Social mix and housing policy: Local effects of a misleading rhetoric. The case of Milan

Introduction: Policies (and shifting arguments) regarding social mix in Europe and North America

This article focuses on different implementations and implications of the social mix rhetoric in housing policies, with special attention on the case of Milan, Italy. ¹ Milan represents an interesting case study for two main reasons: first, there is a gap in the literature about social mix policies in Mediterranean countries and especially in Italy, while there are many investigations into the effects of social mixing in the Anglo-Saxon countries, as well as in northern and continental Europe (Bridge et al., 2012). Second, this case study outlines a new understanding of the use and the effects of social mix policies, enriching the existing body of analysis and classifications (Bolt, 2009; Harlander and Kuhn, 2012). After an overview of the literature accounting for how social mix has been fostered in several contexts, the paper focuses on the experience of Milan in order to investigate how social mix interventions are entering the policy agenda in a country characterised by continuing low levels of spatial segregation and by a Mediterranean welfare regime. These two features are important to understand the peculiarities of the Milan case study, because social mix policies have been traditionally fostered in countries where segregation on an ethnic or socio-economic basis represents an important characteristic of the spatial configuration of the city (i.e. in the USA), and/or the municipal or social housing stock represents a high percentage of the total housing stock (i.e. the Netherlands).

In North America, social mix interventions have been strongly connected with the phenomenon of residential segregation of low-income people (Arbaci, 2007) and the presumed risks connected to ‘neighbourhood effects’ limiting chances of social mobility for the resident population. On the other hand, in Europe, as widely investigated in a recent comprehensive work by Harlander and Kuhn (2012), the image of a compact and mixed city (both in terms of social groups and functions) is widely considered as a main and fundamental feature of the city, representing its consolidated urbanity, stratified identity and culture, as well as for overall conditions favouring social cohesion (Siebel, 2004). In fact, in this respect even the walls of the medieval city, while producing separation, also organised integration. Furthermore, the assumption that the socio-spatial organisation of cities is strongly influenced by urban planning and policies which reflect specific forms of government and conditions of citizenship is highly relevant. The consolidation of the bourgeois city at the end of the 19th century featured in fact a vivid debate between the supporters of forms of urban development which would organise the construction of homogeneous housing areas for the different social classes (as was the case in the Garden Cities or in cooperative housing developments) and those instead supporting the development of a compact city model along principles of density and intense mix of social groups and functions.

The first orientation was mainly widespread in Great Britain and the USA along with a concept of good ‘neighbourliness’ among inhabitants with a similar socio-economic profile and with an urban development pattern which was aimed at preserving the stability of real estate values. Continental Europe mainly featured the second mode. Dense urban developments at the turn of the century featured social mix within compact urban blocks such as in the cases of Paris, Vienna and Berlin (Teuteberg and Wischermann, 1985).

Further to this historical background, social mix has recently become increasingly popular in European cities’ housing policies, where the spatial concentration of recent immigrants in particular has started to be managed as a sort of ‘public emergency’, not only in terms of the risks connected to possible effects of social exclusion for people living in these areas, but also as a perceived threat to ‘social cohesion’ (Cento Bull, 2010; Cucca and Pologruto, 2011; Manzo, 2012; Musterd, 2003; Musterd and Andersson, 2005). Because desegregation/dispersal has been progressively regarded as a form or process of integration or even assimilation (Bolt et al., 2010; Murdie and Ghosh, 2010), several policies have been introduced to reduce or to prevent spatial segregation, with the aim of creating a so-called ‘social mix’ (Arbaci and Malheiros, 2010; van Kempen and Bolt, 2009).

Beside the issue of spatial segregation, a second relevant aspect is that in Europe these policies have tended to be fostered in contexts characterised by a high availability of municipal or social housing. In these countries social housing does not represent a residual segment of the housing stock, it also being a suitable option for the middle class.

The programmes to avoid the concentration of particular social groups are fostered through several measures often following three main overlapping directions (Cities for Local Integration Policy (CLIP), 2007): (1) to reduce or prevent spatial segregation through the diversification of housing (type of housing and tenure) in disadvantaged areas; (2) to contain the possible negative effects of spatial segregation through the definition of criteria and procedures governing the allocation of dwellings in the social and municipal housing sector; (3) to make positive use of the opportunities that segregation (and more specifically ethnic segregation) presents through area-based programmes aiming at the renewal of the housing stock and the promotion of social services and initiatives to increase social capital at a local level.

It is worth considering more specifically the arguments and features of these different orientations.

In the first direction, tenure mix, as a diversification of housing (type of housing and tenure) in disadvantaged areas, has been introduced especially in the UK under a variety of rubrics – ‘sustainable communities’, ‘mixed and inclusive communities’ – and is now a central ingredient of government neighbourhood regeneration policies. The assumption forming the basis of these initiatives is that the promotion of tenure diversification in areas characterised by concentrated social housing will be a driver for the local economy, increasing the local residents’ overall levels of social wellbeing. However, rather than endorsing the policy, research on the efficacy of the policy as a mechanism for neighbourhood improvement has tended to give ambiguous results and highlight the absence of a strong and consistent relationship between tenure mixing and positive social wellbeing outcomes (Arbaci, 2007; Graham et al., 2009)

A second direction concerns the definition of criteria and procedures governing the allocation of dwellings in the social and municipal housing sector, as a combined tool to disperse low-income earners and immigrant families. This orientation is definitely the one guiding recent housing policies in France. From 2003 onwards, under the lead of ANRU (National Agency for Urban Renewal), Péuplement (populating) has been actively used as a tool for spatially reorganising social groups in the vast public housing estates around the country (Desage et al., 2014; Donzelot, 2006) and administrators can act to prevent the concentration of ‘problematic’ families in areas considered ‘at risk’. In other contexts local authorities have even stated a maximum quota of foreign residents that must not be exceeded in designated areas: in some German municipalities, such as Stuttgart, the proportion of foreign inhabitants in public housing estates is limited to 30% (Munch, 2009; Simon, 2003). These policies, however, have a long history of analysis and critical opposition (Agustoni et al., 2012; Bridge et al., 2012). A first critical issue is the questionable belief that the spatial proximity between groups or classes contributes to the social proximity of these groups or classes. According to Elias and Scotson (2004), for example, policies of social mix may collide with a substantial lack of willingness among different social groups to make contact and build cohesive relationships (Atkinson and Kintrea, 2001; Bolt et al., 2010). Indeed, in certain circumstances, specifically when people live together against their will, proximity can exacerbate conflicts and enhance their class and status differentials (Blanc, 2010; De Rudder, 1989; Gans, 1961). In addition, while the overwhelming majority of these interventions try to achieve a balance of socio-ethnic populations at the neighbourhood level, it is not completely clear what exactly constitutes a ‘fair mix’ (Bolt, 2009). To sum up, the policy of deliberately mixing tenures in housing developments in order to improve social wellbeing remains largely unsupported by the research evidence (Graham et al., 2009)

A third orientation is the one that favours policies to renovate the housing stock and to promote social services and initiatives enhancing social capital at a local level. Many European countries have developed such urban policies, known as area-based policies, at different times. The various measures and instruments used to implement these policies are action-oriented, interactive and multi-purpose in nature (Power et al., 2010). Often they focus both on the renovation of housing facilities and on the dynamics of social inclusion, incorporating training and employment programmes. In areas with a strong ethnic concentration, actions are oriented towards strengthening the resources available locally; for example, by consolidating ethnic economies, providing opportunities for vocational training, offering language courses, opening channels of trade with the rest of the city, fostering appreciation for the contribution of ethnic institutions, and facilitating the participation of foreign families in the design of housing (Edgar, 2004). In 1998, the Swedish government launched the ‘Metropolitan Development Initiative’, a project intended to improve the quality of public spaces and facilities, to combat early school-leaving, and to support the acquisition of professional skills. The goal was not to transform the housing mix but to promote the overall development of residents in the area, with a focus on fighting unemployment and improving social mobility.

Given the three orientations that have been discussed above, the aim of this paper is to discuss an additional direction in which the reference to social mix has been adopted in policy making. The case study of Milan can be considered with reference to those experiences in which social mix is employed as a sort of ingredient in urban development projects where a significant stock of new housing is planned. Social mix loses its character as a measure against social and ethnic exclusion, while it appears instead as a rhetoric sustaining a diverse urban environment on the basis of vaguely moral arguments: diversity, the benefit of mixing different populations and tenures, providing more vivid and vibrant living environments, and as a tool to answer to the housing needs of the vulnerable middle class. The discourse on the virtues of social mix in this case tends in fact to obscure more significant economic interests and negotiations among developers and public actors (Bolocan Goldstein and Bonfantini, 2007). These issues are very much at stake in large-scale urban projects dealing with the developments of consistent new urban areas in which, after the dismissal of the modernist approach to zoning and separation of urban functions, the emphasis is very much now on the combination of a variety of local and supra-local functions (Breckner and Menzl, 2012; Bruns-Berentelg et al., 2010). With reference to the issue of social mix, it is interesting to observe the shifting arguments that are used in the design and implementation of new interventions. Rather than preventing the exclusion of the disadvantaged, the aim appears to be more significantly focused on the production of new values and qualities as well as creating conditions for greater feasibility and profitability of real-estate investments. In a rather different way, social mix has been advancing both in new housing projects and in social housing programmes. The principle has increasingly been affirmed as the equitable combination of inhabitants and this is evident in all new privately driven social housing projects which benefit from public funding (for example, through the allocation of publicly owned plots in favourable sites).

In order to explain this approach, before describing the different policies and interventions oriented to foster social mix in the city, we next present some overall information concerning the housing stock, the conditions of socio-economic and ethnic segregation and challenges in the city of Milan. ²

The housing question in Milan and the emerging rhetoric of social mix

Italy, like other southern European countries featuring so-called Mediterranean welfare regimes, is characterised by a vast majority of citizens who enjoy home ownership and a very limited stock of social and public housing (Allen et al., 2008; Poggio, 2008). Italy, Greece, Spain and Portugal have high home ownership rates, and this form of tenure is widely distributed across all social strata. However the ‘democratisation’ of this tenure (Poggio, 2008) is incomplete, because social inequality still fosters differences within the home ownership sector as regards, for example, the timing of home ownership, housing conditions and the implications for household budgets.

Another shared feature among southern European countries is the limited prevalence of both the social housing sector and of rental arrangements with social connotations, such as cooperatives. A general lack of secure and affordable alternatives in the rental sector is due to both the underdevelopment of social housing and to ineffective regulation – or non-regulation – of the private rental market (Allen et al., 2004).

Within this context, access to home ownership has been a rational strategy through which to satisfy housing needs and to invest savings in the context of relatively underdeveloped financial markets.

Among the European countries, Italy is one of those with the highest home ownership rates: in 2008 the proportion of those living in their own flats was 81.5%, with the 17.2% of people renting their dwellings concentrated significantly in the bottom two quintiles of the income distribution (Cittalia, 2010). The public and social housing stock plays a very residual role. Only 4% of the housing stock is in public hands, against 36% in the Netherlands, 22% in the UK and 20% of the EU average.

Moreover, along with similar processes in northern European countries, in recent years, right-to-buy policies have been locally delivered by public housing companies and local authorities without any public debate regarding their consequences for the overall availability of affordable rental housing. While very limited new public housing projects have been undertaken in the last 30 years, the overall public housing stock has shrunk by more than 20%: from 1 million dwellings in 1991 to 900,000 in 2001 and 800,000 in 2007 (Cittalia, 2010). Meanwhile, the annual production of social housing decreased from 34,000 dwellings in 1984 to only 1900 in 2004. It must be assumed that the poor consistency of the stock has limited the severity of the effects of social segregation: although public housing estates tend to witness a significant concentration of distressed and multiply disadvantaged families, these estates are rather limited in their number and extent.

In fact, Milan – with a current population of 1,354,000 – despite being one of the most unequal cities in Europe in terms of income distribution (Cucca, 2011), is not particularly affected by spatial segregation in comparison with urban contexts elsewhere in Europe and particularly in North America. This represents an important issue in any discussion of social mix in Milan. Essentially, although the city centre is the area where the richest inhabitants are more likely to settle, in the rest of the urban context there is a ‘spotted’ distribution of medium- and low-income people. The social housing stock plays a very residual role indeed, representing only 11% of the overall housing stock. Given the very small and unregulated rental market, affordable housing is quite limited and the public housing stock (composed of dwellings either owned by the Regional Public Housing Agency, Aler, or by the city of Milan) is rather small in comparison with the potential number of those who are entitled to access public housing. The official list of those who have made an application to access public housing includes large numbers of families who should be fully entitled to a social dwelling but will never be granted one in the medium term. In 2007 the official list numbered 13,000 families; in 2011, 19,000; in 2013, 23,000; and these numbers are considered to significantly underestimate actual demand. As far as provision is concerned, on an annual basis, the availability of social housing flats was limited to only 1000 in 2013, mostly consisting of older dwellings released because of deaths or made ready after renovation. This has resulted in a situation in which the only people who have a chance of being assigned a social dwelling are those who are under eviction orders (more than 10,000 in 2012) or multiply disadvantaged families. Some areas of municipal and deprived private housing are characterised by the large presence of recent immigrants (Agustoni et al., 2012), who are the social groups at the bottom of the socio-economic ladder in Milan; other municipal housing areas are places where families with particular social problems are concentrated (mentally and physically disabled people, elderly people living alone, evicted families, etc.).

More than residential segregation, the main problem in Milan is the scarcity of affordable housing (Mugnano and Palvarini, 2011), which is the result of a particular approach to urban development characterising the government of the city over the last 20 years. While on the one hand, the production of new social housing was extremely limited, on the other significant quantities of new housing stock have been provided for the private market. This has targeted the more affluent at a time when the real estate market has been booming and driven by rather speculative investments (Bolocan and Bonfantini, 2007; Nicolin, 2007). Over this period, the city has adopted ‘entrepreneurial’ policies aimed at the maximisation of property values that closely mirror the ideal type of pro-growth urban regimes (Harvey, 1989). This has been achieved particularly through strategies of urban planning that have favoured the intensive use of land for private investment while the proportion and quality of public benefits have been very limited. The former city administration explicitly accounted for the rise of real estate values by themselves as a comprehensive index of the city’s good performance, but failed to take account of any negative impacts this may have had. Nor did they consider the tendency towards increasing social and urban polarisation. Consistently with this approach, housing policies aiming to foster affordability have been more or less abandoned for more than two decades, and this can be considered one of the most important institutional factors exacerbating the conditions of social and spatial inequalities in Milan, considered the most unequal city in Italy (Ranci, 2010). The analysis of this process in Milan, and its transformation, both in terms of social and economic trends (Bonomi, 2009; Lodigiani, 2010; Ranci and Torri, 2007) and in terms of style of governance has been the focus of a large body of literature developed over the last decade on the city.

The main result of this approach to urban development has been an unjustified increase in housing costs. Between 1991 and 2009, the average increase in salaries was in the region of 18%, but the boom in the real estate market saw costs in the private rental market rise by up to 105%. This has resulted in a generalised increase of the number of evictions (Cittalia, 2010) and an increasingly tense situation for those who can barely afford housing in the current economic crisis. Reductions in the average family size, increases to their absolute numbers, changes to the traditional family structure (increasing numbers of singles, single-parent families, etc.), the ageing of the urban population and large numbers of new immigrants are significant demographic factors influencing a more articulated and atypical housing demand. At the same time, the spread of uncertainties in the labour market is placing vast numbers of citizens in conditions of housing risk (Baldini, 2010; Torri, 2006).

Social mix and neighbourhood regeneration: Acquiring new tenants and creating value?

Identifying the specific period in which social mix started to be a reference for public policy in Lombardy and Milan is a complex task. According to some key informants, a public discussion on mix started in the late 1980s, during the process of deindustrialisation, when beside the housing demands of the working class, new needs and demand for public intervention in the provision of affordable housing started to be heard from the lower-middle class. For some decades, this request had been satisfied by public support for ownership and partially implemented through the privatisation of public housing stock. Over the same period, decreasing levels of intervention from national and local authorities resulted in a drastic reduction in the production of municipal housing, accompanied by low turnover in existing municipal housing. This fostered a progressive process of ‘filtering down’ the inhabitants into public buildings. This has effectively resulted in an increased concentration of disadvantaged social groups, especially in the most deprived neighbourhoods, as well as a rising stigma associated with the municipal housing stock. Nevertheless, as reported by the key informants, from the very beginning the reference to social mix was not oriented to challenge segregation processes in the most disadvantaged areas, but rather to open up access to housing provisions supported by the state to low- to middle-income social groups, rather than prioritising the most disadvantaged population. Within this framework social mix represented a tool to promote a different type of ‘social justice’, whose beneficiaries were to be the (new) vulnerable people of the middle class. This clearly represents a specific characteristic of the interpretation and reference to social mix adopted in planning documents and housing policies in Milan.

According to some key informants, however, the first introduction of interventions oriented to social mix in Milan dates back to 2000, during a season of urban regeneration programmes, mainly developed through the EU ‘Urban’ programme and the national programme ‘Contratti di quartiere’ (neighbourhood contracts) which specifically targeted the regeneration and development of public housing estates (Bricocoli, 2008). For the first time, attempts to reduce the concentration of particular social groups, such as the elderly or migrants, were introduced as implicit or hidden objectives in the programmes’ agendas. In particular, increasing attention to the relocation of some immigrant families is owed to an emerging public concern about their progressive access to the municipal housing stock. The access of immigrant families to the public housing stock, usually the most deprived housing stock unwanted by Italians (Cucca, 2012), has been seen not only as a potential threat to urban social cohesion, but especially as the ‘expropriation’ of part of the already poor municipal housing stock (Agustoni and Alietti, 2011) in favour of these ‘strangers’. In any case, after a few years, this ‘problem’ was partially ‘solved’ through the introduction of a regional law in 2004, which provided strict entry barriers for people not already resident in Lombardy for at least five years. This paved the way for a very restrictive law at national level (in 2008) excluding a significant proportion of recent immigrants (Gargiulo, 2011), who were now also affected by extremely difficult conditions in the private housing market (Agustoni et al., 2012).

In the neighbourhood regeneration programmes in Milan, social mix has often been promoted through the allocation of the new units obtained as a result of the renewal process to less problematic social groups, such as students; additionally, in situations characterised by an ‘excessive homogeneity’, social mix has been fostered through the mobility of residents: the very rare situation of a concentrated presence of mentally disabled people in some units has systematically provided a scapegoat. It is interesting that these interventions have been driven by remarkably vague criteria about the concept of social mix; both the allocations and new destinations of the tenants obliged to move have been decided by the project managers on the basis of ‘common sense’ regarding what a balanced community should be. The idea of social mix has from the beginning been assumed as an indisputable necessity, without regard for the problems that the coexistence of different social groups may represent. The first experimental programmes, however, were integrated with social interventions vaguely recalling the need to build new communities (Bricocoli and Savoldi, 2010); however, these interventions worked only for a very short time because of the paucity of financial resources available to these experimental programmes. After a very short experimental phase, social mix interventions have been promoted by different actors, simply placing different ‘categories’ of people, also defined through the ‘common sense’ of the operators, closer to other social categories generally able to pay higher rents.

Social mix and new urban development projects: Organising separation?

Besides being used as an argument for the filtering of tenants in terms of access to the existing available stock, social mix in Milan has been used as a main argument for gradually excluding a relevant number of publicly owned flats from the stock which is reserved for social housing and allocated through the official waiting list. In spite of the tremendous demand for housing by socially deprived people, the publicly owned housing company has been increasingly urging more rewarding uses of the stock in its control. Significant parts of the housing stock have been reserved for other users. In many cases the allocation has targeted those who have jobs which are considered valuable to the city (the police, nurses, bus drivers) as well as students. Despite the limited individual resources of students, the agreements developed between the housing company, the state universities and the city generally allow much higher rents than those paid by the social tenants. The interpretation of social mix is somehow very functional as it is implemented though the simple development of single different buildings and populations and there is no consideration of any concrete or real processes of social interaction among the people. In Stadera, a public housing estate from the 1920s located in a semi-central area of dense traditional blocks, social mix is implemented through densification and new construction. Two buildings have ceased to be dedicated to social housing and are allocated to the elderly and to students. In both cases, housing is organised as a specialised function – a nursing home and student housing – which in fact prevents any interaction with the surroundings; no outside use of the facilities is allowed, and even a reading room is not open to a mixed set of users (Marzorati, 2010). In Gratosoglio a housing unit with 45 small apartments has been allocated to a category of tenants recently identified as a target: divorced men who need a place to welcome their children to from time to time. This suggests that for the sake of social mix, new classifications and separations are being produced (Coppola, 2010).

New development projects introduced in the last decade show how a set of devices and rules which have been orienting public–private negotiations as well as the design and organisation of urban space have been systematically producing fine-tuned and significant separation (Bricocoli and Savoldi, 2014). An interesting example is the case of the Pompeo Leoni project. The case is representative of a set of interventions that were devised in the late 1990s for the redevelopment of vacant industrial sites (in this case, a former Fiat plant) (Bricocoli and Savoldi, 2010). The development has been regulated according to an agreement which was widely shared and welcomed even by progressive Milanese planners. The negotiations with the private investors resulted in a public green area (50%), a residential area (private, cooperative, public) (25%) and a commercial area (25%). What was in absolute and general terms considered as a fairly positive planning result for the public is now a very introverted development, with no significant connection to the existing mixed context, dominated by one single shopping centre and by a fenced and separate organisation of each different housing block according to its specific tenure. Whereas all the housing developments consist essentially of buildings detached from the street-line, surrounded by very small yards and strong fences, the green areas clearly denote the separated character of this development from the surrounding existing urban pattern. In the words of many inhabitants, this separation is exactly what gives added value to the project.

Beside the recurrent reference to urbanity and to the need to avoid the shortcomings of socially homogeneous and mono-functional housing areas (which is often assumed to be a criticism of social housing estates) our exploration of the case reveals that many of the features which are supposed to be urban qualities in the eyes of the inhabitants have turned into negative externalities for the city. The possibility of the coexistence of a multiplicity of different uses and populations is definitely seen as a risk rather than a valuable opportunity. Within an urban space that is ideally flat and smooth, any unforeseen and eventually controversial use does not find any opportunity for mediation, but rather produces conflicts which are immediately conceived and treated as issues of public order (de Leonardis, 2008). In Pompeo Leoni as well as in several other urban development projects, the principle of spatial organisation that is experiencing growing momentum is separation. This is the physical separation through barriers and fences, as well as the separation produced by public green areas. Their main role seems to be to act as buffers, distancing and separating people. Separation is moreover enforced by an organisation of collective housing which is mainly driven by a limited investment in the innovation of typologies, that is, avoiding any sort of risk in merging and combining uses and populations in the same building.

The use of social mix in new social housing projects: Profiling good communities and raising profits?

In recent years, there has been an increasingly frequent reference to social mix, especially with reference to new housing projects (Fondazione Housing Sociale, 2010). First, a neoliberal discourse has ideologically stigmatised and rejected public housing estates as areas at risk of ghettoisation, over-representing once more the social homogeneity of these housing areas which cannot be compared with that of other major large cities in Europe and America. On the other hand, the involvement of private actors (for profit and not for profit) as promoters of new social housing initiatives has been presented in public debate as the only feasible approach to the provision of affordable housing, reducing the public actor to the role of enabler of projects mainly driven by private actors. In this respect, social mix has become an appealing argument that may be combined with principles of competitiveness, feasibility and rentability of the projects.

Although a combination of tenures and clients in different economic situations is definitely a relevant factor in the production of social housing by non-profit actors, it is still interesting to observe the intertwining and convergence of these different elements within a vision in which social housing is increasingly defined as a field of action where the target of the lowest rental costs for the social tenant appears no longer to be the main focus. At the core of the debate and of local programmes, the main issues are the economic feasibility and short-term rentability of projects for their promoters. This has resulted in a general rule which regulates social housing projects being planned in plots owned by the city and which is severely limiting the ratio of social housing dwellings rented at social costs in comparison with the quota of flats which are built by the developer and sold at market prices. Although in the general debate this has led to an extension of the notion of social housing to include private ownership, the lack of any consideration and debate on innovation and reducing housing construction costs must be stressed. We can observe that social housing projects are providing a small quota of social rental flats, while representing significant opportunities for developers who are granted significant public support and direct benefits at a time of economic crisis, in which the market is definitely not offering the high profits of the recent past. Besides this, broad emphasis has been laid on a whole set of issues concerning community development within (and through) housing projects, along with a rather moralistic interpretation of the virtues of ‘living together’ and of collective housing which is very much keen on envisioning ‘good’ combinations of ‘good’ tenants as major factors of success in the development of new housing (Bianchetti, 2014; Sampieri, 2011).

For the current state of the art in the development of new social housing projects and public and policy discourse, one needs to consider ‘Abitare Milano 1’, a pioneer programme which introduced relevant innovations and shows how the interpretation of social mix has been controversially orientating not only the social housing initiatives led by the private actors but also the design of housing policies and projects even when they are fully public.

As emerged during intensive field research in 2011 and 2012, the housing programme ‘Abitare Milano 1’ offers a very interesting perspective on the influence of the social mix rhetoric on the design of local public policies (Bricocoli and Cucca, 2012). Abitare Milano 1 is the first huge project for new social housing units launched by the city of Milan in the last two decades (Infussi, 2012). The programme was launched in 2005 with four new housing projects to be developed on publicly owned land. The projects were developed along guidelines provided by the city and selected though an international competition. The principle of social mix was targeted through a very normative approach aiming at identifying and selecting specific numbers of flats and categories of tenants. A new category of rent, the so-called convenzionato (‘conventioned’), calculated on the basis of the return on the investment by the developer, was introduced and adopted, although the land and the construction costs were fully covered by public funding. In the convenzionato solution, the rent is proportional to the size of the apartments and independent of family income. During the implementation of the programme, the previous city administration even increased the number of flats to be allocated at a higher rent to specific categories of tenants. The reduction in the number of flats to be let at social rent was motivated by the desire for a variety of socio-economic profiles and to avoid the concentration of disadvantaged families. In fact, these housing projects are quite small in size (a maximum of 180 units) and are located in socially mixed areas. Also, the imperative of mix produces controversial effects: the identification of limited and specific categories of tenants on the one hand, and a new allocation procedure on the other, resulting in paradoxical effects of homogeneity. In one of the housing projects, for example, more than half of the tenants are police officers.

In a context in which the housing situation is very tense, affordable rental housing for low-income families is so limited and socio-spatial segregation very modest, the systematic return to social mix seems to contribute to the exclusion from welfare policies of a significant number of potential recipients. In fact, the reference to social mix – considered and debated as would be any technical issue – produces a sort of de-politicisation of the social nature of housing policies (Desage et al., 2014).

Conclusions

The aim of this paper has been to focus on the possible different implementations and effects of social mix in housing policies. As already stated by a large body of literature, social mix cannot be considered a scientific concept, but rather as a legitimative construct suggesting social problems can be solved indifferently to their reasons. As stated by Bacqué et al. (2011), social mix cannot settle the issue of poverty through social dispersion, while this can be done only through vital social redistribution. In fact, many investigations have already shown not only the weak impacts of social mix policies on the presumed negative impact of the ‘neighbourhood effect’, but have also demonstrated possible unexpected consequences, such as the creation of new conflicts, more stigmatisation and the weakening of social networks.

By analysing the case study of Milan, this article has discussed a new use and new risks in the implementation of this rhetoric in urban planning, especially in countries characterised by a low level of spatial segregation and a low availability of municipal and social housing solutions.

In Milan, policies are promoting social mix in the name of a broad and often generic set of arguments, such as the necessity to respond to the housing needs not only of the most disadvantaged social groups but also of other targets (key workers, students, divorced fathers with children, young couples, etc.) in terms of access to public housing provision. In general, what we argue here is that in introducing social mix as a value guiding new social housing projects in Milan – most of which have a private (although non-profit) nature – the drivers of social mix are different, and what is often being traded is a reduced blend of social tenants for the advantage of higher rents and overall rentability of housing projects. This is definitely greatly beyond the aim of including disadvantaged people into the benefits and wellness of a new housing environment. Within this context, social mix can foster more new inequalities and forms of spatial segregation than innovative approaches to urban justice. While in other countries the risk is more related to gentrification (Bridge et al., 2012), in Italian cities the effects are more connected to displacement, implications for household budgets and rights of access to the city. Additionally, the case of Milan highlights the problems related to the implementation of social mix strategies in contexts where affordable housing and the availability of social and municipal housing stock is low, because it usually fosters the exclusion of low-income groups from access to affordable housing. As the case of the Abitare 1 programme shows, it is an issue that the 30% of the new 500 dwellings provided have been devoted to households not on the waiting list of the municipal housing sector, while the number of poor families entitled to public provision is greater than 20,000. Furthermore, when social mix has been promoted through strategies of tenure mix, it has led largely to a permanent loss of public land that is potentially suitable for public investments, while obtaining a low percentage of affordable housing that is usually placed in the most deprived locations, fostering separation, segregation and stigmatisation. Additionally, in a context of the decline of the residential sector in Milan resulting from the crisis, the strategy to provide affordable housing through these programmes is not forward-looking.

In general, we can argue that although the call for social mix has been driven by a will to focus on the housing demands of the impoverished lower-middle classes, recent dynamics rather suggest the need for a closer investigation of the housing conditions of the most disadvantaged social groups, who are definitely more exposed to the current economic recession.

Finally, another concern is related to the process of ‘translation’ of the concept in different contexts. The issue of social mix has mostly arisen in the academic and policy debate in North America and in northern Europe, but we need to discuss whether and how it has become a type of catchword, a passe-partout category, which is transported into the academic and policy debate of urban and national contexts in which conditions are quite different from the ones in which the desire for it originated. This investigation, for example, has highlighted the risks involved in promoting social mix strategies in countries characterised by a Mediterranean housing regime. However, this may be true for several other concepts and categories as well; the case of Milan shows the extent to which an instrumental use of a shared and recognised concept can be achieved.

This is a matter of significant interest to urban studies and critical academic work at the international level, which may be influential in identifying differences and contextualising issues in mainstream discourse, avoiding the ambiguity of a generic application of analytical categories which are very often translated into normative attitudes.