with the time spent in France.

  3. Technically, and relying on the taxonomy created by Harrison and List (2004), we have conducted a “framed field experiment” since, as we shall describe, we rely on a nonstandard subject pool, and these subjects receive an information set from the real world (the names of their game partners) that they can use in their game participation. We also ran a follow-up experiment in 2010 that also relied on a nonstandard subject pool that received an information set from the real world. Henceforth, for economy of expression, we refer to both interventions as “field experiments.”

  4. To avoid bias, we did not rely on landlines, as they would tap only the wealthiest and, least typical, of the immigrant population.

  5. The sample remains disproportionately Muslim, and this is not surprising given that Senegal is more than 90 % Muslim. Our identification of two Senegalese ethnic groups with substantial numbers of Christians is key, given the predominance of Islam in Senegal.

  6. Assimilation into French culture can be measured in part by the reduction in the level in which religion infuses other aspects of life. Indeed, in the 2006 World Values Survey for France, 13 % say that religion is very important in their lives; by contrast, in the 2008 Afrobarometer survey for Senegal, 95.8 % say that religion is very important in their lives. Relying on our definition of assimilation, those who claim that religion is less important in their lives are thus assimilating into French norms.

  7. Note that there is no difference in the probability of participating in an association between SM and SX.

  8. Politics in Senegal do not follow a typical left-wing/right-wing scale the way they do in France. Instead, the most common political dimensions in Senegal are region, Sufi order, and cousinage (see Smith (2010)).

  9. Alternatively, coefficient d could capture the fact that the first migrants in SX families systematically differ with regard to assimilation, depending on when they arrived to France. We rule out this possibility in the robustness checks.

  10. Alternatively, the sum of coefficients c and d could capture the fact that the first migrants in SM families systematically differ with regard to assimilation, depending on when they arrived to France. We rule out this possibility in the robustness checks.

  11. We thank an anonymous referee for bringing this issue to our attention.

  12. Controlling for the variable Time, as well as for the respondent’s socioeconomic characteristics, substantially reduces the number of observations due to missing data for these controls. However, the significance level for the dummy SM does not change if we impute missing data (results available upon request).

  13. The significance level for the dummy SM does not change if we impute missing data, with the exception of row 2 (whether the respondent believes immigrants should do whatever possible to avoid conflict with the hosting society) where the dummy SM loses significance (results available upon request).

  14. Given the coding of our measures of secularization, a positive coefficient indicates less secularization.

  15. The significance level for the dummy SM does not change if we impute missing data, with the exception of row 3 (the degree of sympathy of the respondent toward people sharing the same religion) where the dummy SM becomes significant (results available upon request).

  16. These results hold when we impute missing data, with the exception of column 6 where the positive coefficient of the interaction term SM.Time becomes significant (i.e., SM and SX attachment to Senegal and Africa diverges over time). More precisely, the desire to be buried in Senegal or in Africa increases over time for SM, while it decreases over time for SX (results available upon request).

  17. These results hold when we impute missing data (results available upon request).

  18. These results hold when we impute missing data (results available upon request).

  19. We thank an anonymous referee for bringing this issue to our attention.

  20. According to the 2009 Report on International Religious Freedom by the US Department of State, 97 % of the population in Bosnia and Herzegovina is either Christian or Muslim. Muslims stand for 46.5 % of this population (while Christians account for 53.5 %).

  21. Our subjects are coded by religious self-identification or, when that information is missing, ascribed religious heritage based on the advice of an ethnographer with expertise on Senegalese culture, who served on our research team.

  22. Note that this bias, if anything, runs against us finding FFF discrimination against SM since SM participants are more moderate in their religious practices.

  23. According to the 1999 French census, the percentage of individuals living in this district who are born in France is 63.5 (against 82.4 for all of Paris). A good picture of the diversity in the 19th district is offered in the French film “Entre les murs” (“The Class” in its English language version, a film directed by Laurent Cantet, 2008) that received the Palme d’Or at the 2008 Cannes Film Festival.

  24. In the exit questionnaire, we asked: “Que pensez-vous que notre équipe aura appris sur vous à travers vos décisions aujourd’hui?” [What do you think our team will have learned about you from the decisions you made today?]

  25. This stands for roughly 8.5 times the hourly minimum wage in France in 2009.

  26. At registration, we collected demographic data from participants, potentially priming them about identity issues, and thereby biasing our results. This is unlikely, however, given that at least 2 weeks separated the registration and game phases. Moreover, as explained in the text, what we told the players about our games and where we held the sessions served to downplay any suggestion that religious identities had any role in our intervention. The success of this strategy was revealed in our exit questionnaires, which asked participants what they thought our team had learned about them throughout the games: only one respondent out of a total 80 mentioned religion.

  27. Throughout the session, and out of sight from the players, monitors kept a full account of all answers and earnings for each player. At the end of the session, as players answered an exit survey, the winnings for each player were placed in sealed envelopes for them to take home. Full protocols (in French, but with English translations) are available upon request. Here, we review only what is necessary for interpreting the results presented in the subsequent section. We take this opportunity to thank our six recruiters and monitors for their incredible hard work, intellectual contributions throughout, and dedication to the project: Mathieu Couttenier, Jacinto Cuvi Escobar, Karine Marazyan, Etienne Smith, Josselin Thuilliez, and Séverine Toussaert.

  28. In order to avoid the equal split option from becoming a focal point in this game (Schelling 1960), we did not offer it.

  29. The novelty of our simultaneous trust game with respect to the original trust game introduced by Berg et al. (1995) is in the simultaneity of the decisions made by the sender and by the receiver. We preferred the simultaneous trust game over the original trust game for several reasons. Our objective was to treat each trust game played by our subjects as a one-shot game in order to mimic everyday life random encounters between strangers. It was therefore critical to avoid any reputation effect that would have occurred if the receivers learned how much particular senders had sent in previous games. This procedure also brings a touch of realism since most interactions in real life happen under incomplete information. In this respect, removing sequentiality in the decision process looks less artificial. Furthermore, since our protocol introduced a socialization phase after the simultaneous trust game, in which players would get to know each other, we did not want their conversations to be biased by knowledge of their partners’ actions during the simultaneous trust game.

  30. A third possible confound of positive transfers by the sender in the trust game is risk aversion, yet risk aversion has not been shown to be a serious confounding factor. For instance, Eckel and Wilson (2004) do not find behavioral risk measures to be significantly correlated with the decision to send in the trust game. They conclude that “subjects do not see trust as a problem of risk” (Eckel and Wilson (2004), 464).

  31. See Trivers (1971) on the role that reciprocal altruism plays in cooperation.

  32. Originally, Arrow introduced statistical discrimination to account for employers discriminating racially when they believe that the unobserved determinants of performance are correlated with race, which is observable (Arrow (1998), 96).

  33. Recall that we find that our sample of rooted French players in 2009 is more left-wing in political ideology than is the ESS sample of rooted French, suggesting that our sample is more open to diversity than the average rooted French. We do not have a 2010 ESS to conduct such an analysis for our 2010 sample of rooted French players. However, our 2010 sample of rooted French players was recruited from the same district as our 2009 sample and was introduced to the games in the exact same manner. It is therefore likely that the direction of the bias in 2010 is the same as the one we found in 2009.

  34. Only two of the 11 SM with first names of Arabic origin were not characterized as “Muslim” by a majority of FFF (these first names are Ibou and Sidy). Similarly, only one of the seven SM with first names not of Arabic origin were characterized as “Muslim” by a majority of FFF (this first name is Ndeye).

  35. And indeed, although the bulk of Serers and Joolas have converted to Islam or Christianity, many of them continue to follow traditional beliefs (see Berg and Ruth (2009)).

  36. Player 1 is the sender when we analyze the amount sent and the receiver when we analyze the amount sent back. Player 2 is the receiver when we analyze the amount sent and the sender when we analyze the amount sent back. In Eq. 3, we not only focus on the behavior of SM and SX toward their game partners, but also on the behavior of FFF toward SM and SX. This is because we rely on Eq. 3 in the following section to test whether FFF discriminate against SM in the simultaneous trust game.

  37. We ignore columns 2, 4, 6, and 8 for now, which are analyzed in Section 5.2.1

  38. Our finding contradicts the claim by Jacquemet and Yannelis (2012) according to which anti-minority discrimination by the dominant group in Western countries does not reflect discrimination against any specific minority, but rather discrimination against all groups which sound foreign (what they call “ethnic homophily”).

  39. The fact that FFF do not treat SM and SX differently when FFF are senders but that they do discriminate against SM when FFF are receivers brings additional support to the fact that FFF exhibit taste-based, not statistical, discrimination. Indeed, the amount sent in the trust game, whether simultaneous or not, is commonly considered by experimental economists as a signal of trust (hence the name given to this game). In this context, if FFF exhibit taste-based discrimination, it is not surprising that they do not discriminate against SM when they behave as senders, but that they do discriminate when they behave as receivers. Besides belief-based reciprocal altruism, the amount sent back in the simultaneous trust game indeed captures unconditional altruism.

  40. See Algan et al. (2012) who show that even in the HLM sector where immigrant communities and rooted French live close to each other, housing surveys show very low levels of social interaction between them.

  41. Note that relying on this game runs against us finding any difference between SM and SX beliefs about FFF behavior toward SM and SX, respectively. As already mentioned, the 2009 strategic dictator game and its companion, the 2009 dictator game, took place after the speed-chatting game, once we can no longer identify anti-Muslim discrimination on the part of FFF, yet our results show that this phenomenon is not anticipated (or fully anticipated at least) by Muslims since the 2009 strategic dictator game reveals differences between SM and SX beliefs about FFF behavior toward SM and SX, respectively.

  42. Although this result holds both in a difference of means analysis (not shown here) and in the first three columns of Table 12, it loses significance in column 4 when the multiple imputation analysis is run (but the p value of 0.16 associated with the coefficient in row (1) is close to statistical significance). This weakening of the significance is likely due to the fact that multiple imputation typically generates high standard errors when the number of observations is low (Eq. 6 is estimated on 41 observations only).

  43. The project employed two ethnographers to run these interviews: Etienne Smith (then a Ph.D. candidate in political science at Sciences-Po, who had conducted extensive field research in Senegal, and speaks Wolof, the lingua franca of Senegal, a language spoken by nearly all of our respondents) and Mahnaz Shirali (a Ph.D. in Sociology at the Ecole des Hautes Etudes en Sciences Sociales who has published extensively on Muslim youth and gender in France and Iran).

  44. The significance level for the dummy SM does not change if we impute missing data, with the exception of column 3 (i.e., distrust toward the parliament) and column 5 (i.e., distrust toward the judicial system) where the dummy SM loses significance (results available upon request).

  45. The significance level for the dummy SM does not change if we impute missing data, with the exception of column 1 (i.e., agreement with the fact that the police treats individuals on an equal basis) and column 2 (i.e., agreement with the fact that immigration authorities treat individuals on an equal basis) where the dummy SM loses significance (results available upon request).

  46. In our simultaneous trust game, the first mover is the sender; in the voting game, it is the leader; and in our dictator game, it is the dictator.

  47. These results hold if we run a multiple imputation analysis.

  48. These results hold if we run a multiple imputation analysis.

  49. For expediency, a random half of the 2010 FFF players were shown half of our 2009 Senegalese and FFF players; the other random half were shown the other half of our 2009 Senegalese and FFF players.

  50. Adida et al. (2010) identify that Senegalese Muslim households in France earn, on average, 400 euros less than their Christian counterparts each month (the equivalent of 14 % of the average monthly household income for France in 2009). See Constant and Zimmermann (2008) and Bisin et al. (2011b) for further evidence on the relationship between low assimilation and poor economic performance in Europe.

  51. Adida et al. (2010) confirm that significant anti-Muslim discrimination prevails in the French labor market. They compare the rate of interview callbacks received by two French applicants of Senegalese background showing the same educational and work experience but differing on religion, with a similar experimental design as in Bertrand and Mullainathan (2004). They confirm that the Muslim applicant faces high prejudice in France in 2009: she is 2.5 times less likely to receive a callback for an interview than is her Christian counterpart.

  52. Our experimental results clearly show that rooted French players do not expect SM to be less cooperative than SX. However, rooted French more exposed to contact with minorities, like recruiters, may hold correct beliefs. Relying on 400 interviews with human resources managers in large French firms, Bouzar and Bouzar (2010) indeed show that H.R. personnel report partly conditioning their hiring decisions (this is at least a post facto rationalization of such decisions) on their beliefs about what Muslims will do to the esprit de corps of their work teams. Perception of Muslims’ higher attachment to their religion and culture of origin is listed among the factors underlying their preferences for non-Muslims over Muslims.


  • Adida CL, Laitin DD, Valfort M-A (2010) Identifying barriers to Muslim integration in France. Proc Natl Acad Sci 107(52):384–390

    Article  Google Scholar 

  • Adida CL, Laitin DD, Valfort M-A (2014) Gender norms, Muslim immigrants, and economic integration in France. Econ Polit 26(1):79–95

    Article  Google Scholar 

  • Adida CL, Laitin DD, Valfort M-A (2012) One Muslim is enough! Evidence from a field experiment in France. Working paper

  • Adida CL, Laitin DD, Valfort M-A (2013) Region of origin or religion? Understanding why immigrants from Muslim-majority countries are discriminated against in Western Europe. Working paper

  • Akerlof GA, Kranton RE (2000) Economics and identity. Q J Econ 115(3):715–753

    Article  Google Scholar 

  • Aleksynska M (2011) Civic participation of immigrants in Europe: assimilation, origin, and destination country effects. Eur J Polit Econ 27(3):566–585

    Article  Google Scholar 

  • Algan Y, Hémet C, Laitin DD (2012) Diversity and local public goods: a natural experiment with exogenous residential allocation. Working paper

  • Allport G (1954) The nature of prejudice. Doubleday, New York

    Google Scholar 

  • Arrow KJ (1972) Some mathematical models of race discrimination in the labor market. In: Pascal A H (ed) Racial discrimination in economic life. D.C. Heath, Lexington, pp 187–204

    Google Scholar 

  • Arrow KJ (1998) What has economics to say about racial discrimination.J Econ Perspect 12(2):91–100

    Article  Google Scholar 

  • Asante MK (2009) Serer. In: Asante MK, Mazama A (eds) Encyclopedia of African religion. Sage, Thousand Oaks

    Google Scholar 

  • Ashraf N, Bohnet I, Piankov N (2006) Decomposing trust and trustworthiness. Exp Econ 9(3):193–208

    Article  Google Scholar 

  • Austen-Smith D, Fryer RG (2005) An economic analysis of acting White. Q J Econ 120(2):551–583

    Google Scholar 

  • Axelrod R (1984) The evolution of cooperation. Basic Books, New York

    Google Scholar 

  • Becker G (1957) The economics of discrimination. University of Chicago Press, Chicago

    Google Scholar 

  • Berg EL, Ruth W (2009) Cultures of the sorld: Senegal. Benchmark Books, New York

    Google Scholar 

  • Berg J, Dickhaut J, McCabe K (1995) Trust, reciprocity and social history. Games Econ Behav 10(1):122–142

    Article  Google Scholar 

  • Bertrand M, Mullainathan S (2004) Are Emily and Greg more employable than Lakisha and Jamal? Am Econ Rev 94(4):991–1013

    Article  Google Scholar 

  • Bisin A, Patacchini E, Verdier TA, Zenou Y (2008) Are Muslim immigrants different in terms of cultural integration? J Eur Econ Assoc 6(2-3):445–456

    Article  Google Scholar 

  • Bisin A, Patacchini E, Verdier TA, Zenou Y (2011a) Errata corrige are Muslim immigrants different in terms of cultural integration? J Eur Econ Assoc 9(5):1012–1019

    Article  Google Scholar 

  • Bisin A, Patacchini E, Verdier TA, Zenou Y (2011b) Immigrants and the labour market. Econ Policy 26(65):57–92

    Article  Google Scholar 

  • Bisin A, Patacchini E, Verdier TA, Zenou Y (2011c) Formation and persistence of oppositional identities. Eur Econ Rev 55(8):1046–1071

    Article  Google Scholar 

  • Blalock HM (1967) Toward a theory of minority-group relations. Capricorn Books, New York

    Google Scholar 

  • Bouzar D, Bouzar L (2009) Allah a-t-il sa place dans l’entreprise? Albin Michel, Paris

    Google Scholar 

  • Caldwell C (2009) Reflections on the revolution in Europe: immigration, Islam, and the West. Doubleday, New York

    Google Scholar 

  • Camerer CF (2003) Behavioral game theory. Princeton University Press, Princeton

    Google Scholar 

  • Cannot-Brown W (2009) Jola. In: Asante M K, Mazama A (eds) Encyclopedia of African religion. Sage, Thousand Oaks

    Google Scholar 

  • Constant AF, Zimmermann KF (2008) Measuring ethnic identity and its impact on economic behavior. J Eur Econ Assoc 6(2-3):424–433

    Article  Google Scholar 

  • Constant AF, Gataullina L, Zimmermann KF (2009) Ethnosizing immigrants. J Econ Behav Org 69(3):274–287

    Article  Google Scholar 

  • Cox JC (2004) How to identify trust and reciprocity. Games Econ Behav 46(2):260–281

    Article  Google Scholar 

  • Davila A, Mora M (2005) Changes in the earnings of Arab men in the US between 2000 and 2002. J Popul Econ 18(4):587–601

    Article  Google Scholar 

  • Dustmann C (1996) The social assimilation of immigrants. J Popul Econ 9(1):37–54

    Article  Google Scholar 

  • Diop AM (1988) Stéréotypes et stratégies. In: Leveau R, Kepel G (eds) Les Musulmans dans la société française. Presses de la Foundation Nationale des Sciences Politiques, Paris, pp 77–98

    Google Scholar 

  • Eckel CC, Wilson R K (2004) Is trust a risky decision? J Econ Behav Org 55(4):447–465

    Article  Google Scholar 

  • Eckel CC, Wilson RK (2006) Judging a book by its cover: beauty and expectations in the trust game. Polit Res Q 59(2):189–202

    Article  Google Scholar 

  • Eguia JX (2012) Discrimination and assimilation. Working paper

  • Fang H, Loury GC (2005) Dysfunctional identities can be rational. Am Econ Rev 95(2):104–111

    Article  Google Scholar 

  • Fershtman C, Gneezy U (2001) Discrimination in a segmented society: an experimental approach. Q J Econ 116(1):351–377

    Article  Google Scholar 

  • Fordham S, Ogbu JU (1986) Black students’ school successes: coping with the burden of acting white. Urban Rev 18(3):176–206

    Article  Google Scholar 

  • Fordham S (1996) Blacked out: dilemmas of race, identity and the success at capital high. University of Chicago Press

  • Fryer RG (2009) An empirical analysis of acting white. J Public Econ 94(56):380–396

    Google Scholar 

  • Gastellu J-M (1981) L’égalitarisme économique des Serer du Sénégal. ORSTOM, Paris

    Google Scholar 

  • Georgiadis A, Manning A (2011) Change and continuity among minority communities in Britain. J Popul Econ 24(2):541–568

    Article  Google Scholar 

  • Gorinas C (2014) Ethnic identity, majority norms, and the native–immigrant employment gap. J Popul Econ 27(1):225–250

    Article  Google Scholar 

  • Gould E D, Klor E F (2012) The long-run effect of 9/11: terrorism, backlash, and assimilation of Muslim immigrants in the West. Working paper

  • Harrison GW, List JA (2004) Field experiments. J Econ Lit 42(4):1009–1055

    Article  Google Scholar 

  • Hertwig R, Ortmann A (2001) Experimental practices in economics: a methodological challenge for psychologists? Behav Brain Sci 24(3):383–403

    Google Scholar 

  • Jacquemet N, Yannelis C (2012) Indiscriminate discrimination: a correspondence test for ethnic homophily in the Chicago labor market. Labour Econ 19(6):824–832

    Article  Google Scholar 

  • Jamison J, Karlan D, Schechter L (2008) To deceive or not to deceive: the effect of deception on behavior in future laboratory experiments. J Econ Behav Org 68(3-4):477–488

    Article  Google Scholar 

  • Kahneman D, Knetsch JL, Thaler RH (1986) Fairness and the assumptions of economics. J Bus 59(4):285–300

    Article  Google Scholar 

  • Laitin DD (1995) Marginality: a microperspective. Ration Soc 7(1):31–57

    Article  Google Scholar 

  • Linares O (1992) Power, prayer and production: the Jola of Casamance, Senegal. Cambridge University Press, Cambridge

    Google Scholar 

  • Mackie G (1996) Ending footbinding and infibulation: a convention account. Am Soc Rev 61(6):999–1017

    Article  Google Scholar 

  • Manning A, Roy S (2010) Culture clash or culture club? National identity in Britain. Econ J 120(542):F72–F100

    Article  Google Scholar 

  • McWhorter JH (2000) Losing the race: self-sabotage in black America. Free, New York

    Google Scholar 

  • Mecham Q (2006) From the sacred to the state: institutional origins of Islamist political mobilization. PhD Dissertation, Stanford University

  • Paluck EL, Green DP (2009) Prejudice reduction: what works? A critical look at evidence from the field and the laboratory. Annu Rev Psychol 60:339–367

    Article  Google Scholar 

  • Pew Research Center Report (2011) The future of the global Muslim population. Projections for 2010–2030

  • Pope D, Price J, Wolfers J (2011) Awareness reduces racial bias. Working paper

  • Price J, Wolfers J (2010) Racial discrimination among NBA referees. Q J Econ 125(4):1859–1887

    Article  Google Scholar 

  • Riphahn RT (2003) Cohort effects in the educational attainment of second generation immigrants in Germany: an analysis of census data. J Popul Econ 16(4):711–737

    Article  Google Scholar 

  • Schelling TC (1960) The strategy of conflict. Harvard University Press, Cambridge

    Google Scholar 

  • Shields MA, Price SW (2002) The English language fluency and occupational success of ethnic minority immigrant men living in English metropolitan areas. J Popul Econ 15(1):137–160

    Article  Google Scholar 

  • Smith E (2010) Des arts de faire société: parentés à plaisanteries et constructions identitaires en Afrique de l’Ouest (Sénégal), Doctorat de Science Politique. Institut d’Etudes Politiques de Paris

  • Trivers RL (1971) The evolution of reciprocal altruism. Q Rev Biol 46(1):35–57

    Article  Google Scholar 

  • van Ours JC, Veenman J (2003) The educational attainment of second-generation immigrants in The Netherlands. J Popul Econ 16(4):739–753

    Article  Google Scholar 

Download references


The research reported in this paper is not the result of any for-pay consulting relationship. It was funded by the National Science Foundation “Muslim Integration into EU Societies: Comparative Perspectives,” Grant SES-0819635, David Laitin, P.I. We thank Ben Adida for building the web-based game for the March 2010 experiments. We also thank two anonymous referees, Yann Algan, John Bowen, Pierre Cahuc, Rafaela Dancygier, Henry Farrell, Harvey Feigenbaum, James Vreeland, and seminar/conference participants at the Georgetown University Center for Contemporary Arab Studies, Council of European Studies in Boston, and George Washington University Comparative Workshop for their very helpful comments.

Author information

Authors and Affiliations


Corresponding author

Correspondence to Marie-Anne Valfort.

Additional information

Responsible editor: Erdal Tekin

Electronic supplementary material

Below is the link to the electronic supplementary material.

(PDF 1.14 MB)

Rights and permissions

Reprints and Permissions

About this article

Cite this article

Adida, C.L., Laitin, D.D. & Valfort, MA. Muslims in France: identifying a discriminatory equilibrium. J Popul Econ 27, 1039–1086 (2014).

Download citation

  • Received:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI:


  • Assimilation
  • Muslim and Christian immigrants
  • Discrimination
  • France