The logo of Pole Emploi, the French national employment agency, sits on documents inside a job center in Cahors, France, on Thursday, Aug. 28, 2014. Balint Porneczi/Bloomberg

Kim Yi Dionne: This is the second post in this week’s series on immigrant integration in Europe.

****

This summer, the Monkey Cage ran a series of posts debating whether the news about Muslim integration in France was good or bad. On one hand, scholars have argued that large proportions of Muslims claim they feel French. Yet others have emphasized instead that Muslims in France remain distinctive on a number of key integration measures. A possible explanation for apparently contradictory findings is that integration can be an elusive term, and our conclusions depend on how integration is measured.

This may be true, but a more likely explanation is that existing research is not actually measuring Muslim integration in France. To do so, one would have to compare two groups of immigrants who are similar in all respects except for their religious identity, an incredibly difficult task in France, where most Muslim immigrants come from North Africa and with few non-Muslim counterparts. In the Trajectories and Origins survey employed by Bleich and Maxwell, for example, 56 percent of the Muslim sample is Maghrebi, such that any characteristic of that sample can be attributable to the fact that it is Maghrebi as much as to the fact that it is Muslim. If we want to be able to say something about Muslim integration in France, and attribute findings to the Muslim factor, rather than the Maghrebi factor, we need to compare Muslim and non-Muslim immigrants from the same region – or better yet, from the same country. By doing so, we gain confidence that any difference in integration we observe is attributable to religious identity rather than to region-of-origin.

Our research, summarized in a forthcoming book (Why Muslim Integration Fails: an inquiry in Christian-heritage societies, Harvard University Press), does just that. It identifies a group of immigrants to France who hail from the same two ethnic groups and the same socio-economic status in Senegal, vary in their religious identity, and migrated to France at the same time (the 1970s) and for the same economic reasons. Furthermore, the historical forces that led some of these Senegalese to convert to Christianity and others to convert to Islam in the mid-19th century were purely accidental. Southwest Senegal, where our sample originates, was penetrated by jihadists and Christian missionaries at the same time. And conversion to one of the Western religions was determined by crop cultivation: Christian missionaries helped develop and commercialize palm oils and wines, an enterprise that was scorned by Muslim jihadists who turned instead to the cultivation and development of the groundnut trade. These crops offered similar economic returns, and they were the easiest ones in the region to cultivate. We thus gain confidence that converts to Islam or Christianity did not differ in their levels of economic or social ambition. Finally, historical accounts show that the French colonial administration in Senegal did not favor Catholic over Muslim converts. In other words, we are not comparing apples to oranges.

In 2009, we ran a CV experiment comparing responses to Senegalese Christian and Muslim job applicants; we then surveyed 511 Senegalese Muslim and Christian immigrants in France, and obtained data on a battery of integration measures, which we conceptualize as both an individual’s attachment to the country-of-origin and to the host country. One advantage of this approach is that it offers a comprehensive account of the immigrant integration experience. A limitation is that it may mask or underplay important exceptions to the rule, such as Bleich and Maxwell’s finding that a significant proportion of Muslim immigrants in France say they feel French.

Still, our data reveal four disturbing trends.

First, Muslim immigrants in France face discrimination on the job market precisely because of their religious identity. Results from our CV experiment, in which employers responded to matched CV applications that differed solely on the religious identity of the applicant (holding constant the applicant’s Senegalese origins), indicate that a Muslim candidate is 2.5 times less likely to receive a job interview callback than is her matched Christian counterpart. Consistent with such discrimination, our survey of Senegalese Christian and Muslim immigrants in France reveals a significant income gap between the two. Muslim households make, on average, 400 euros less than matched Christian households each month, the equivalent of 15 percent of the average monthly income for France in 2007.

Distribution of household income, by religion. Senegalese Christian immigrants are more likely to earn more than their Muslim counterparts in France.

Third, and speaking directly to the debate above, Senegalese Muslims are less attached to their host country than are Senegalese Christians. They express less sympathy toward French people, and are less likely to believe they share much in common with the French. They are significantly less secular than their Christian counterparts, hence distinguishing themselves in a country that defines itself by its laïcité. Finally, they express significantly lower trust than do Senegalese Christians toward key French institutions from schools to firms, to the French administration. Given the discrimination they face on the French labor market, such distrust is hardly surprising.

Finally and perhaps most disturbingly, these patterns do not improve over time. The distinct separation of Senegalese Muslims from French norms and attitudes is characteristic not just of first-generation immigrants: it persists, and even increases in some instances, when we look at second- and third-generation immigrants as well.

In order to be able to say something about the fate of Muslim immigrants as Muslims, our strategy has identified a small group of immigrants from a single country, who vary in their religious affiliation. The flip side of our strategy is that we end up focusing on a small slice of the Muslim immigrant population, and certainly not on the prototype of the Muslim immigrant to France – which is North African. And yet this brings us back to our initial conundrum: the countries of North Africa do not contain large enough populations of Christian immigrants to compare to matched Muslims in France.

But by focusing on Muslims who are not spontaneously associated with the collective imagination of Islam in France, our findings offer an underestimate of the Muslim effect on immigrant integration. And even with an underestimate, the news for Muslim immigrant integration in France is on the pessimistic side of the debate.

****

Claire Adida is an assistant professor of political science at the University of California San Diego. 

David Laitin is the James T. Watkins IV and Elise V. Watkins Professor of Political Science at Stanford University. 

Marie-Anne Valfort is associate professor of economics at Paris School of Economics-Paris 1 Panthéon Sorbonne University.