Ethnic discrimination in social housing

par François Bonnet & traduit par Oliver Waine [12-03-2014]

Mots-clés : housing | racism | social housing | discrimination | minorities | race

Valérie Sala Pala has investigated the extent to which ethnic criteria come into play in the allocation of social housing in Marseille in France and in Birmingham in England. In her recent book, she presents a detailed analysis of discriminatory processes that questions the usefulness of the notion of racism.

Valérie Sala Pala’s book considers the way in which ethnic criteria are employed in the allocation of social housing in Marseille (in France, the nation of republican universalism) and Birmingham (in multiculturalist Britain). It adopts a dual perspective combining the sociology of ethnic discrimination and an analysis of public action. The author’s research is based on 50 or so interviews with people involved in the social-housing sectors in each city. Housing (social or otherwise) is a central part of all our lives, whether in terms of the cost it represents in family budgets, the personal emotional investment we make in our homes, or the way it structures space. It is therefore a fertile field for the study of ethnic issues. The approach adopted by the author is comprehensive : she analyses the culture, routines and rules produced by institutions (p. 70). Her aim is to show how public action helps create ethnic boundaries. To this end, she studied both employees of social-housing bodies who are in contact with users (“street-level bureaucrats” [1]) and the strategies implemented by institutions – in other words, public policy.

A comparative approach to the co-production of discrimination

Talking about ethnic discrimination means carefully describing certain concepts and ridding them of unwanted conventional connotations. To achieve this, Valérie Sala Pala makes extensive use of contributions from English-speaking – and more specifically British – academic literature. The challenge here is to avoid reducing racism to individual animosity, hence the concept of “institutional racism”. The author concludes a detailed and well-informed debate with two definitions : she defines racism as all representations that essentialise and inferiorise a particular group, and discrimination as practices that may or not be racist. Theoretical debates on racism are so labyrinthine that any theoretical or definitional choice can be criticised from a number of angles. For example, American sociologists such as Devah Pager and Lincoln Quillian are cautious in linking discrimination and racism. For these authors, discrimination involves observable, measurable, quantifiable, tangible practices that can be used from a scientific viewpoint. Racism, on the other hand, is such a nebulous word that its only function is to condemn (or to distance oneself from certain beliefs and behaviours). Here, Valérie Sala Pala has made the (perfectly legitimate) opposite choice, and constructed her research object in such a way as to determine whether the allocation of social housing is, ultimately, “racist” or not.

Her interviewees can breathe easy : the answer is no, in both Birmingham and Marseille. But – and this is the key argument of the book – the fact that institutions are not inherently racist does not mean that they do not discriminate heavily in terms of ethnicity.

In Birmingham, the allocation of social housing on ethnic criteria is commonplace and justified by the residential choices of households : Asians (i.e. from the Indian subcontinent) are supposed to want to live together, as are Afro-Caribbeans. Residential segregation results from an aggregation of individual choices, thereby legitimising the fact that minorities of colour tend to be relegated to the least desirable housing, in a context of mass privatisation of social-housing stock. Social housing represented 33% of all housing stock in 1980, compared with 18% in 2009.

In France, the importance accorded to “social diversity” and the stigma associated with the use of ethnic categories produce contradictory injunctions around which those involved in the “HLM” (habitations à loyer modéré – low-rent housing) sector must work. Social housing in France has, since the 1960s, moved away from an approach based on “necessity” (the need to build massively because of a rapidly growing population unable to find housing) and towards an approach based on the “institutionalisation of diversity” (p. 115). But what kind of diversity are we talking about here ? In theory, strictly socio-economic diversity, i.e. to avoid the concentration of poverty. In practice, though, it is a combination of economic and ethnic criteria, in order to avoid the development of “ghettos”. The use of ethnic criteria, however, is altogether unofficial, since the French Republic officially proscribes the use of any ethnic categories. In reality, those involved in social housing manage to identify Comorans by their surnames, or alternatively by their social security numbers (which indicate the département or country of birth), their dates of birth (more specifically, those that are incomplete), or the quality of their written communication in French on the forms they fill in (p. 140). The use of ethnic categories is viewed as necessary by employees of social-housing bodies in Marseille to enable “fine-tuned housing allocations”, such as not housing a large family above an elderly couple, not mixing “Arabs” and “Gypsies”, and so forth. The stated aim of such unofficial policies is to minimise conflict between neighbours. For Sala Pala, the case of “Arabs” and “Gypsies” is a typical example of “racialisation” without there necessarily being any “racist” intent, since there is no question of suggesting the inherent inferiority of a group with regard to another.

Valérie Sala Pala is interested in the way public action is interpreted and implemented locally, and the effects of local systems of stakeholders. Her analysis takes into consideration the construction of action frameworks and implementation constraints, as well as institutional strategies : in particular, the ethnicised accommodation strategies of social landlords, but also urban policies. The revelation of what takes place within networks of local stakeholders (including state bodies, municipal bodies and other organisations) and within the institutions responsible for defining various strategies (e.g. accommodation strategies for social-housing bodies, housing strategies for local authorities, anti-discrimination strategies, etc.) is central to understanding the co-production of discrimination, which extends far beyond the practices or representations of social-housing employees, which form another “stratum” of analysis. The approach adopted therefore does not merely concern individual routines or individual stereotypes.

Discrimination without racism ?

The key merit of Discriminations ethniques lies in the fact that it has a core argument, and that it makes the case for this argument. The conclusion is twofold : the various bodies and individuals involved “do not need to be racist in order to discriminate” (p. 180), but their daily routines ultimately produce ethnic borders. It is with the very best of intentions that those who work in social housing categorise, sort, separate and allocate applications because the candidates are “white”, “Arab”, “Gypsy” or “Comoran”. In short, we must not allow the tree of racism to obscure the wood of discrimination, as Patrick Simon puts it (2006). Valérie Sala Pala’s aim is not to minimise racism, but to use a precise vocabulary to show how, insidiously, discrimination takes place without the employees themselves being racist.

In this context, how can ethnic discrimination be combated ? Valérie Sala Pala looks at both institutional efforts and examples of mobilisation by residents. In Birmingham, the “enchanted discourse” on cultural difference resulting from multiculturalism leads to a euphemisation of discrimination, and relative institutional apathy. In Marseille, republican universalism forces stakeholders to reduce specific ethnic problems (such as racial discrimination) to problems of poverty, which also leads – for different reasons – to a euphemisation of discrimination. Universalism does, however, have one “advantage” over multiculturalism : it can be used to refute any argument citing “communitarianism” that might be put forward by ethnic minorities that dare to take action (for better housing, against racism, etc.). Overall, Discriminations ethniques is an accomplished work : the comparative approach, the mobilisation of English-language literature and an impartial, unemotional outlook make for a frank, serious work that will be useful for anyone interested in ethnic issues, social housing, public policy or comparative analysis.

Was Valérie Sala Pala right or wrong not to describe her interviewees as racist ? In my view, this question is not especially relevant, for the following reasons. First, racism is not a concept that lends itself to an unequivocal definition, as it is a “practical” category (a notion in common parlance), as opposed to an analytical category (a scientific concept). “Life expectancy at birth”, for example, is not a concept that changes its meaning according to the speaker. Similarly, “symbolic violence” or “areas of uncertainty” have relatively stable definitions in the field of French sociology. “Racism”, on the other hand, is without doubt a concept whose meaning is disputed in common parlance. As Albert Bastenier noted, racism has in Western democracies become a “secular version of Evil”, which “weighs heavy, with the weight of a deceptive moralism, on the comprehension that Europeans manage to develop regarding the nature of the social issues that they perceive within this theme”(Bastenier 2004). Nobody, therefore, wants to be racist ; and any accusation of racism is so morally loaded that it cannot be distanced by any scientific definition of racism.

Valérie Sala Pala defines racism narrowly, precisely to avoid this label being applied indiscriminately, and concludes that her interviewees are not, according to her definition, racist. In my view, Sala Pala was wrong, from the outset, to take the notion of racism seriously : in her review of the literature examined, she should have stopped at Wacquant (1997), decided that the notion of racism is unusable, and proceeded to build an alternative analytical framework. As she did not make this choice, she is now exposed to moral criticism. By exempting her respondents from the ignominious accusation of racism, even though the interview extracts cited are full of references to “Arab” tenants who are less desirable than “Armenian” tenants, Valérie Sala Pala exposes herself to the accusation of “euphemising” and “covering up” the racism of her respondents. Such accusations fall squarely into the category of judgements based on presumed intent, but it is nonetheless inevitable, as anything related to racism provokes a normative response, such that only stances (“White people are necessarily racist”/“Racism is an invention of political correctness”) are pure enough to at least escape reproach.

Another criticism that can be levelled at this work is that the author has produced a book of 300 or so dense, well-informed pages, with a remarkable analysis of the literature on the subject, and yet manages not to discuss the question of pure discrimination versus statistical discrimination. In a nutshell, this theory (originating in the field of economics) makes a distinction between irrational discrimination (“I reject the minority applicant, regardless of his or her characteristics, through pure hostility towards the minority in question”) and probabilistic discrimination (“whether for the right or wrong reasons, I think that members of minority X have a greater probability of exhibiting a particular characteristic that is rationally undesirable, and so I discriminate against minority X”). For example, if I think men have a particular propensity for violence against children, I will take steps to prevent this risk by refusing to hire men to work in a nursery, even though I know that all men are not violent. We are constantly making such statistical discriminations, with varying levels of (in)accuracy and (mis)information, and it would appear the employees of social-housing bodies in Discriminations ethniques are no different.

The distinction between pure discrimination and statistical discrimination is significant in relation to the cognitive interpretation made by Valérie Sala Pala. Here, “cognitive interpretation” involves identifying the categories in terms of which people think and act. This challenging exercise is brilliantly executed, and readers will delight in the chapters of the second part of the book, particularly those set in Marseille, which are worth the cover price alone. However, once these cognitions have been described, they need to be explained. Where do they originate ? Are they the same from one organisation to another ? Do they reflect stereotypes that are present in the media or in the professional literature ? Ultimately, are these cognitions more or less in line with the empirical reality, or are they somewhat left-field ? Why are ethnic categories so pervasively present among employees of social housing, and why do they not use other categories that they might consider relevant ?

These questions arise for the following reason : the individuals who work in social-housing bodies in Marseille resort to ethnic stereotypes in order to fine-tune allocations in order to avoid conflicts between neighbours and the deterioration of dwellings and buildings. If these stereotypes are false, then that would mean that the employees in question are grossly incompetent, and do literally whatever they feel like. This would be an interesting research result, both for the social sciences and for public policy. Conversely, if the stereotypes have some basis in reality, then these employees are just doing their job. Merely describing cognitions, as Valérie Sala Pala does here, can only be the first step in truly understanding what is at stake in the use of ethnic stereotypes. For example, according to one person in the social-housing sector (p. 166), the problem with Comorans is that they will happily play host to family and friends for any length of time, to the extent that a dwelling allocated to a family of four is often very quickly inhabited by 10 people. North Africans, on the other hand, do not tend to share their apartments. Are these characteristics true, marginally true, or completely false ? Another example is the supposed antagonism between “Arabs” and “Gypsies” : racist nonsense or telling stereotype ?

Obviously, just because a stereotype has a basis in reality does not make it legitimate : comprehension is not the same thing as justification. It is clear that very large families have a higher probability of causing noise or other disruption for other residents, and it is also clear that we must not prevent very large families from accessing social housing. Above all, though, stereotypes can be wrong and self-fulfilling : it is all too easy to imagine how conventional “wisdom” of the kind cited above is shared around the coffee machine or water cooler. In short, we would have liked to see this issue explored in greater detail. Nevertheless, Discriminations ethniques is a work to be highly recommended to anyone interested in the city and in contemporary Britain and France.


  • Bastenier, Albert. 2004. Qu’est-ce qu’une société ethnique ? Ethnicité et racisme dans les sociétés européennes d’immigration, Paris : Presses Universitaires de France.
  • Lipsky, Michael. 1980. Street-Level Bureaucracy : Dilemmas of the Individual in Public Services, New York : Russell Sage Foundation.
  • Simon, Patrick. 2006. “L’arbre du racisme et la forêt des discriminations”, in Guénif-Souilamas, Nacira (ed.), La République mise à nu par son immigration, Paris : La Fabrique, pp. 160–177.
  • Wacquant, Loïc. 1997. “For an analytic of racial domination”, Political Power and Social Theory, vol. 11, pp. 221–234.


[1This term is borrowed from a classic work on the implementation of public policy (Lipsky 1980).

Pour citer cet article :

François Bonnet & traduit par Oliver Waine, « Ethnic discrimination in social housing », Métropolitiques, 12 mars 2014. URL :
© - Toute reproduction interdite sans autorisation explicite de la rédaction