Abstract

Building on emerging research into intergenerational contextual mobility, I use longitudinal data from France (1990–2008) to investigate the extent to which second-generation immigrants and the French majority continue to live in similar neighborhood environments during childhood and adulthood. To explore the persistence of ethnoracial segregation and spatial disadvantage, I draw on two measures of neighborhood composition: the immigrant share and the unemployment rate. The analysis explores the individual and contextual factors underpinning intergenerational contextual mobility and variation across immigrant-origin groups. The results document a strong stability of neighborhood environments from childhood to adulthood, especially with regard to the ethnoracial composition of the neighborhood. Individual-level factors are quite weak in accounting for these patterns compared with the characteristics of the city of origin. Moreover, the degree of contextual mobility between childhood and adulthood varies across groups. I find that neighborhood environments are more stable over time for non-European second-generation immigrants. The findings offer important new empirical contributions to the French literature on the residential segregation of immigrants and will more broadly be of interest to scholars of intergenerational spatial and social mobility.

Introduction

With socioeconomic inequalities on the rise in most Western societies, intergenerational social mobility has garnered renewed attention. The latest research from the United States indicates that upward social mobility is stagnating and that contemporary adults are less likely to achieve the income levels of their parents compared with previous generations (Chetty et al. 2014b, 2016a). Despite country variation, earnings and educational levels are also strongly persistent between parents and their children in Europe (Causa et al. 2009). These findings have triggered concern that contrary to the promise of meritocratic societies, social advantage is largely inherited from the previous generation rather than achieved by individuals over the life course.

Although the focus of intergenerational mobility studies has typically been placed on class (Blau and Duncan 1967; Grusky 2014), substantial research has suggested that social mobility between parents and their children is highly contingent on residential environments and race/ethnicity. Residential segregation and spatial disadvantage, affecting primarily blacks and black Hispanics in the United States (Crowder et al. 2012; Iceland 2009; Massey 2007; Pais et al. 2012; Quillian 2002; South et al. 2005, 2008), have important consequences for life chances. A growing body of evidence has indicated that growing up or living for long periods of time in poor segregated spaces negatively affects a wide range of socioeconomic, health, and residential outcomes, with the potential of perpetuating inequalities over long periods of time, notably along ethnoracial lines (Ellen and Turner 1997; Massey 2007; Sampson 2012; Sharkey and Faber 2014). Opportunities for intergenerational social mobility, moreover, vary substantially across geographical contexts in the United States and according to local characteristics (Chetty et al. 2014a). The insight that social and spatial mobility processes are interrelated—in ways that disadvantage minorities in particular—has pushed scholars in recent years to investigate whether spatial disadvantage, like the intergenerational transmission of socioeconomic positions, is inherited from the previous generation (Sharkey 2008; Van Ham et al. 2014; Vartanian et al. 2007).

These questions have yet to be empirically examined in the context of France, even while research has pointed to reduced social and spatial mobility prospects for immigrants and their offspring. This is particularly true for immigrants from north and sub-Saharan Africa: geographically concentrated within a few major metropolitan areas predominately within the public housing sector, these groups experience the greatest degree of residential segregation with respect to French natives (Pan Ké Shon and Verdugo 2015; Préteceille 2009; Safi 2009; Verdugo 2011) and hold the most disadvantaged positions on the labor market (Safi 2006). Despite the predominance of France’s republican tradition, which emphasizes sociocultural integration through citizenship, this crystallization of social and spatial inequalities does not appear to decline substantially among French-born second-generation immigrants. Evidence has shown that African-origin French are more likely to be unemployed (Aeberhardt et al. 2010; Meurs et al. 2006; Silberman et al. 2007; Simon 2003), reside in disadvantaged neighborhoods (Pan Ké Shon 2010), and face discrimination on labor and housing markets (Bunel et al. 2017; Jacquemet 2013). Such findings suggest that sociospatial outcomes might indeed be quite stable over time and across generations.

Using panel data from the L’échantillon démographique permanent (INSEE 1990–2008) matched with contextual variables from the French census, this article has three broad aims. The first is to describe intergenerational contextual mobility in France, or following the definition used by Sharkey (2008:931), the extent to which “inequalities in neighborhood environments persist across generations.” Unlike extant studies, which focus only on spatial disadvantage, this study uses two indicators of neighborhood environments: ethnoracial composition (the immigrant share) and socioeconomic composition (the unemployment rate). The second aim is to identify the individual and contextual factors that account for why neighborhood environments might be stable. For example, is contextual mobility over the life course attributable to social class alone? To what extent does it depend on contextual factors linked to the broader urban area? And what role does immigrant origin play? Distinguishing between four groups (the French majority, Europeans, Africans, Asians/Turks), I pay particular attention to whether second-generation immigrants experience a comparable degree of intergenerational contextual mobility as the French majority.

Why Would Inequalities in Neighborhood Environments Persist (or Decline) Across Generations?

Although the empirical effort to investigate intergenerational contextual mobility is recent, intergenerational dynamics have played a central role in theoretical approaches to spatial inequalities since the First Chicago School. The earliest model of residential segregation—the spatial assimilation theory—describes the process by which immigrants living in poor segregated areas will gradually improve housing and neighborhood quality as they become socioeconomically and culturally integrated into the host society (Alba and Logan 1993; Alba and Nee 2003; Massey and Denton 1985; Park and Burgess 1921). Inherent in this theory is the idea that with the successful upward social and residential mobility of their parents, children of immigrants will live in similar neighborhood environments as the majority, such that any differences in residential outcomes between groups will dissipate across generations.

This straight-line narrative—the assumption of a gradual process of integration—has, however, been challenged by subsequent theoretical models that highlight the role of race/ethnicity in shaping spatial inequalities. With the theory of segmented assimilation, Portes and Zhou (1993) worked with the assumption that the success of immigrant integration is dependent not only on time and acculturation but also on structural factors that affect ethnic/racial groups differently. If racial stratification is strong in the host society, ethnoracial groups may experience disparate assimilation trajectories: some groups quickly achieve similar social and residential positions to the majority, while others experience downward mobility into the lower ranks of society (Alba and Nee 2003; Portes and Zhou 1993). In this case, second-generation immigrants belonging to ethnoracial minorities may confront a comparable or even greater level of spatial disadvantage than their parents.

Segmented assimilation also highlights the role of ethnic enclaves. Living in proximity to coethnics has been shown to affect immigrants’ socioeconomic integration, even though debate exists as to whether these effects are positive or negative (Xie and Gough 2011). Regardless, the continued presence of immigrants and their offspring in enclaves contributes to enduring spatial disparities between groups.

The importance of race/ethnicity is also the linchpin of the place stratification perspective. This model highlights how two central processes—namely, racial discrimination and race-based residential preferences—obstruct opportunities for upward residential mobility by channeling minorities toward the least-valued segments of the housing market (Charles 2003; Massey and Denton 1993). Decades of research in the United States have revealed a wide range of discriminatory practices by landlords, real estate agents, and banks. In addition to direct refusals for housing, racial steering and redlining have contributed to enduring segregation. Place stratification further highlights the role of race-based residential preferences or preferences for living in proximity to members of one’s own group. Ethnoracial minorities may choose to remain in ethnic enclaves, as segmented assimilation posits, but whites tend to cluster in neighborhoods where minorities are not present in large numbers. Evidence of white flight or white avoidance strategies is widespread in the United States (Quillian 2002; Crowder 2000; Pais et al. 2009, Crowder et al. 2011). Hence, by emphasizing race/ethnicity, the segmented assimilation and place stratification models challenge the assumption of upward contextual mobility between generations for all groups, insisting instead on the persistence of inequalities as children inherit the residential outcomes of their parents based on their position in the racial hierarchy.

Beyond race-based processes, residential locations themselves may be considered an independent mechanism that inhibits or favors sociospatial mobility. The works of Wilson (1987) and Massey and Denton (1993), at the origin of the neighborhood effects literature, documented that growing up in severely disadvantaged neighborhoods is decisive in shaping a wide range of individual outcomes later in life. Exposure to poor neighborhoods (or cities and regions) during childhood has been found to be negatively associated with cognitive development, educational outcomes, employment, wages, physical and mental health, and residential outcomes during adulthood (Chetty et al. 2016b; Diez Roux 2001; Ellen and Turner 1997; Mayer and Jencks 1989; Sampson 2012; Sampson et al. 2008; Sharkey and Elwert 2011; Sharkey and Faber 2014).1 This research further underscores some mechanisms that explain why residential environments might have negative effects on individuals net of other factors. In poor neighborhoods, individuals face greater barriers to accumulating, maintaining, and transmitting material resources. This is because spaces that concentrate poverty also tend to concentrate a wide range of negative social and economic characteristics, such as crime, poor educational environments, limited access to jobs, low social trust, and a dearth of local institutions and amenities, all of which have consequences on individuals’ ability to move up the social ladder and leave behind disadvantaged neighborhoods (Sampson 2012; Sharkey 2013).

The effects of growing up in spaces with low resource endowment is compounded by the stigma associated with living in neighborhoods that are perceived to be disreputable, unattractive, or dangerous, forging status distinctions between individuals that serve as a basis for discrimination (Sampson 2012). Finally, early neighborhood environments shape residential preferences implemented later in life, such that spatial inequalities can be largely self-perpetuating (Sharkey 2013).

Two extant studies of intergenerational contextual mobility have drawn on these theoretical frameworks to explore whether living in poor neighborhoods in the first generation is significantly linked to living in poor neighborhoods in the second generation (Sharkey 2008; Van Ham et al. 2014). These studies concurred that neighborhood environments are relatively stable across generations, even after controlling for a wide range of factors. Sharkey (2008) showed that more than one-half of neighborhood poverty is inherited from the previous generation and that family socioeconomic background is quite weak in accounting for the transmission of residential context. In Europe, Van Ham et al. (2014) used Swedish data to show that the correlation between neighborhood poverty between parents and their children is quite durable over the long term. They found a significant association between neighborhood poverty across generations that continues up to 18 years after children leave the parental home. Finally, both Sharkey (2008) and Van Ham et al. (2014) found that the degree of intergenerational contextual mobility is stronger for ethnic/racial minorities than for white or majority populations.

Mechanisms of Intergenerational Contextual Mobility in France

A number of European scholars have challenged the application of theories forged in the United States to account for spatial inequalities affecting minorities on the other side of the Atlantic (Finney and Simpson 2009; Peach 2009; Wacquant 1992, 2008). In France, the predominance of the color-blind republican paradigm, which refuses to acknowledge racial/ethnic distinctions, has hindered the development of research into ethnoracial inequalities. Rather, class is emphasized as the legitimate lens through which to apprehend social or spatial outcomes. Moreover, the diverging histories of racial discrimination and urban development in the United States and Europe, as well as the strong role of the welfare state in contemporary European societies, have undermined claims that ethnoracial mechanisms are contributing to minority spatial concentration or that neighborhoods have effects in the European context.

However, growing evidence of ethnoracial inequalities in France suggests otherwise. Immigrants are strongly geographically concentrated within a few large urban areas: specifically, the peripheries of Paris, Lyon, and Marseille. Although overall levels of residential segregation are found to be lower than in the United States,2 non-Europeans—and African immigrants in particular—are the most segregated groups (Pan Ké Shon and Verdugo 2015; Préteceille 2009; Safi 2009; Verdugo 2011). Immigrants are less likely to move out of areas with high coethnic concentrations, and French natives tend to avoid spaces with immigrants (Rathelot and Safi 2014). Moreover, Africans are less likely net of other factors to exit disadvantaged neighborhoods, and they have a greater probability of moving into them (Pan Ké Shon 2010). Similar patterns are found in housing: net of other factors, Africans have greater chances of living in public housing than other groups (Fougère et al. 2013; Verdugo 2011) and have low access to homeownership (Gobillon and Solignac 2015). These findings coincide with evidence of discrimination on housing markets. Audit studies have shown that Africans are less likely to obtain an interview for housing than French native candidates with identical characteristics (Bonnet et al. 2016; Bunel et al. 2017; HALDE 2006), a finding that is corroborated by subjective declarations of discrimination in survey data (Safi and Simon 2013).

The public housing market and school system are key for understanding these forms of disparities in France and Europe more generally. State-funded public housing makes up a sizable share of the total French housing stock (approximately 18 %) and accommodates a large immigrant population. The concentration of immigrants in public housing is a result of urban policies aimed at resettling slum residents into the sector during the 1970s (Bernardot 1999) as well as the development of public housing projects on the industrial outskirts of urban areas where immigrants largely found employment. Hit hard by deindustrialization and rising unemployment, public housing neighborhoods increasingly came to concentrate low-income immigrant households and have become highly stigmatized, particularly following the urban riots of 2005 (Lagrange and Oberti 2006). Discrimination within the public housing sector—notably through the informal use of ethnoracial criteria in housing assignments—further contributes to the overrepresentation of non-European immigrants in the most disadvantaged projects (Bourgeois 2013; Sala Pala 2013; Simon et al. 2001). On the other hand, the presence of French natives in public housing projects has declined for socioeconomic reasons and because of native flight and avoidance strategies. The influx of large shares of immigrants in public housing coincides with a perceived decline in neighborhood and housing quality, triggering the departure of natives (Barou 2005).

Schools are also highly segregated on the basis of class and immigrant status (Felouzis 2003; Merle 2011; Oberti 2007; Oberti et al. 2012; van Zanten 2001). Children of non-European immigrants are concentrated in the most disadvantaged schools with poor educational achievement. School segregation is partly a consequence of districting policies that assign students to schools on the basis of place of residence, and it is further exacerbated by school flight—that is, strategies implemented by upper-middle class parents living in socieconomically or ethnoracially mixed areas to keep their children out of poor-quality zones. As with public housing, the presence of ethnoracial minorities is often perceived as a signal of an unfavorable school environment (Oberti et al. 2012).

This nexus of housing market segmentation and school segregation has resulted in the spatial concentration of disadvantage that disproportionately affects ethnoracial minorities. Initial quantitative research on neighborhood effects in France has indicated that spatially concentrated disadvantage has consequences for social mobility. Unemployed residents of immigrant neighborhoods have lower chances of finding a job, and residents of France’s most deprived neighborhoods have an increased risk of being unemployed (Aeberhardt et al. 2015; Gobillon et al. 2011; Rathelot 2014). Address-based discrimination targeting residents of France’s peripheral urban areas is an additional obstacle to jobs and housing (Bonnet et al. 2016; Bunel et al. 2017). These forms of disadvantage in turn likely restrict opportunities for contextual mobility over the life course and across generations.

Hypotheses

Building on these frameworks, I work with the assumption that neighborhood environments are in part inherited from the previous generation. I predict that neighborhood composition in childhood and adulthood is tightly correlated and that this correlation holds net of other factors (Hypothesis 1 (H1)). I further hypothesize about why neighborhood environments might be stable over time. One explanation lies in the socioeconomic and demographic characteristics of individuals and their parents. Because these factors sort individuals into neighborhoods of varying quality, controlling for such variables should strongly attenuate the correlation between neighborhood composition in childhood and adulthood. Higher education, upper-class occupations, and access to homeownership—all central to the spatial assimilation hypothesis—should favor intergenerational contextual mobility (Hypothesis 2a (H2a)). On the other hand, following the argument that context matters, structural constraints within the broader urban areas in which children grow up may also explain why neighborhood environments are transmitted. People may get mired in spaces with lower-quality schools and job and housing markets that provide reduced possibilities for social and residential mobility. In this case, controlling for contextual factors should weaken the correlation between neighborhood composition in childhood and adulthood (Hypothesis 2b (H2b)).

I also predict that ethnoracial neighborhood composition will be more persistent net of other factors than socioeconomic neighborhood composition (Hypothesis 3 (H3)). Ethnoracial-based processes, such as residential preferences and discrimination, could result in a stronger propensity for immigrants to remain in immigrant neighborhoods. In contrast, because spatial disadvantage is perhaps less rooted in preferences and/or discrimination and is more sensitive to individual socioeconomic characteristics, the correlation between socioeconomic neighborhood composition in childhood and adulthood may be less salient net of other factors.

The final hypothesis states that minorities may experience a lesser degree of intergenerational contextual mobility, such that they remain in neighborhoods of similar ethnoracial and socioeconomic composition over time. If certain neighborhoods offer reduced possibilities for social and residential mobility that affect everyone equally, mobility patterns should be similar for all groups regardless of origin (Hypothesis 4a (H4a)). If, on the other hand, discrimination and ethnoracial-based residential preferences tie certain groups to certain neighborhoods, a stronger correlation of neighborhood composition should be expected over time among the children of immigrants, and non-Europeans in particular (Hypothesis 4b (H4b)).

Data and Methods

Data come from the French panel L’échantillon démographique permanent (EDP), an ongoing panel conducted by the French National Institute of Statistics and Economic Studies (INSEE) approximately every seven to nine years since 1968 using a random sampling design based on day of birth.3 Information is compiled over time from two main sources: the French census and civil registries on vital events. Persons born on the sampling days enter the panel when their birth is recorded in the civil registries or when they are first observed in the French census. The panel is then updated with each new census or as events in the lives of EDP individuals are recorded in the registries.4 EDP can be merged with census data on neighborhoods (aggregated units for statistical information, or IRIS) using a geographic ID code. Because IRIS are not available prior to 1990 and are primarily used in municipalities (communes) of more than 10,000 inhabitants, I include only such municipalities in metropolitan France and the three most recent dates of observation (1990, 1999, and 2008).

Sample Selection

EDP is a rich source for studying intergenerational dynamics. When an EDP individual enters the panel at birth, she may be observed for one or several survey periods as a child. Children in the panel are identified on the basis of a variable indicating the position of the individual (father, mother, child, or other) within the household.5 For the years during which individuals are observed as children, a number of variables are provided referring to their parents. Consequently, I include in the sample only those individuals who transitioned to adulthood: namely, persons who appear in the panel at least once as children (in 1990 or 1999) and who are observed consecutively as an adult in a new household (in 1999 or 2008). This approach makes it possible to compare outcomes during childhood (the parental or childhood variables) with outcomes later in life (the adulthood variables). These restrictions raise some selection issues that demand caution when interpreting the contextual mobility patterns observed. First, because individuals appear as children in the panel, the resulting sample of adults (N = 33,913) is relatively young and observed over a short period (i.e., a period of nine years, from 1990 to 1999 or from 1999 to 2008). Summary statistics on age are provided in Table 1. Approximately 75 % of the sample individuals were first observed as adults aged 32 or younger, reflecting that most left the parental home in their late teens and early 20s. Five percent were first observed as adults at age 39 or older, suggesting a delayed transition to adulthood among some. I do not implement an age restriction so as to avoid selection into the sample based on the age at which individuals left the parental home.

However, this age-based heterogeneity likely has implications for contextual mobility. If residential choices are limited during the 20s—notably due to incomplete education, professional instability, singlehood, or lower income—individuals’ neighborhood environments may appear quite stable simply because they are observed early in the life course. Patterns may be different among adults in their 30s. I address this issue by testing the robustness of the main results on a subsample of adults older than 30 (N = 12,538). I also conduct an analysis on a subsample of individuals who were observed three times in the panel: once as a child in 1990, and twice as an adult in 1999 and 2008 (N = 10,149). Drawing on the panel nature of the data in this way allows better assessment of long-term patterns of contextual mobility—that is, up to 18 years after leaving the parental home.

The issue of panel attrition also requires attention. Attrition concerns children whose transition to adulthood is never observed because they disappeared from the panel over time. In the  appendix, I discuss this issue in further detail and provide summary statistics (Table 9) on EDP children according to whether a transition to adulthood was observed. Compositional differences between groups are minor, particularly with regard to the neighborhood composition variables. Children who did not transition to adulthood over the period originated in neighborhoods with comparable immigrant shares and unemployment rates as compared with children who did.

Immigrant Origin Categories

As is typical in the majority of statistical sources in France, EDP does not collect information on race or ethnicity (Simon 2008), nor is it possible to formally identify the origin of French-born descendants of immigrants in these data. Yet, because of the availability of parental variables among EDP individuals who were observed as children, a national origin can be assigned to children of immigrants. Drawing on the official definition used in France, immigrants are persons born outside of France without French citizenship at birth.6 Taking parental country of birth as a proxy for the origin of EDP children, I create three aggregated categories, representing the largest immigrant groups in France: southern Europeans (Spain, Italy, and Portugal), Africans (Algeria, Morocco, Tunisia, and SSA), and Asians/Turks (Vietnam, Laos, Cambodia, and Turkey).7 The latter are grouped because of small sample sizes. Collectively, these groups are referred to as children of immigrants. By definition, children of immigrants may have been born in France or migrated as children. To account for nativity, I use an indicator of having been born in France in all models that include immigrant origin. Finally, the French majority are individuals born with French citizenship whose parents were also born with French citizenship. Table 2 displays the sample (N = 33,913). More than four-fifths (83 %) belong to the French majority. The largest immigrant origin groups are Europeans and Africans.

Neighborhood Composition Variables

To explore intergenerational contextual mobility in terms of both ethnoracial segregation and spatial disadvantage, I use two indicators of neighborhood (IRIS8) composition pulled from the French census: the neighborhood immigrant share and the neighborhood unemployment rate. The neighborhood immigrant share is the proportion of immigrants (persons born outside of France without French citizenship at birth) of the entire IRIS population.9 The unemployment rate refers to the proportion of unemployed individuals of the working population of the IRIS.

The neighborhood composition variables measured during adulthood are used as dependent variables in the models, and the main independent variables of interest are the respective neighborhood composition variables measured during childhood.10 The correlation between these variables is interpreted as the degree of intergenerational contextual mobility, with higher values indicating greater stability within similar neighborhood environments. In some analyses, I divide these variables into quartiles to examine childhood/adulthood transitions between low-concentration (the first quartile) and high-concentration (the fourth quartile) neighborhoods. Table 1 shows the quartile distributions. The French majority are quite evenly spread out in quartiles of the childhood and adulthood immigrant share and unemployment rate, with a somewhat greater proportion residing in low-concentration areas. Children of immigrants, on the other hand, are overrepresented in high-concentration immigrant and unemployment areas.

Covariates

A number of variables collected during childhood and adulthood are included as covariates. Those measured during childhood describe the characteristics of parents11 and include age and age squared, occupation, education, marital status, number of children in the household, and housing tenure. To capture heterogeneity linked to the broader urban area, some models control for municipality fixed effects for the municipality where individuals lived as children. The attributes of individuals collected during adulthood and used as controls include age and age squared, gender, occupation, education,12 marital status, number of children, housing tenure, city size, and year of observation. Immigrant origin and nativity are included in some models. Summary statistics for all covariates are included in Table 1.

Modeling Strategy

I apply ordinary least squares (OLS) regressions predicting neighborhood composition during adulthood. Models 1 and 2 use, respectively, the immigrant share and unemployment rate as the dependent variable. I first estimate three specifications of these models to explore intergenerational contextual mobility and its correlates. The first includes only the respective neighborhood composition variable measured during childhood as a covariate (specification a). The second integrates all individual-level controls relating to individuals and their parents (specification b). The third adds municipality fixed effects (specification c).13 Introducing these controls incrementally makes it possible to break down the original correlation observed in the a specifications and identify their relative importance. As a robustness check, the models are replicated on the subsample older than 30 (N = 12,538).14

I run additional model specifications with the aim of identifying whether patterns differ for children of immigrants and the French majority. The fourth specification introduces immigrant origin and nativity as covariates (specification d). I use this specification to test an interaction between immigrant origin and childhood neighborhood composition on the full sample and on a subsample of individuals who originated in the Paris region15 (N = 7,489).16

Finally, Models 3 and 4 are random-effects OLS regressions predicting the immigrant share and the unemployment rate in the neighborhood during adulthood, drawing on the subsample observed twice during adulthood (N = 10,149). I focus on the effect of the respective neighborhood composition variable during childhood and its interaction with year of observation during adulthood to determine whether neighborhood environments are stable up to 18 years after leaving the parental home (1990 to 2008). The models include the same controls as those used in the previous models.

Results

Table 3 displays mobility between quartiles of the neighborhood immigrant share and unemployment rate from childhood to adulthood. The first finding of note is that the chances of remaining within the same types of neighborhoods are quite high. Regardless of the origin group, 41 % or more remain in the same immigrant share quartile between childhood and adulthood. Stability is somewhat lower when measured by the unemployment rate, with approximately one-third staying in the same quartile, suggesting greater intergenerational contextual mobility in terms of socioeconomic neighborhood composition.

Yet, substantial disparities are found across origin groups in the likelihood of remaining in similar spaces over time. Nearly one-half of the majority (42 %) and Europeans (48 %) who originated in high-concentration immigrant areas (>75th percentile) remain in similar neighborhoods as adults. The story for children of non-European immigrants is much different. Africans and Asians/Turks who grew up in neighborhoods with the highest shares of immigrants have, respectively, a 63 % and a 69 % chance of living in those types of spaces as adults—an increase of approximately 20 points. Similarly, the French majority and Europeans have the lowest risk of staying in high-concentration unemployment areas (39 % for the former and 33 % for the latter); at 54 %, Africans have the highest chances, followed by Asians/Turks (47 %).

The trend is reversed in low-concentration immigrant areas (<25th percentile). French and Europeans remain in these spaces at the greatest rates (54 % for the former and 43 % for the latter). On the other hand, non-Europeans—especially Asians/Turks—have considerably lower chances of originating in these neighborhoods and staying in them over time. The same pattern by origin group is found in low-concentration unemployment neighborhoods.

Mobility into neighborhoods with higher and lower shares of immigrants or the unemployed can also be discerned in Table 3. Because mobility to a higher quartile is not possible for those who grew up in the fourth quartile, nor is mobility to a lower quartile possible for those who grew up in the first quartile, these cells are empty. Generally speaking, the probability of moving up quartiles from low-concentration areas—that is, toward neighborhoods with greater shares of immigrants or the unemployed—is stronger than the probability of moving down quartiles from high-concentration areas. The French majority is the only exception. More than one-half (58 %) of those originating in a high-concentration immigrant area move to a lower quartile over time. This is true of 52 % of Europeans but only about one-third of non-Europeans. On the other hand, among those who originated in the bottom quartile of the immigrant share, the risk of entering a higher quartile in adulthood is lowest among the French majority (46 %) and highest among Africans (64 %) and Asians/Turks (95 %). Similar disparities are observed in mobility patterns with regard to the neighborhood unemployment rate. Once again, non-Europeans have the lowest likelihood of leaving high-concentration areas for a lower quartile and the greatest chances of entering a higher quartile from low-concentration areas.

These trajectories between neighborhood quartiles reflect not only individual moves to different neighborhoods, but they also capture neighborhood change. If the original neighborhood underwent an increase or decrease in the immigrant share or unemployment rate, changes in neighborhood quartiles may be observed for individuals who stayed in the same IRIS over time. As shown earlier in Table 1, geographic mobility—defined as moving to a different IRIS between childhood and adulthood17—is strong (between 81 % and 86 %), indicating that individual moves contribute substantially to the patterns in Table 3.

Table 4 helps shed light on the role of neighborhood change by indicating, among individuals who are not geographically mobile, the share remaining in the same neighborhood quartile between childhood and adulthood. Neighborhood change is not predominant in terms of ethnoracial composition: 66 % of nonmovers remained in the same quartile of the immigrant share. Greater change is observed on the socioeconomic indicator, with 45 % of nonmovers remaining in the same quartile of the unemployment rate. Moreover, the majority and Europeans live in neighborhoods that are more likely to undergo change. This finding suggests that in addition to geographic mobility, a lower degree of neighborhood change also contributes to the lower levels of contextual mobility observed among non-Europeans.

Hence, these descriptive patterns emphasize a strong continuity between the types of neighborhoods where individuals live as children and adults. Ethnoracial segregation appears to be particularly persistent relative to spatial disadvantage, especially for children of non-Europeans immigrants. I next use OLS regressions to model the correlation between childhood and adulthood neighborhood composition as an indicator of the stability of neighborhood environments. Results for the coefficients of interest are reported in Table 5.

I begin by exploring the correlation between neighborhood composition in childhood and adulthood in the models not including controls (Models 1a and 2a). Both coefficients are positive and significant: higher shares of immigrants and the unemployed in the neighborhood during childhood are correlated with higher shares of immigrants and the unemployed in the neighborhood during adulthood. Yet, the magnitude of this correlation varies across indicators and appears to be most intense in immigrant neighborhoods (Model 1a). A 1 % increase in the childhood neighborhood immigrant share translates into a 0.38 % increase in the neighborhood immigrant share during adulthood. The coefficient from the model predicting the unemployment rate (Model 2a) is comparatively somewhat lower. As the child neighborhood unemployment rate increases by 1 %, the unemployment rate in the neighborhood during adulthood rises by 0.25 %. Results are similar for the subsample of adults older than age 30.

Next, I examine whether individuals remain in certain neighborhoods over time as a result of their own demographic and socioeconomic characteristics or those of their parents. Models 1b and 2b report the coefficients of interest when all individual-level factors are added to the estimation. Adding these controls results in a reduction of the coefficients by approximately 8 % for the immigrant share coefficient (0.347 / 0.377) and approximately 15 % for the unemployment rate coefficient (0.211 / 0.248). Again, findings are confirmed in the age-restricted subsample. Individual-level factors relating to individuals and their families thus at least partially account for why neighborhood composition is stable over time. Nonetheless, these factors appear more decisive in accounting for neighborhood stability in terms of socioeconomic composition than ethnoracial composition.

The last specifications in Table 5 report the coefficients for the child neighborhood variables after controlling for municipality fixed effects (Models 1c and 2c). The aim here is to determine whether remaining in neighborhoods with similar ethnoracial and socioeconomic composition can be attributed to the broader context in which children of immigrants and the French majority grew up. Compared with the empty models, the correlations between the childhood and adulthood neighborhood immigrant share and unemployment rate are reduced substantially, by approximately 75 % (0.102 / 0.377 and 0.063 / 0.248, respectively). Thus, the continuity of neighborhood environments over the life course appears to have a lot to do with geographic location during childhood.

As the full model results in Table 6 show, socioeconomic status in both generations matters to neighborhood environments experienced in adulthood. Individuals with higher education, and whose parents have higher occupational status and education, show decreased unemployment rates and immigrant shares in their neighborhoods as adults, although these factors show lower significance in the age-restricted models. Accessing homeownership is linked to lower shares of immigrants and the unemployed in the neighborhood, whereas public housing occupancy in both generations is positively correlated with ethnoracial segregation and spatial disadvantage in adulthood.

Hence, both individual and contextual factors are prominent in accounting for the types of neighborhoods in which people live. Nonetheless, the characteristics of individuals and their parents seem to be relatively weak compared with the salience of the broader geographic context in which one grew up. Still, net of controls, childhood neighborhood composition continues to exert an influence on where individuals live as adults.

I now shift the analysis to explore whether contextual mobility is restricted among children of non-European immigrants. To do so, I add the immigrant origin and nativity covariates (Models 1d and 2d). Origin-based disparities in neighborhood composition during adulthood emerge from both models. As shown in Table 7, net of controls, children of immigrants live in neighborhoods with significantly higher shares of immigrants compared with the majority (Model 1d). Disparities between non-Europeans and the majority are the most pronounced: Africans and Asians/Turks live in neighborhoods with higher average immigrant shares of 2 to 5 percentage points. Likewise, the greatest levels of disadvantage are found in non-European neighborhoods, and no significant differences are found between Europeans and the majority (Model 2d). Nativity is significantly negatively correlated with higher shares of immigrants in the neighborhood during adulthood.

To confirm whether patterns differ between non-Europeans and the majority net of other factors, I use Models 1d and 2d to test an interaction between immigrant origin and childhood neighborhood composition on the full sample as well as on a subsample of individuals who grew up in the Paris region. Results are displayed in Fig. 1. The slope for each group indicates the effect of the childhood immigrant share or unemployment rate on the respective neighborhood composition variable during adulthood. Findings on the full sample show a weak slope for the majority on both indicators, pointing to a relatively low stability of neighborhood environments over time. For all immigrant origin groups, and particularly non-Europeans, the effect of childhood neighborhood composition is significantly stronger. As the local immigrant share in childhood increases, the local immigrant share in adulthood rises sharply. The same is true for the neighborhood unemployment rate. However, in this case, Europeans do not significantly differ from the majority. Non-Europeans stand out from the pattern, showing a distinctly stronger effect of the unemployment rate in the original neighborhood.

Among the subsample of the Paris region, the slopes are stronger overall, suggesting a greater degree of stability in neighborhoods over time. The same disparities across immigrant origin groups are found on the immigrant share indicator, with significant differences between the French majority and children of immigrants. However, in terms of neighborhood socioeconomic disadvantage, only Africans are distinct from the French majority, with greater stability on the unemployment rate indicator. Differences between other groups and the majority are not significant.

Finally, Table 8 displays Models 3 and 4. These models are run on the subsample of individuals observed as children in 1990 for whom two observations during adulthood are available, in 1999 and 2008 (N = 10,149). The models again predict adulthood neighborhood composition, this time with an interaction between childhood neighborhood composition and the year of observation in adulthood (1999 and 2008) in order to test the stability of the correlation over time. The first part of the interaction term shows the positive and significant effect of the respective childhood neighborhood composition variable at the first adult observation in 1999. At this point in time, a stronger coefficient is found for the neighborhood unemployment rate than for the neighborhood immigrant share. The second part of the interaction term shows the difference in the effect of the childhood neighborhood between 1999 and 2008. The correlation between neighborhood unemployment in childhood and adulthood drops significantly, by approximately 9 percentage points in 2008. The difference in the effect of the childhood neighborhood immigrant share, however, is not significant. These patterns suggest that individuals may access neighborhoods with a lower degree of disadvantage later in life. Ethnoracial segregation, on the other hand, remains strong up to 18 years after transitioning to adulthood.

Discussion and Conclusion

This study used recent longitudinal data to investigate intergenerational contextual mobility in France. The analysis had three broad aims: (1) identify the extent to which individuals live in similar neighborhood environments over the life course as measured by ethnoracial and socioeconomic composition; (2) determine how well individual and contextual characteristics account for these patterns; and (3) assess whether neighborhood environments are more stable for the children of immigrants compared with the French majority.

The findings are in line with prior research on the topic (Sharkey 2008; Van Ham et al. 2014; Vartanian et al. 2007) and indicate that in France, neighborhood environments are partially inherited from the previous generation (H1). Both neighborhood composition variables remain significantly correlated between childhood and adulthood net of other factors. Neighborhood ethnoracial composition, in comparison with neighborhood socioeconomic composition, was particularly stable over time (H3). Although these patterns must be considered with regard to the young age of the sample and limited time frame of observation, additional analysis showed that the findings are robust among older adults and hold, in the case of neighborhood ethnoracial composition, up to 18 years after children leave the parental home.

The results further shed light on the weight of individual and contextual factors in explaining contextual mobility. The demographic and socioeconomic characteristics of individuals and their parents only somewhat contribute to these patterns, undermining H2a. Socioeconomic characteristics in both generations were found to affect adult ethnoracial segregation and disadvantage yet only weakly accounted for the correlation between neighborhood composition over the life course. Of course, variables omitted from this analysis because of data restrictions, such as income and wealth, influence where people live and may explain why individual factors appear to have limited importance in these models. Yet it seems that socioeconomic factors alone do not explain why neighborhood environments are transmitted between parents and their children.

On the other hand, introducing contextual controls in the models resulted in a sharp decrease in the effect of neighborhood composition during childhood. Moreover, stability within similar neighborhoods was markedly stronger for children of immigrants from the Paris region. Given that previous studies of intergenerational contextual mobility focused primarily on individual mechanisms, this finding offers a new contribution about the role of the broader geographical environments where individuals grow up (H2b). The municipality fixed effects included in the models capture unobserved characteristics of cities, such as housing prices, the size of the public housing stock, and the quality of local education and employment. Growing up in areas with tight housing markets and large shares of public housing may greatly reduce neighborhood choice, and social mobility may be inhibited among individuals who are embedded within low-quality school and job environments. These findings are in line with theoretical arguments within the neighborhood effects literature, according to which residential contexts (neighborhoods, but also cities and regions) bolster or restrict opportunities for social and residential mobility (Chetty et al. 2014a; Sharkey and Faber 2014).

Immigrant origin also emerged as a salient predictor of intergenerational contextual mobility (H4b). Net of other factors, the effect of childhood neighborhood composition was significantly stronger for non-Europeans. This conclusion joins the findings of Sharkey (2008) in the United States and Van Ham et al. (2014) in Sweden, who documented a pronounced risk of remaining in poor neighborhoods among racial minorities. The children of Europeans, on the other hand, like the French majority, are less exposed to segregated and disadvantaged spaces over time.

These origin-based disparities, as well as the relatively stronger persistence of ethnoracial neighborhood composition over generations and over time, point to the salience of race/ethnicity in spatial patterns in France. Several mechanisms help account for why non-Europeans in particular experience reduced contextual mobility. The first may be related to residential preferences. The segmented assimilation literature (Logan et al. 2002) emphasizes that living in proximity to coethnics provides networks that may be useful to integration on job and housing markets. Toma (2016) found evidence of ethnic enclave returns to job market integration in France among immigrants. Furthermore, preferences to remain in coethnic neighborhoods may be heightened in France, where anti-immigrant sentiment and racism are not negligible (Mayer et al. 2014). The residential segregation of non-Europeans in previous generations could also be reproduced among their children because the latter, having been socialized in specific neighborhoods, form preferences for the residential environments experienced during childhood (Sharkey 2013).

If neighborhood choice is indeed driven by preferences for coethnic neighborhoods, this could explain the lower degree of contextual mobility among non-Europeans, even when upward social mobility has occurred. On the other hand, the residential preferences of the French majority could also be contributing to these patterns. Research in the United States has provided substantial evidence of the role of white flight in maintaining segregated neighborhoods, and quantitative findings from France (Rathelot and Safi 2014) and Europe (Bolt et al. 2008; Bråmå 2006; Van Ham and Clark 2009) align with this hypothesis. Native flight and avoidance processes are linked to ethnoracial concentration in public housing projects and disadvantaged schools. Middle- and upper-class French natives can more easily buy into attractive neighborhoods on open housing markets, with the consequence of reducing the range of neighborhoods available to minorities, confining them to similar spaces over time.

The restricted contextual mobility among non-Europeans may also be linked to their presence in public housing. Verdugo (2011) found that the concentration of non-Europeans in public housing accounts for increased residential segregation among these groups. Preferences may again be a driving factor of this outcome: prior research has documented that entering the public housing sector can be a veritable residential strategy among immigrants to improve housing quality and access to homeownership, especially in expensive urban housing markets areas where the public sector is particularly financially attractive (Dietrich-Ragon 2011; Goffette-Nagot and Sidibé 2016). Second-generation immigrants who grew up in public housing may choose to remain in the sector or may have greater access to information about housing opportunities there. Yet, beyond preferences, discrimination in both public and private housing markets, widely documented in France (Bonnet et al. 2016; Bourgeois 2013; Bunel et al. 2017; Sala Pala 2013; Simon et al. 2001), durably restricts opportunities for contextual and social mobility among non-European minorities. Direct refusals for housing by real estate agents and landlords, as well as racial steering, undoubtedly play a key role. Little is known in France about redlining, but evidence of banks denying loans to geographical areas with large immigrant populations has been documented in other European countries (Aalbers 2005).

This article presents some analytical limits that invite avenues of future research. As new EDP waves become available, intergenerational contextual mobility may be assessed over the long term. This will overcome the present limit of observing young adults over a short period of time and help grasp mobility dynamics as a process that unfolds over multiple decades. Moreover, given the importance of context highlighted in these analyses, further analysis is needed to explore how geographic mobility (between urban areas, departments, or regions) specifically contributes to intergenerational contextual mobility. Finally, future studies could explore the articulations between intergenerational social mobility and contextual mobility across ethnoracial groups. Research into these questions are of particular importance in France, where scholarly assessments of ethnoracial stratification are undermined by the statistical invisibility of race/ethnicity (Safi 2013; Simon 2008), even while segregation, spatial disadvantage, and social mobility among immigrants are politically salient topics. Much would be gained by using the lens of intergenerational inequalities to continue this investigation of the interconnectedness of space, social mobility, and ethnoracial inequalities in France and other national contexts.

Acknowledgments

This research was supported by the Flash Asile program of the French Agence Nationale de la Recherche (ANR-16-FASI-0001). Data access was made possible by the Centre d’accès sécurisé aux données (CASD), supported by a French state grant (Grant No. ANR-10-EQPX-17). I also thank Denis Fougère, Mirna Safi, Yannick Savina, Gregory Verdugo, and the three anonymous reviewers for their helpful comments.

Appendix

Restricting the sample to children who transition to adulthood raises issues of attrition and censoring. Attrition concerns individuals who are observed as children in t but leave the panel before they can be observed as adults. As Table 9 shows, 18 % of children observed in t disappeared from the panel in t + 1. Another 26 % were observed twice as children but then left the panel in t + 2. Thus, the transition to adulthood is not observable for approximately 44 % of all children because of attrition. Although this rate is high, sample attrition must be put into perspective with fact that after 2004, the French census and EDP switched from collecting data on the entire population to only 20 % of the population. As a result, not all EDP individuals were relocated after 1999, increasing the attrition rate after this period. However, the loss of EDP individuals because of this change can be considered random.

Censoring, on the other hand, concerns individuals whose outcomes cannot be observed because they have not yet occurred. This analysis presents two cases. The first, concerning 31 % of the sample, comprises children who entered the panel at the last available date of observation (2008). These are presumably young children whose births were recently recorded in the civil registries. The second, 6 % of the sample, concerns individuals who remained children at all three dates. Such persons may also have been young at the first date of observation or left the parental home at a later age. In both cases, the transition to adulthood cannot be observed until a future EDP date is available.

To get a sense of how attrition and censoring may affect the analysis, Table 9 provides descriptive statistics on the sample of EDP children according to whether a transition to adulthood was observed. The most substantial differences between the samples concern immigrant origin and housing tenure. Lower rates of non-European children of immigrants and greater rates of homeowners are found in the transition to adulthood sample, which may suggest that the analysis sample is somewhat positively selected on socioeconomic characteristics. The higher attrition of immigrant populations may be due to remigration patterns. Nonetheless, the similar composition of the samples suggests that the analyses are not severely biased by these differences.

Notes

1

Not all studies within this literature are able to make a causal claim about the negative effects of living in poor neighborhoods. A number of unobserved factors influence where one lives while simultaneously shaping individual outcomes, thus making it difficult to distinguish the effect of neighborhoods from the effect of other characteristics that select individuals into neighborhoods. The most robust evidence comes from experimental studies based on the Moving to Opportunity (MTO) and Gautreaux programs. These findings have shown that moving out of poor neighborhoods improves residential outcomes and health later in life (Keels et al. 2005; Leventhal and Brooks-Gunn 2003) but has only a minor effect on economic and educational achievement (Ludwig et al. 2008; Sanbonmatsu et al. 2006). The most recent research from MTO, however, documented that moving out of a poor neighborhood does have consequences for education and earnings, but that these effects are contingent on other factors, particularly the age at the time of the move and the duration of exposure to concentrated disadvantage (Chetty et al. 2016a).

2

Nevertheless, comparing segregation levels in France and the United States is complicated by categorical differences between ethnic/racial minorities and immigrants across contexts. Although U.S.–based measures of segregation draw on ethnoracial categories declared in the census, research in France is generally confined to first-generation immigrants distinguished by national origin. Segregation measures in France thus exclude immigrant offspring from the second generation and beyond, likely resulting in an underestimation of minority spatial concentration.

3

From 1968 to 1999, individuals entered the panel if they were born on the first four days of October. Since 2006, the sample was broadened to integrate individuals who are born on 16 days of the year (four days respectively in January, April, July, and October).

4

The periodicity of EDP follows that of the French census. From 1968 until 1999, the French census was conducted on the entire population at an interval of every seven to nine years (1968, 1975, 1982, 1990, and 1999). EDP was enriched with new information from the census at this regularity. Since 2004, however, the French census has been conducted every year on 20 % of the population. A cycle of five years is thus required for the census to be completed. Likewise, although EDP data are now updated annually with each new census, five years must be aggregated to obtain a complete wave. In addition to the five previous waves (1968, 1975, 1982, 1990, and 1999), I thus compile years 2004, 2005, 2006, 2007, and 2008 to form the sixth wave of the panel. I control for year of observation in all models.

5

Noninstitutional, private households.

6

In France, two criteria are used to define immigrants: nationality at birth, and country of birth. French natives are defined only on the basis of nationality at birth because of France’s colonial history: French citizens by birth who were born in the former colonies and who returned to France following decolonization are distinguished from immigrants.

7

The variables referring to the origin of the father are used first; when the latter are not available, the mother’s origin is used. This choice is justified by the fact that the father transmits the last name, which can be a marker of difference and source of discrimination. Even though the analysis does not use any dates prior to 1990, I draw on all available years of observation to identify the immigrant origin of EDP individuals.

8

IRIS are inframunicipality units of between 1,800 and 5,000 inhabitants, somewhat smaller than U.S. census tracts, on average. All French municipalities of more than 10,000 inhabitants, and the majority of those with more than 5,000 inhabitants, are broken down into IRIS. IRIS were not implemented until 1999. Prior to this date, the inframunicipality division used was the îlot. I use the îlot/IRIS correspondence table provided by INSEE to match the 1990 îlots with the 1999 IRIS code. The poor quality of the geographic ID codes in EDP before 1990 makes a neighborhood-level spatial analysis difficult.

9

Because they are not formally identified in the census, second-generation immigrants are not included in the calculation of the immigrant share.

10

For EDP individuals who were observed more than once in childhood or adulthood, the childhood variables correspond to the last observation in childhood, and the adulthood variables correspond to the first observation during adulthood. The same is true of the covariates.

11

As with parental country of birth and nationality at birth, the variables referring to the father are used first; when the latter are not available, the variables referring to the mother are used.

12

Occupation and education are used to measure social class in the absence of income and wealth in EDP. These are typical measures of social class used in empirical research in France, notably on intergenerational social mobility (Lemel 1991; Vallet 1999), and are in line with the salience of the cultural dimensions of class inequalities in France (Bourdieu 1984).

13

Because municipality fixed-effects models require multiple individual observations within the same municipality, the models are restricted to individuals living in municipalities in which at least another observation is available. This results in a small reduction of the sample size.

14

For concision, the age-restricted models are not included but are available from the author upon request.

15

The Paris region (Ile-de-France) consists of municipalities in eight departments: Paris, Essonne, Hauts-de-Seine, Seine-Saint-Denis, Seine-et-Marne, Val-de-Marne, Val-d’Oise, and Yvelines.

16

For concision, the interaction models are not included but are available from the author upon request.

17

Of a total 12,387 unique IRIS codes of individuals observed during childhood, 1,781 IRIS were observed again for individuals during adulthood. The individuals living in these IRIS form the sample of nonmovers.

References

Aalbers, M. B. (
2005
).
Place-based social exclusion: Redlining in the Netherlands
.
Area
,
37
,
100
109
. 10.1111/j.1475-4762.2005.00609.x
Aeberhardt, R., Fougère, D., Pouget, J., & Rathelot, R. (
2010
).
Wages and employment of second-generation immigrants in France
.
Journal of Population Economics
,
23
,
881
905
. 10.1007/s00148-009-0266-3
Aeberhardt, R., Rathelot, R., & Safi, M. (
2015
).
Les difficultés scolaires et professionnelles des jeunes issus de l’immigration: Effet de l’origine ou effets géographiques? [Academic and professional difficulties of young people from immigrant backgrounds: Effect of origin or geographical effects?]
.
Population
,
70
,
599
635
.
Alba, R., & Logan, J. (
1993
).
Minority proximity to whites in suburbs: An individual-level analysis of segregation
.
American Journal of Sociology
,
98
,
1388
1427
. 10.1086/230193
Alba, R., & Nee, V. (
2003
).
Remaking the American mainstream: Assimilation and contemporary immigration
.
Cambridge, MA
:
Harvard University Press
.
Barou, J. (
2005
).
HLM, le risqué d’une homogénéisation par le bas: Les effets des ségrégations [Low-rent housing in France (HLM): Could the lowest become the standard? The effects of segregation]
.
CNAF/Informations Sociales
,
123
(
3
),
74
87
.
Bernardot, M. (
1999
).
Chronique d’une institution: La Sonacotra (1956–1976) [Chronicle of an institution: Sonacotra (1956–1976)]
.
Sociétés Contemporaines
,
33–34
,
39
58
. 10.3406/socco.1999.1750
Blau, P. M., & Duncan, O. D. (
1967
).
The American occupational structure
.
New York, NY
:
John Wiley & Sons
.
Bolt, G., Van Kempen, R., & Van Ham, M. (
2008
).
Minority ethnic groups in the Dutch housing market: Spatial segregation, relocation dynamics and housing policy
.
Urban Studies
,
45
,
1359
1384
. 10.1177/0042098008090678
Bonnet, F., Lalé, E., Safi, M., & Wasmer, E. (
2016
).
Better residential than ethnic discrimination! Reconciling audit and interview findings in the Parisian housing market
.
Urban Studies
,
53
,
2815
2833
. 10.1177/0042098015596107
Bourdieu, P. (
1984
).
Distinction: A social critique of the judgement of taste
.
Cambridge, MA
:
Harvard University Press
.
Bourgeois, M. (
2013
).
Choisir les locataires du parc social? Une approche ethnographique de la gestion des hlm [Choose the tenants of the social park? An ethnographic approach to HLM management]
.
Sociologie du Travail
,
55
,
56
75
. 10.1016/j.soctra.2012.12.002
Bråmå, A. (
2006
).
“White flight”? The production and reproduction of immigrant concentration areas in Swedish cities, 1990–2000
.
Urban Studies
,
43
,
1127
1146
. 10.1080/00420980500406736
Bunel, M., L’Horty, Y., Du Parquet, L., & Petit, P. (
2017
).
Les discriminations dans l’accès au logement à Paris: Une expérience contrôlée
[Discrimination in access to housing in Paris: A controlled experience].
Paris, France
:
TEPP
.
Causa, O., Dantan, S., & Johansson, A. (
2009
).
Intergenerational social mobility in European OECD countries
(OECD Economics Department Working Papers, No. 709).
Paris, France
:
OECD Publishing
.
Charles, C. Z. (
2003
).
The dynamics of racial residential segregation
.
Annual Review of Sociology
,
29
,
167
207
. 10.1146/annurev.soc.29.010202.100002
Chetty, R., Grusky, D., Hell, M., Hendren, N., Manduca, R., & Narang, J. (
2016
).
The fading American dream: Trends in absolute income mobility since 1940
(NBER Working Paper Series, No. 22910).
Cambridge, MA
:
National Bureau of Economic Research
.
Chetty, R., Hendren, N., & Katz, L. F. (
2016b
).
The effects of exposure to better neighborhoods on children: New evidence from the Moving to Opportunity Experiment
.
American Economic Review: Papers & Proceedings
,
106
,
855
902
.
Chetty, R., Hendren, N., Kline, P., & Saez, E. (
2014
).
Where is the land of opportunity? The geography of intergenerational mobility in the United States
.
Quarterly Journal of Economics
,
129
,
1553
1623
. 10.1093/qje/qju022
Chetty, R., Hendren, N., Kline, P., Saez, E., & Turner, N. (
2014
).
Is the United States still a land of opportunity? Recent trends in intergenerational mobility
.
American Economic Review
,
104
,
141
147
. 10.1257/aer.104.5.141
Crowder, K. (
2000
).
The racial context of white mobility: An individual-level assessment of the white flight hypothesis
.
Social Science Research
,
29
,
223
257
. 10.1006/ssre.1999.0668
Crowder, K., Hall, M., & Tolnay, S. E. (
2011
).
Neighborhood immigration and native out-migration
.
American Sociological Review
,
76
,
25
47
. 10.1177/0003122410396197
Crowder, K., Pais, J., & South, S. J. (
2012
).
Neighborhood diversity, metropolitan constraints, and household migration
.
American Sociological Review
,
77
,
325
353
. 10.1177/0003122412441791
Dietrich-Ragon, P. (
2011
).
Le logement intolérable
[Intolerable housing].
Paris, France
:
PUF
.
Diez Roux, A. V. (
2001
).
Investigating neighborhood and area effects on health
.
American Journal of Public Health
,
91
,
1783
1789
. 10.2105/AJPH.91.11.1783
Ellen, I. G., & Turner, M. A. (
1997
).
Does neighborhood matter? Assessing recent evidence
.
Housing Policy Debate
,
8
,
833
866
. 10.1080/10511482.1997.9521280
Felouzis, G. (
2003
).
La ségrégation ethnique au collège et ses conséquences [Ethnic segregation in middle school and its consequences]
.
Revue Française de Sociologie
,
44
,
413
447
.
Finney, N., & Simpson, L. (
2009
).
“Sleepwalking to segregation”?: Challenging myths about race and migration
.
Bristol, UK
:
Policy Press
.
Fougère, D., Kramarz, F., Rathelot, R., & Safi, M. (
2013
).
Social housing and location choices of immigrants in France
.
International Journal of Manpower
,
34
,
56
69
. 10.1108/01437721311319656
Gobillon, L., Magnac, T., & Selod, H. (
2011
).
The effect of location on finding a job in the Paris region
.
Journal of Applied Econometrics
,
26
,
1079
1112
. 10.1002/jae.1168
Gobillon, L., & Solignac, M. (
2015
).
Homeownership of immigrants in France: Selection effects related to international migration flows
(IZA Discussion Paper, No. 9517).
Bonn, Germany
:
Institute for the Study of Labor
.
Goffette-Nagot, F., & Sidibé, M. (
2016
).
Housing wealth accumulation: The role of public housing
.
Regional Science and Urban Economics
,
57
,
12
22
. 10.1016/j.regsciurbeco.2015.11.004
Grusky, D. (
2014
).
Social stratification: Class, race, and gender in sociological perspective
. (4th ed.).
Boulder, CO
:
Westview Press
.
Haute Autorité de Lutte Contre les Discriminations et Pour l’Égalité (HALDE)
. (
2006
).
La discrimination dans l’acces au logement locatif prive
[Discrimination in access to private rental housing].
Paris, France
:
ASDO Études
.
Iceland, J. (
2009
).
Where we live now: Immigration and race in the United States
.
Oakland
:
University of California Press
.
Jacquemet, N. (
2013
).
Discriminations à l’embauche: Quelle ampleur, quelles solutions? [Discrimination in hiring: How much, what solutions?]
.
Regards Croisées sur l’Économie
,
1
(
13
),
49
63
. 10.3917/rce.013.0049
Keels, M., Duncan, G. J., DeLuca, S., Mendenhall, R., & Rosenbaum, J. (
2005
).
Fifteen years later: Can residential mobility programs provide a long-term escape from neighborhood segregation, crime, and poverty
.
Demography
,
42
,
51
73
. 10.1353/dem.2005.0005
Lagrange, H., & Oberti, M. (
2006
).
Emeutes urbaines et protestations
[Urban riots and protests].
Paris, France
:
Presses de Sciences Po
.
Lemel, Y. (
1991
).
Stratification et mobilité sociale
[Stratification and social mobility].
Paris, France
:
Armand Colin
.
Leventhal, T., & Brooks-Gunn, J. (
2003
).
Moving to opportunity: An experimental study of neighborhood effects on mental health
.
American Journal of Public Health
,
93
,
1576
1582
. 10.2105/AJPH.93.9.1576
Logan, J. R., Zhang, W., & Alba, R. D. (
2002
).
Immigrant enclaves and ethnic communities in New York and Los Angeles
.
American Sociological Review
,
67
,
299
322
. 10.2307/3088897
Ludwig, J., Liebman, J. B., Kling, J. R., Duncan, G. J., Katz, L. F., Kessler, R. C., & Sanbonmatsu, L. (
2008
).
What can we learn about neighborhood effects from the Moving to Opportunity Experiment?
.
American Journal of Sociology
,
114
,
144
188
. 10.1086/588741
Massey, D. S. (
2007
).
Categorically unequal: The American stratification system
.
New York, NY
:
Russell Sage Foundation
.
Massey, D. S., & Denton, N. A. (
1985
).
Spatial assimilation as a socioeconomic outcome
.
American Sociological Review
,
50
,
94
106
. 10.2307/2095343
Massey, D. S., & Denton, N. A. (
1993
).
American Apartheid: Segregation and the making of the underclass
.
Cambridge, MA
:
Harvard University Press
.
Mayer, N., Michelat, G., Tiberj, V., & Vitale, T. (
2014
).
Le regard des chercheurs: Un refus croissant de “l'autre”
[The perspective of researchers: A growing refusal of the other]. In
Commisision nationale consultative des droits de l’homme
(Ed.),
La lutte contre le racisme, l’antisémitisme et la xénophobie. Année 2013
(pp.
157
208
).
Paris
:
La Documentation française
.
Mayer, S. E., & Jencks, C. (
1989
).
Growing up in poor neighborhoods: How much does it matter?
.
Science
,
243
,
1441
1445
. 10.1126/science.243.4897.1441
Merle, P. (
2011
).
La carte scolaire et son assouplissement. Politique de mixité sociale ou de ghettoïsation des établissements? [The school map and its relaxation. Social diversity or ghettoization of schools?]
.
Sociologie
,
2
,
37
50
. 10.3917/socio.021.0037
Meurs, D., Pailhé, A., & Simon, P. (
2006
).
The persistence of intergenerational inequalities linked to immigration: Labour market outcomes for immigrants and their descendants in France
.
Population (English ed.)
,
61
,
645
682
. 10.3917/pope.605.0645
Oberti, M. (
2007
).
L’école dans la ville: Ségrégation-mixité-carte scolaire
[School in the city: Segregation-mix-school map].
Paris, France
:
Presses de Sciences Po
.
Oberti, M., Préteceille, E., & Rivière, C. (
2012
).
Les effets de l’assouplissement de la carte scolaire dans la banlieue parisienne
[The effects of the relaxation of the school map in the suburbs of Paris] (Report for the HALDE).
Paris, France
:
Défenseur des Droits and the DEPP/French Ministry of Education
.
Pais, J., South, S. J., & Crowder, K. (
2012
).
Metropolitan heterogeneity and minority neighborhood attainment: Spatial assimilation or place stratification?
.
Social Problems
,
59
,
258
281
. 10.1525/sp.2012.59.2.258
Pais, J. F., South, S. J., & Crowder, K. (
2009
).
White flight revisited: A multiethnic perspective on neighborhood out-migration
.
Population Research and Policy Review
,
28
,
321
346
. 10.1007/s11113-008-9101-x
Pan Ké Shon, J-L (
2010
).
The ambivalent nature of ethnic segregation in France’s disadvantaged neighbourhoods
.
Urban Studies
,
47
,
1603
1623
. 10.1177/0042098009356123
Pan Ké Shon, J-L, & Verdugo, G. (
2015
).
Forty years of immigrant segregation in France, 1968–2007: How different is the new immigration?
.
Urban Studies
,
52
,
823
840
. 10.1177/0042098014529343
Park, R. E., & Burgess, E. W. (
1921
).
Introduction to the science of sociology
.
Chicago, IL
:
University of Chicago Press
.
Peach, C. (
2009
).
Slippery segregation: Discovering or manufacturing ghettos?
.
Journal of Ethnic and Migration Studies
,
35
,
1381
1395
. 10.1080/13691830903125885
Portes, A., & Zhou, M. (
1993
).
The new second generation: Segmented assimilation and its variants
.
Annals of the American Academy of Political and Social Science
,
530
,
74
96
. 10.1177/0002716293530001006
Préteceille, E. (
2009
).
La segregation ethno-raciale a-t-elle augmenté dans la métropole parisienne? [Has ethnoracial segregation increased in the Paris metropolis?]
.
Revue Française de Sociologie
,
50
,
489
519
. 10.3917/rfs.503.0489
Quillian, L. (
2002
).
Why is black-white residential segregation so persistent? Evidence on three theories from migration data
.
Social Science Research
,
31
,
197
229
. 10.1006/ssre.2001.0726
Rathelot, R. (
2014
).
Ethnic differentials on the labor market in the presence of asymmetric spatial sorting: Set identification and estimation
.
Regional Science and Urban Economics
,
48
,
154
167
. 10.1016/j.regsciurbeco.2014.06.007
Rathelot, R., & Safi, M. (
2014
).
Local ethnic composition and natives’ and immigrants’ geographic mobility in France, 1982–1999
.
American Sociological Review
,
79
,
17
42
. 10.1177/0003122413514750
Safi, M. (
2006
).
Le processus d’intégration des immigrés en France: Inégalités et segmentation [The process of integration for immigrants in France: Inequalities and segmentation]
.
Revue Française de Sociologie
,
47
,
3
48
. 10.3917/rfs.471.0003
Safi, M. (
2009
).
La dimension spatiale de l’intégration: Évolution de la ségrégation des populations immigrées en France entre 1968 et 1999 [The spatial dimension of integration: Evolution of the segregation of immigrant populations in France between 1968 and 1999]
.
Revue Française de Sociologie
,
50
,
521
552
. 10.3917/rfs.503.0521
Safi, M. (
2013
).
Les inégalités ethnoraciales
[Ethnoracial inequalities].
Paris, France
:
La Découverte
.
Safi, M., & Simon, P. (
2013
).
Les discriminations ethniques et raciales dans l’enquête trajectoires et origines: Représentations, expériences subjectives et situations vécues [Ethnic and racial discrimination in the survey trajectories and origins: Representations, subjective experiences and lived situations]
.
Economie et Statistique
,
464
,
245
275
. 10.3406/estat.2013.10240
Sala Pala, V. (
2013
).
Discriminations ethniques: Les politiques du logement social en France et au Royaume-Uni
[Ethnic discrimination: Social housing policies in France and the United Kingdom].
Rennes, France
:
Presses Universitaires de Rennes
.
Sampson, R. J. (
2012
).
Great American city: Chicago and the enduring neighborhood effect
.
Chicago, IL
:
University of Chicago Press
.
Sampson, R. J., Sharkey, P., & Raudenbush, S. W. (
2008
).
Durable effects of concentrated disadvantage on verbal ability among African-American children
.
Proceedings of the National Academy of Sciences
,
105
,
845
852
. 10.1073/pnas.0710189104
Sanbonmatsu, L., Kling, J. R., Duncan, G. J., & Brooks-Gunn, J. (
2006
).
Neighborhoods and academic achievement results from the Moving to Opportunity Experiment
.
Journal of Human Resources
,
41
,
649
691
. 10.3368/jhr.XLI.4.649
Sharkey, P. (
2008
).
The intergenerational transmission of context
.
American Journal of Sociology
,
113
,
931
969
. 10.1086/522804
Sharkey, P. (
2013
).
Stuck in place: Urban neighborhoods and the end of progress toward racial equality
.
Chicago, IL
:
University of Chicago Press
.
Sharkey, P., & Elwert, F. (
2011
).
The legacy of disadvantage: Multigenerational neighborhood effects on cognitive ability
.
American Journal of Sociology
,
116
,
1934
1981
. 10.1086/660009
Sharkey, P., & Faber, J. W. (
2014
).
Where, when, why, and for whom do residential contexts matter? Moving away from the dichotomous understanding of neighborhood effects
.
Annual Review of Sociology
,
40
,
559
579
. 10.1146/annurev-soc-071913-043350
Silberman, R., Alba, R., & Fournier, I. (
2007
).
Segmented assimilation in France? Discrimination in the labour market against the second generation
.
Ethnic and Racial Studies
,
30
,
1
27
. 10.1080/01419870601006488
Simon, P. (
2003
).
France and the unknown second generation: Preliminary results on social mobility
.
International Migration Review
,
37
,
1091
1119
. 10.1111/j.1747-7379.2003.tb00171.x
Simon, P. (
2008
).
The choice of ignorance: The debate on ethnic and racial statistics in France
.
French Politics, Culture & Society
,
26
(
1
),
7
31
. 10.3167/fpcs.2008.260102
Simon, P., Kirszbaum, T., & Chafi, M. (
2001
).
Les discriminations raciales et ethniques dans l’accès au logement social
[Racial and ethnic discrimination in access to social housing] (GELD Public report).
Paris, France
:
Groupe d’Etude et de Lutte contre les Discriminations
.
South, S. J., Crowder, K., & Chavez, E. (
2005
).
Exiting and entering high-poverty neighborhoods: Latinos, Blacks and Anglos compared
.
Social Forces
,
84
,
873
900
. 10.1353/sof.2006.0037
South, S. J., Crowder, K., & Pais, J. (
2008
).
Inter-neighborhood migration and spatial assimilation in a multi-ethnic world: Comparing Latinos, Blacks and Anglos
.
Social Forces
,
87
,
415
443
. 10.1353/sof.0.0116
Toma, S. (
2016
).
The role of migrant networks in the labour market outcomes of Senegalese men: How destination contexts matter
.
Ethnic and Racial Studies
,
39
,
593
613
. 10.1080/01419870.2015.1078480
Vallet, L. A. (
1999
).
Quarante années de mobilité sociale en France: L’évolution de la fluidité sociale a la lumière de modèles récents [Forty years of social mobility in France: The evolution of social fluidity in the light of recent models]
.
Revue Française de Sociologie
,
40
,
5
64
. 10.2307/3322517
Van Ham, M., & Clark, W. A. V. (
2009
).
Neighbourhood mobility in context: Household moves and changing neighbourhoods in the Netherlands
.
Environment and Planning A
,
41
,
1442
1459
.
Van Ham, M., Hedman, L., Manley, D., Coulter, R., & Östh, J. (
2014
).
Intergenerational transmission of neighbourhood poverty: An analysis of neighbourhood histories of individuals
.
Transactions of the Institute of British Geographers
,
39
,
402
417
. 10.1111/tran.12040
van Zanten, A. (
2001
).
L’école de la périphérie: Scolarité et ségrégation en banlieue
[The school of the periphery: Schooling and segregation in the suburbs].
Paris, France
:
PUF
.
Vartanian, T. P., Walker Buck, P., & Gleason, P. (
2007
).
Intergenerational neighborhood-type mobility: Examining differences between blacks and whites
.
Housing Studies
,
22
,
833
856
. 10.1080/02673030701474792
Verdugo, G. (
2011
).
Public housing and residential segregation of immigrants in France, 1968–1999
.
Population
(English ed.),
66
,
169
193
.
Wacquant, L. (
1992
).
Banlieues Françaises et ghetto noir Américain: De l’amalgame à la comparaison [French suburbs and American black ghetto: From amalgam to comparison]
.
French Politics and Society
,
10
(
4
),
81
103
.
Wacquant, L. (
2008
).
Urban outcasts: A comparative sociology of advanced marginality
.
Cambridge, MA
:
Polity Press
.
Wilson, W. J. (
1987
).
The truly disadvantaged: The inner city, the underclass, and public policy
.
Chicago, IL
:
University of Chicago Press
.
Xie, Y., & Gough, M. (
2011
).
Ethnic enclaves and the earnings of immigrants
.
Demography
,
48
,
1293
1315
. 10.1007/s13524-011-0058-8