Abstract
This paper documents the existence of sorting patterns based on region of origin in the French public housing market (HLM). We provide evidence that first-generation African immigrants who benefit from the program tend to live in poorer neighborhoods than their French-born counterparts, once controlled for the broader geographical area of residence. This differential is comparable in magnitude to the one observed in the private rental market, even though the pricing of a public rental is almost uncorrelated with its location characteristics. Whereas we cannot rule out the possibility that this sorting partly reflects the specificity of African immigrants’ housing demand, we mitigate this concern by using information on the HLM offers that were previously rejected by the households and on the households’ declared level of satisfaction regarding their current neighborhood. Given that about half of African immigrants live in HLM (against one sixth of the non-immigrant population), this finding has possibly strong consequences on African immigrants’ segregation and welfare in France.
This is a preview of subscription content, log in via an institution to check access.
Similar content being viewed by others
Notes
This is the point made by Glaeser (2003, p. 187): “In the free market, an apartment is allocated to an individual if his desire for the apartment exceeds the price of the apartment. This means that poorer residents who are willing to sacrifice enough financially can in principle choose to live in a richer community in order to get more successful peers for themselves or their children. […] In a shortage economy, the allocation of apartments gets far less straightforward. Desire for the apartment plays some role, but since there are more people who want apartments (at that price) than get them, other variables come in, including nepotism, favoritism and so forth.”
In a general equilibrium framework with the quality of education as an endogenously determined local public good, Leung et al. (2012) show that housing vouchers are strictly preferred to public housing units for their consequences on the labor supply, as well as on the quality of education and the level of private rents. In a recent summary of the empirical studies, Olsen (2008) writes that “the most important finding of the empirical literature on the effects of different housing programs from the viewpoint of housing policy is that recipient-based housing assistance has provided equally good housing at a much lower total cost than any type of unit-based assistance.”
In 2000, this belief gave ground to a new legislation: the SRU law (Solidarité et Renouvellement Urbain, or Solidarity and Urban Renewal). The SRU law stipulates that, to avoid financial sanctions, a large majority of municipalities must progressively increase their HLM supply up to 20 %. Such disposition stemmed from a compromise between feasibility (municipalities are the smallest administrative units with executive power) and the idea that, the most relevant scale to look at segregation being the metropolitan area, ensuring equal distribution between cities of a common metropolitan area was a reasonable first-order approximation to reduce segregation. While the evaluation of the compliance of municipalities has recently begun (Bono and Trannoy 2011), the final impact of SRU on segregation will also crucially depend on the condition that these new public housing units are allocated to the relevant social groups, which was only assumed by the lawmaker (Schmutz 2013).
See Aslund et al. (2010) for one of the very few examples of another institutional feature: the compulsory scattering of incoming refugees in Sweden.
One way to overcome this issue is to look at administrative data, such as waiting lists and requests to move (Bird 1976). Centralized administrative data has recently started being gathered in France and will hopefully soon become available to researchers.
Bonnal et al. (2012) make a step in the same direction. They use the ENL to compare average waiting times between non-European HLM applicants and European applicants and show that the former experience longer waiting times, even after controlling for observable characteristics. Given the declarative nature of the ENL, information on waiting times is not fully reliable and also subject to various kinds of censoring. However, their result is an additional indication of the likelihood of discriminatory practices against non-European immigrants by public housing agencies.
While the program dates back to the 1920s, most of the current stock was effectively built after 1958.
A French political tradition makes it controversial to collect racial or ethnic statistics. See Simon (2003) for details.
For example, our control group of non-immigrants will include people born in the former French colonies and who were given French citizenship at birth, French West Indians and, increasingly, second, third and even fourth-generation immigrants of African origin.
According to the 2007 Census, the share of HLMs at the district (“département”) level is distributed with a mean of 0.139, a standard deviation of 0.058, going from 0.047 to 0.359 and it is positively correlated with the total population in the district.
These neighborhoods, which were selected to be as homogeneous as possible in terms of population size, are formed by a reunion of census tracks in municipalities with a population greater than 10,000 inhabitants (45 % of cases), by municipalities themselves when their population is smaller than 10,000 inhabitants and greater than 5,000 inhabitants (19 %), and by an administrative reunion of municipalities (called “canton” or “arrondissement”) otherwise.
Other indicators of neighborhood quality could be used, that would lead to very similar results. See subsection “Robustness of the results” for details.
In the ENL, the respondent declares whether he or she faces a daily commute to work and if so, provides an estimate of the average duration of this commute. An ordinary-least-squares regression of the log of this duration on the log of the NQI with municipality fixed effects yields a significant estimate for elasticity of transportation duration with respect to neighborhood quality of−13 % on the sample of private tenants and of−16 % on the sample of HLM tenants.
Indeed, the mean of the log of the NQI is equal to−1.232 for African HLM tenants and to−1.278 for non-immigrant HLM tenants (the difference is significant at the 10 % confidence level) and this is even truer on the private rental market (−0.898 against−1.082, the difference being significant at the 1 % confidence level).
Départements are roughly equivalent to US districts. Employment areas are defined statistically as areas where most of the workforce both lives and works. Their delimitation follows regional boundaries and, most of the time, also the boundaries of the département.
Out of the fifteen unpaired t-tests conducted on the sample of tenants and comparing the household characteristics of African immigrants and non-immigrants, eleven yield statistically significant differences.
We thank a referee for pointing that point out.
Results, available upon request, are indeed very similar.
In practice, we set μ = 1, 000. Provided μ is large enough, results are not sensitive to this choice.
This program was launched in 1996. According to the lawmaker, ZUS are characterized by large-scale housing schemes, derelict habitat and a strong imbalance between local population and local employment. There are 717 ZUS in continental France. The 714 of them that were part of a municipality with more than 10,000 inhabitants gathered more than 4.25 millions inhabitants in 1999. As for the NQI, the ZUS indicator is not only a measure of low socioeconomic status. In 2004, 70 % of the housing stock in home-ownership that is targeted by a nationwide rehabilitation program against urban slums is located in the ZUS. The proportion of students who are already at least 2 years behind in sixth grade is 3 points higher in ZUS middle schools, whereas the graduation rate for the middle-school degree (“Brevet”) is 10 points lower (Onzus 2004).
This process of differential attrition yields a positive correlation between the seniority of public tenants and the quality of their dwelling. Glaeser (2003) alludes to this mechanism in a theoretical paper on rent control and it has been established repeatedly on the HLM market concerning the characteristics of the dwelling (Laferrère et al. 1999; Jacquot 2007; Laferrère 2008).
In fact, they are even sometimes significantly higher for African immigrants. We do not interpret this result here, which may be driven by a larger unobserved heterogeneity in the population of African immigrants.
Moreover, they may not address the possibility of lexicographic preferences, with the local share of immigrants as the first-order criterion, and the wealth of the neighborhood as a second-order criterion.
Indeed, the log of the NQI for HLM tenants who have refused at least one proposal before accepting the dwelling where they currently reside is equal to−1.301, against−1.220 for those who have not (the difference is significant at the 1 % confidence level) and the gap increases, with values−1.294 and−1.172, if one isolates refusals that were explicitly motivated by neighborhood quality (the difference is still significant at the 1 % confidence level).
The results are very similar if we isolate the refusals that were explicitly motivated by neighborhood quality.
An alternative theory for the reason why African immigrants less often refuse public housing offers could also be that they are less savvy about navigating the public housing allocation system because they have less experience about the French housing market. This theory cannot be fully ruled out: according to the ENL 2002, only 12 % of the African immigrants who were living outside of France 5 years before the survey and who are currently on the waiting list for a HLM have refused a previous offer, against 27 % of their counterparts who were already living in France in 1997. However, controlling for current household income is enough to make this difference disappear. In addition, this kind of explanation would imply that seniority in France should be negatively correlated with the level of segregation experienced by public housing tenants: Verdugo (2011), who studies this specific question, does not find any relationship of the kind.
References
Algan Y, C. Hémet and D. Laitin (2011): “Diversity and public goods in a republican setting,” mimeo, Sciences-Po.
Aslund O, Osth J, Zenou Y. How important is access to jobs? Old question– improved answer. J Econ Geogr. 2010;10(3):389–422.
Bird H. Residential mobility and preference patterns in the public sector of the housing market. Trans Inst Br Geographers New Series. 1976;1(1):20–33.
Black D, Smith J. How robust is the evidence on the effects of college quality? Evidence from matching”. J Econ. 2004;121:99–124.
Bonnal L, Boumahdi R, Favard P. Nonexpected discrimination: the case of social housing in France. Appl Econ Lett. 2012;19:1909–16.
Bono P-H, Trannoy A. Logement social: les quotas sont-ils utiles. Regards Croisés sur l’Econ. 2011;9:247–57.
Bouvard L, Combes P-P, Decreuse B, Laouenan M, Schmutz B, Trannoy A. Géographie du chômage des personnes d’origine africaine. Rev Fr d’Econ. 2009;23(3):4–106.
Combes P.-P., G. Duranton, L. Gobillon, D. Puga and S. Roux (2012): “The productivity advantages of large cities: distingushing agglomeration from firm selection”, Econometrica, 80(6), pp. 2543-
Combes P.-P., B. Decreuse B. Schmutz and A. Trannoy (2014): “Neighbor Discrimination,” Working papers, CEPR 9160 (updated).
Decreuse B, Schmutz B. Residential mobility and unemployment of African immigrants in France: A calibration approach. Ann Econ Stat. 2012;107–108:51–92.
Dujardin C. and F. Goffette-Nagot (2005): “Neighborhood effects, public housing and unemployment in France,” Working Papers 0505, GATE.
Dujardin C, Goffette-Nagot F. Does public housing occupancy increase unemployment? J Econ Geogr. 2009;9(6):823–51.
Dujardin C, Goffette-Nagot F. Neighborhood effects on unemployment?: A test à-la Altonji. Reg Sci Urban Econ. 2010;40(6):380–96.
Fack G. Are housing benefits an effective way to redistribute income? Evidence from a natural experiment in France. Labour Econ. 2006;13(6):747–71.
Glaeser EL. Does rent control reduce segregation. Swed Econ Policy Rev. 2003;10:179–202.
Gobillon L. and H. Selod (2007): “The effect of segregation and spatial mismatch on unemployment: evidence from France,” Working papers, CEPR
Goux D, Maurin E. Close neighbours matter: Neighbourhood effects on early performance at school. Econ J. 2007;117(523):1193–215.
Henderson J, Karn V. Race, classe and state housing: Inequality and the allocation of state housing. Cambridge University: Press; 1987.
Jacquot A. (2007): “L’occupation du parc HLM: éclairage à partir des Enquêtes Logement de l’INSEE,” Working papers, INSEE
Kintrea K, Pawson H. Part of the problem or part of the solution? Social housing allocation policies and social exclusion in Britain. J Soc Policy. 2003;31(4):643–67.
Laferrère A. (2008): “La paupérisation du parc HLM: essai de vision dynamique,” mimeo, INSEE
Laferrère A., and D. Leblanc (2006): “Housing policy: low-income households in France,” The Blackwell Companion to urban economics, pp. 159–178
Laferrère A, Leblanc D, Pigois R. Les effets de l’existence du parc HLM sur le profil de consommation des ménages. Econ Stat. 1999;328(8):37–60.
Leung CKY, Sarpca S, Yilmaz K. Public housing units vs. housing vouchers: Accessibility, local public goods, and welfare. J Hous Econ. 2012;21(4):310–21.
Mistral J. and V. Plagnol (2009): “Loger les classes moyennes : la demande, l’offre et l’équilibre du marché du logement,” Report, Conseil d’Analyse Economique
Nechyba T. J. (2006): Income and Peer Quality Sorting in Public and Private Schools, vol. 2 of Handbook of the Economics of Education, chap. 22, pp. 1327–1368.
Olsen E. O. (2008): “Low-income housing policy,” The new Palgrave dictionary of economics, Second Edition
Onzus (2004): Rapport de l’Observatoire National des Zones Urbaines Sensibles, 254pp.
Oreopoulos P. The Long-Run Consequences Of Living In A Poor Neighborhood. Q J Econ. 2003;118(4):1533–75.
Pan Ke Shon, J.-L. and G. Verdugo (2014): “Forty years of immigrant segregation in France, 1968-2007. How different is the new immigration?”, Urban Studies, forthcoming.
Patacchini E. and Y. Zenou (2007): “Intergenerational Education Transmission: Neighbor- hood Quality and/or Parents’ Involvement?,” Working papers, IZA
Preteceille E. La ségrégation ethno-raciale dans la métropole parisienne. Rev Fr Sociol. 2009;50(3):489–519.
Quigley JM, Raphael S. Is housing unaffordable? Why isn’t it more affordable. J Econ Perspect. 2004;18(1):191–214.
Safi M. La dimension spatiale de l'intégration des populations immigrées (1968-1999). Rev Fr Sociol. 2009;50(3):521–52.
Schmutz B. Public housing quotas and segregation. Urban Publ Econ Rev. 2013;18:68–94.
Simon P. Les sciences sociales françaises face aux catégories ethniques et raciales. Ann Démograp Hist. 2003;105(1):111–30.
Tabard N. and G. Martin-Houssart (2002): “Représentation socio-économique du terri- toire : typologie des quartiers et communes selon la profession et l’activité économique de leurs habitants : France métropolitaine, recensement de 1999,” Working papers, INSEE
Verdugo G. Public housing and residential segregation of immigrants in France (1968-1999). Population. 2011;66(1):171–96.
Verdugo G. (2014): “Social housing magnets,” mimeo
Author information
Authors and Affiliations
Corresponding author
Additional information
This paper has benefited from discussions with Pierre-Philippe Combes, Clément de Chaisemartin, Bruno Decreuse, Laurent Gobillon, Florence Goffette-Nagot, Rodney Green, Francis Kramarz, Anne Laferrère and Alain Trannoy. Special thanks to Edwin Leuven and Maxime Tô. Earlier versions were presented at GREQAM lunch Seminar (Marseilles), CREST LMI internal seminar (Paris), TEPP Conference (Evry) and APET Conference (Galway). Thorough comments from two referees greatly improved the paper. The usual caveat applies.
About this article
Cite this article
Schmutz, B. Spatial sorting of African Immigrants in the French Public Housing Market. Rev Black Polit Econ 42, 247–270 (2015). https://doi.org/10.1007/s12114-014-9205-y
Published:
Issue Date:
DOI: https://doi.org/10.1007/s12114-014-9205-y