Social and Ethnic Transformation of Large Social Housing Estates in Milan, Italy: From Modernity to Marginalisation

This chapter examines the case of large estates of social housing in Italy’s economic capital, Milan. Production of this housing occurred in the period of intensive industrialisation and associated urbanisation from the late 1940s to the mid-1970s. Development of these schemes occurred mainly in the periphery of the city, and led to land speculation and changed the social geography of the city. These estates initially housed Italian economic migrants attracted to Milan during the ‘economic miracle’, and since the 1990s have been the residence of a growing number of international migrants. Housing estates ceased to be developed after the 1980s, and a large part of the stock has been privatised since the 1990s. Today housing estates are more heterogeneous in terms of tenancy regimes and the social and ethnic groups who live there. The majority of the stock shows signs of (often serious) physical deterioration. The resident population has aged in situ, with ethnic segregation occurring in some residual parts of the stock. This chapter studies the evolution of these large social housing estates in spatial and social terms, using published and unpublished data from 1951 to 2017, pointing out their critical points and their potential.

from city to international metropolis. During the period of growth , Milan consolidated its position as Italy's capital of industrial and housing production. Milan's urban landscape was transformed by the development of large industrial sites and housing estates that respectively employed and housed thousands of Italian immigrants.
After this period of growth, decline set in characterised by processes of de-urbanisation and deindustrialisation (Petsimeris 1998). In the late 1970s, the production of large housing estates drastically ended. Since the 1990s, the internationalisation of Milan's demographic structure due to global economic migration flows has become an established trend. The city has experienced a transition from domestic migration to international migration and a change of its economic base from an industrial to service economy (Bolocan Goldstein and Botti 2006). Other processes are the ageing of both housing estates and their tenants (Mugnano and Zajczyk 2008).
Our analysis will focus on the evolution of large housing estates and their populations using census data from 1991 to 2011, special data produced by the Piano Intercomunale Milanese (PIM) and interviews with the inhabitants and actors in the city of Milan. Distribution of housing estates of the public sector (ALER former IACP and council housing) in intra-urban space will be examined in terms of their temporality, and their qualitative and qualitative characteristics by taking into account the social and ethnic division of the city (Petsimeris and Rimoldi 2016).
Large housing estates constitute an important factor that quantitatively and qualitatively conditions the urban landscape of the city and its social geography. Despite their overwhelming appearance, they are not homogeneous but different entities in terms of housing density, planning standards, architectural aesthetics, degree of maintenance and social and ethnic composition of their populations. This chapter explores how these estates transitioned from Fordism to post-Fordism, the relation between location and socio-ethnic segregation, and the outlook for these estates as an integral part of a north Italian metropolis.

Genesis and Evolution of Large Housing Estates in Milan
In order to outline who was responsible for the development of large housing estates in Italy, one has to start with the question of what is 'social housing' in the Italian context. An official definition was provided as recently as 2008 (D.M. 22/04/ 2008), which defines social housing as 'mainly dwellings rented on a permanent basis; also to be considered as social housing are dwellings built or rehabilitated through public and private contributions or with the use of public funding, rented for at least eight years and also sold at affordable price, with the goal of achieving a social mix' (Caruso 2017, p. 23). As Caruso goes on to make clear, social housing, therefore, includes a mix of tenancy categories (rentals and owner occupation), and at the national level does not set out criteria by which housing should be allocated. The underlying principle for social housing in Italy is to promote social cohesion by means of reducing housing stress through the provision of housing to low-income households (Caruso 2017, p. 23).
In contrast to those countries that have received the most attention in literature on social housing, Italy had a national policy on social housing for a relatively brief period (Padovani 1996;Priemus and Dieleman 2002), and is currently at the bottom of the European league table in terms of number of social housing units by country (Tosi and Cremaschi 2001). Commenting on Italian housing policy in general, Tosi and Cremaschi note: 'The traditional model of housing policy that has characterised the Italian system is one in which public intervention in housing has been conceived mainly as a side by side intervention with respect to the market, in order to satisfy the demand which is not able to access the market, rather than having the objective of regulating the market' (Tosi and Cremaschi 2001, p. 14).
Private actors include: individuals, building societies, insurance companies and state pension funds. Not only has the relative importance of these actors changed over time, so has the line between the public sector and insurance companies and state pension funds been blurred due to government intervention aimed at using the housing stocks of these two private actors in order to try to control rental levels. Despite the subsidiary role of social housing, Italy is home to some of the largest social housing projects developed in Europe, including a development of close to 4,000 apartments called Gratosoglio in Milan (developed between 1962 and1965), which helps to account for the relatively large share of housing accounted for by the public sector in the city. Responsibilities for housing are regional, which in the case of Milan means the region of Lombardy. Below the region, a high (and increasing) degree of discretionary power is afforded to the municipalities.
Social housing has a long history in Milan. In 1909, the City of Milan developed a large housing project on a 32,000 m 2 site to the north of the city. The development, named Mac Mahon, included five large four-storey buildings, a small number of rows of housing and independent cottages, in an experimental fashion combining housing and community facilities typical of housing projects elsewhere in Europe at the time. The IACP (Istituto Autonomo Case Popolari, the Institute for Public Housing) was established soon after and started the production of quartieri of working class affordable housing with good standards in line with the hygienist ideology of the time. These experimental schemes acted as ways of innovating the building process in terms of housing, and would progressively lead to the provision of internal services, tidy layouts and self-sufficiency. As was to be the case throughout Milan's history of social housing provision, the private sector was active alongside the public sector, with the Società Umanitaria building two large complexes in the city in 1904 and 1909. In the 1930s, the economic crisis had an impact on the production of social housing, and their size was reduced. After World War II IACP played a leading role in the process of housing production in order to satisfy the growing demand for dwellings, which was the result of the historically accumulated shortage combined with the massive destruction during the war and the pressing demand of domestic immigrants attracted to the city during the post-war economic boom (Castronovo 1970). After World War II, the IACP produced housing for soldiers and for people who were displaced. Housing was produced for white-collar households, the lower middle class and the working class.
After the late 1940s, two national programmes for the provision of social housing began: INA Casa (Istituto Nazionale Assicurazioni Casa, Institute for Social Housing) and, from 1963, Gescal (Gestione Case per Lavoratori, Institute for the Administration of Worker' Houses) (Di Biagi 2001). Essentially, these were means of raising finances for the development of quartieri autonomi (selfsufficient neighbourhoods) such as Harrar, Lorenteggio and Comasina. These autonomous neighbourhoods were financed in part by 'forced' contributions (deductions from wages) from the workers (private sector or state employees), and in part from contributions by employers and the state (Ferracuti and Marceloni 1982). These contributions were centralised by the CER (Comitato per l'Edilizia Residenziale, Committee for Residential Building) before being distributed to the regions. The aim was to reduce the housing shortage but also to create employment in the building industry (segmented in small units with a low-qualified, mainly manual workforce). The ideology behind the programme was to provide a means of access to home ownership by means of riscatto (i.e. after a number of years the tenant would become the owner). Both the ideology and spatial practices of these programmes have been criticised by a number of scholars-such as Tafuri (1986) and Secchi (1972) and Indovina (1972) -while others have criticised the social segregation generated by these schemes (Carozzi and Mioni 1970). This source of capital was important for housing production although insufficient to cover all housing needs.
Even if Milan has played an important economic, social, and political role throughout the history of Italy, its urban fabric came to be most dramatically transformed during the period of the so-called 'economic miracle' from the end of World War II to the late 1960s. The development of large social housing estates was a key element of this urban transformation. Main characteristics of this brief period were the processes of industrialisation and urbanisation, and the modernisation of Italian society and economy. Milan was not only the protagonist and main generator of these processes in terms of decision-making and investment, but also the territory in the so-called 'industrial triangle' (Milan, Turin, Genoa) that experienced the most dramatic quantitative and qualitative changes passing from the status of a city to a metropolis with significant transformation in its urban landscape and housing stock.
Milan has a concentric form that still characterises the city's spatial structure and social geography. The core is composed of the historic centre of the city delimited by the Spanish walls. Many prestigious institutions and firms, and the residences of the elite are located in this area. The core accounts for 4.5% of the total area of the city. The second ring is the extension of the city at the end of the 19th century (Corpi Santi, Greco Milanese, and Turro Milanese) (36%), and the outer part of the city (60% of the total area) is a heterogeneous area formed by the annexation of 12 contiguous rural municipalities (Baggio, Trenno, Musocco, Affori, Niguarda, Greco, Gorla, Precotto, Crescenzago, Lambrate, Chiaravale, and Vigentino) completed in the 1920s. It was mainly the municipalities of the outer rings that absorbed the rapid processes of urbanisation and industrialisation, and which were transformed from agricultural and rural areas into zones with high concentrations of large housing estates that gave shelter to large numbers of migrants in collective housing (see below). In these areas, there were contradictory processes of production and appropriation of space with the coexistence of small historic nuclei, large housing estates, industrial plants, self-promoted housing developments by the migrants of the 1950s and 1960s (coree), and shrinking green and agricultural areas.
In the 1950s and 1960s, Milan attracted many hundreds of thousands of Italian economic migrants originating from a range of places: from the city's more proximate rural areas to Italy's most remote regions, particularly in the South. Between 1951 and 1971 Milan gained just under half a million inhabitants (Table 12.1). The processes by which these migrant workers and their families were housed occurred rapidly and largely in the absence of land use planning, resulting in the development of virtually every free space towards the periphery by large housing estates located on greenfield sites. Many of the sites had poor accessibility; some were located between industrial estates or along (but without direct access to) highways or railway infrastructure, while others were in the middle of historic centres and farming areas of contiguous rural municipalities (Gambi 1973;Dalmasso 1972). At the same time, there was another process of major renewal and intensive development in the centre of the city, consolidating its historical role as the elected space for residences of the upper middle classes, and the location of financial and high tertiary institutions (although it still included enclaves of derelict housing, 'cheap' bedsits, and short-term rentals). Production of these peripheral housing estates was-at the time -considered to be a miraculous solution, giving shelter to new urbanites. Nevertheless, it consolidated the territoriality of social polarisation between the centre and the periphery (ILSES 1964;Cerasi and Ferraresi 1974).
From Table 12.1, it appears that during the six decades, there has been a continuous improvement in terms of overcrowding: from 1.21 persons per room in 1951 to 0.58 in 2011.
Also impressive is the improvement in the provision of dwellings with basic facilities: from 30% in 1951 to 99.88% in 2001. But the most important change that has occurred is in tenure regime. Homeownership tripled its share, jumping from   26.3% in 1951 to 74.3% in 2011. The picture, however, is far from euphoric. Even if housing densities have declined and housing conditions have improved, these have occurred in a context of economic crisis characterised by deterioration in employment opportunities. Associated with this are threats to continuity of employment and income from employment for the less skilled, the continuous decline of the welfare state and reduced availability of social housing. As a consequence, the most vulnerable households find it increasingly difficult to solve their condition of housing stress.
A comparative analysis by IAURIF showed that the intensity of production of housing in Milan was greater than in metropolises such as Chicago, Amsterdam, Los Angeles and London (Dalmasso 1972). As a consequence, the power of the real estate sector grew enormously both in Milan and nationally, with the group of developers, contractors, landowners and real estate companies coming to be known as the blocco eldilizio (the real estate block) (Indovina 1972). This sector has been able to exert significant influence on national policy and city finances, and is widely acknowledged to be linked to mafia interests. The greatly increased pace of housing production was nevertheless unable to redress both the historical shortage of decent quality housing and to provide shelter for all of the new arrivals (ILSES 1964;Garzena and Petsimeris 1984). Unsatisfied demand for housing was expressed towards the end of the 1960s and in the 1970s by protest movements, which occasionally erupted into violent tensions as protests over housing issues were incorporated into protests over working conditions.
Since the housing reform introduced in 1971 (Law 865), responsibility for social housing has progressively passed from central to regional and local governments, with regional government being responsible for setting the objectives of social housing, and establishing regulations for management and funding. This trend reflects the general process of liberalisation, which in this case involves the privatisation of the institutions responsible for public housing and the redefinition of the public sector's role relative to the market in terms of welfare provision. Such processes have, of course, not been confined to Italy. At the same time, levels of construction have continuously declined. This has occurred within the context of reduced central government finance to regions, which makes regions more reliant on their own resources and places them in the position whereby the sale of social housing presents one means by which to balance their budgets.

Location and Diffusion of Large Housing Estates in Milan
Large housing estates are cause and consequence of the transition of Milan from regional centre to Italy's largest metropolitan area and economic capital. They reshaped the city's skyline and its social geography. Figure 12.1 represents the location of large housing estates in Milan by period of construction and gives an indication of their size, form, typology and spatiality of their diffusion. Large housing estates are many, scattered, different in size and form, and adhere to After Centro Studi PIM different visions, architectural fashions, and planning practices. Milanese architects produced rationalist schemes (Belgiojoso in Grattosoglio), modernist (Aymonino) or postmodernist (Aldo Rossi) in Gallaratese (Belgiojoso 1972;Tafuri 1986;Grandi and Pracchi 1980). These estates are now at different stages in terms of ageing, decay and maintenance, and continue to occupy a significant amount of space mainly towards the periphery of the city. These large housing estates are scattered across the urban fabric but not in a haphazard manner because they are located more towards the periphery-where land was available-and their concentration is stronger in the west-the more industrialised part of the city-than the east. The height of the buildings has increased over time. Large estates developed before World War II were four storeys high, those developed in the post-war period were six storeys high, and later developments in the 1960s and 1970s reached 12-16 storeys in height. These taller developments were constructed using industrialised methods of heavy prefabrication 'à la française'. Establishment of Gescal and the introduction of a national 10-year plan for housing provided a context for experimentation by public housing authorities, and Milan led the way. An agreement was signed with five providers of prefabricated housing for a construction programme of 21,000 houses within 1968, including the Olmi neighbourhood at Baggio. The impact of these high-rise estates on the cityscape has been considerable, not least because in most cases, the engineers, planners and architects failed to integrate them with the rest of the city.
There is a positive relationship between the size of the estates and their distance from the city centre: the larger the floor-print of an estate, the more peripheral its location. Part of the explanation for this pattern is afforded by decreasing land values along a centre-periphery gradient. However, this has to be qualified by the fact that these developments contributed to fuelling increases in land values through speculation on free spaces located between the periphery and the centre (Graziosi and Vigano 1970).
Large housing estates were distant from any sort of centrality, symbolic values, public and cultural services and institutions (Gambi 1973). Generally, they were located in the middle-of-nowhere, forming ghettos without infrastructure, with poor accessibility to the city centre and to the work places of their inhabitants. There was a severe shortage of shops, public and private services, cultural activities and public space. Justification for their marginal and inaccessible location put forward at the time had to do with lower land values toward the periphery, which would be the means by which rents could be made affordable for households with low incomes. But this was part of a speculative mechanism invented by the landowners and tolerated by the state. Landowners frequently sold less valuable land holdings on the administrative boundaries of the city to ICAP and the city authority at prices close to nothing. These landowners gifted these peripheral sites so that they could subsequently benefit from the betterment of their land holdings located between these sites and the centre, a betterment which was provided by the infrastructure and social services that the state financed in order to dis-enclave the peripheral housing estates (Ferracuti and Marceloni 1982). This process was dislocated from city planning. The drawing board designs of internationally acclaimed architects went on to become mega containers for the segregation and dislocation of lower income groups.
The last developments of IACP  are located in the peripheral parts of the city (such as Gallaratese, Quarto Oggiaro, Sant'Abrogio and Gratosoglio) and are much larger than previous ones. Development was undertaken by IACP in association with the municipality of Milan. As the pressure of urbanisation continued, overspill increased in the form of large housing estates in the contiguous municipality of Rozzano. These are characterised by a high concentration of housing by IACP, which accounts for more than 50% of the housing stock of the municipality. There has been a process of diffusion of the housing estates from the centre to the periphery and from the ground to the sky, which has brought about the rapid transformation of agricultural land to built-up areas, and the further diffusion of Milan to the first, second, and third rings.
Gratosoglio is a massive housing estate built in the southern part of the city in 1962 and designed by the famous architecture studio BBPR. It consists of 21,000 prefabricated units, arranged in 52 buildings on 9 floors. In 1972, an extension to this development was added in the form of eight white towers of 16 floors each ( Fig. 12.2).
The spatial arrangement of apartment towers is depicted in Fig. 12.2. A rectilinear configuration along a main axis route, via dei Missaglia, connects the development to the southern periphery-and eventually-the centre of Milan. In the context of their immediate and not so immediate surroundings, these 16-storey buildings are imposing. Furthermore, the juxtaposition of these buildings with a business park and commercial units trading in luxury brands on the other side of Via dei Missaglia is accentuated by this axis route, which has now become a new frontier.
Many of the problems that have occurred since the 1970s in this development were foreseen by the architect Belgiojoso (1972): rigidity in terms of prefabricated structural elements, poor accessibility, segregation due to the isolation of the working class, and failure to provide services concurrent with families taking up residence in the development. There were also a number of positive aspects: orientation favouring good luminosity, hygiene, conception, spatial integration of services, and the provision of green areas. It will be interesting to see how the current attempts at regeneration will draw on the positive aspects of the original architectural design and help move towards the amelioration of the more negative aspects.
The following observations were made during fieldwork conducted at Grattosoglio in February and December 2017. Most of the public space at ground floor level is currently empty (Fig. 12.3), the most notable exception being the queue of people outside the ALER offices.
While the laundry and pharmacy continue to function, all of the other shop units originally provided in the development are now vacant and boarded up. A covered market, together with a small supermarket, serves the needs of the local population. Both of these opened 20 years after the housing development was completed. The people who use the public open space are mainly elderly and Italian. When questioned about their living conditions they stated that they were generally content. However, as the conversation developed many of the older residents agreed that the problem of the development's isolation during the 1970s had come to be replaced by an increasing sense of insecurity among its residents. This they The green areas are large and generally well maintained although-at least at the time of my fieldwork-the children's play facilities were under-used ( Fig. 12.4). The road and tram connections with Milan are good, and have played an important role in reducing the perceived and practical isolation of the development. There is evidence of upgrading of a number of individual apartments and the renewal of the façades of some blocks of housing in via Baroni.
In the 1990s, a process of denationalisation of social housing occurred (Cremaschi 1994), with the devolution of the housing question to the local authorities. This is not only a question of change of name from IACP to ALER (Azienda Lombarda Edilizia Residenziale) in 1996 but also marks a period that corresponds to a decrease in resources for social housing. This has occurred alongside an increase in the size of the more marginal population due to international migration and a decrease in welfare provision by the state.
Since 1992 the public housing stock in Italy has been progressively eroded. In Milan, 60,000 units out of 100,000 have been privatised (Marini 2007;Granata and Lanzani 2008). The public housing stock managed by the ex-IACP is now subdivided into two parts: one owned and managed by ALER and the other owned by the city and managed by MM.

Social and Ethnic Succession in Public Housing Estates
In the beginning, large housing estates provided residences for domestic workers in skilled manual labour and white-collar employment related to the manufacturing industry. Skilled labourers were predominantly from the north-east of Italy, while unskilled labour was drawn from the south. After the 1980s, processes of deindustrialisation and migration from the Global South and Eastern Europe had an impact on the social structure of the city. These are reflected in processes of 'succession' of residential location in derelict and overcrowded areas in the inner city and in substandard large housing estates. Are housing estates containers of social and ethnic segregation, or are housing estates going through a shift from a social divide to an ethnic divide? Data are not available at the appropriate scales and periods to carry out detailed analysis of the changes in the micro-social geography of large housing estates. However, with data available from the PIM (Piano Intercomunale di Milano, Inter-municipal plan of Milan) and the council, the relationship between container and content will be outlined.
According to the Observatory for Public Housing, in 2006 Milan had 39,923 dwellings, of which 14,333 (36%) were owned by ALER and 25,580 (64%) by others (mainly the city of Milan). It is noteworthy to underline that amongst them there were 2,631 dwellings occupied illegally and 1,741 non-attributed. This means that there is a problematic situation in terms of governance and turnover.
As is seen in Table 12.2, more than half of social housing has an age that ranges from 70 to 100 years and more recent housing has an age between 30 and 50 years. This means that there is a process of natural decay that is very important and big investments are needed for areas that are oldest. The same table also indicates that more recent housing estates and extensions are larger. This is very important in terms of the quality of standards both for domestic space and the areas assigned to green spaces and social and private services.
One can observe the impact of ageing housing on the quality of housing conditions of the public stock that can also have serious consequences on the social geography of the neighbourhood.
Large public estates are heterogeneous in terms of quality of housing conditions (Table 12.2). Decay is greater in the central city rather than in the rest of the metropolitan area, and it concerns mainly small dwellings of two rooms (58.2%) and three rooms (31.2%) in Milan, and dwellings of two and three rooms in the suburbs (54.6% and 25.4%, respectively). These dwellings belong to older estates in Milan and to the most recent in the periphery that are also relatively larger (Table 12.3).
Previous data concern the stock owned by ALER and the city council. More recent data are only available from the city council that owns 27,945 dwellings occupied by 24,648 households for a total of 50,500 inhabitants.
There is an over-representation of the elderly (Table 12.4), particularly that of the oldest age group (over 65 years old). These big housing estates were designed for young families with children, and services like schools, playgrounds and sport fields-even if often insufficient-were addressed to this type of household. Today the services are inadequate for a population which has aged in situ.
Images diffused by the media give the impression that large housing estates are mainly occupied by different groups of 'others': immigrants from southern Italy in the 1960s and from the Global South since the 1990s. But as seen in Table 12.5, public council housing in Milan is mainly occupied by Italians (83.17%) while non-Italians represent only 16.13%. Amongst non-Italians, the highest shares are recorded by nationalities from outside the European Union, such as Morocco (2.9%), the Philippines (2.3%), Sri Lanka (1.6%), and Peru and Ecuador (1.1% each). A comparison of this data with data on the occupation of housing stock in Milan shows that the Italian population is two percentage points lower, and that migrant groups record correspondingly larger shares: Philippines (2.6%), followed by Egypt and China (1.6% each), and Peru (1.4%). Each of these groups records a higher concentration in the city than in the large housing estates. Only immigrants from Sri Lanka and Ecuador (0.9% each at the city level) have higher concentrations in the housing estates. Therefore, Italians record higher concentrations in the large housing estates relative to the rest of the city, and the majority of larger ethnic minority groups have the tendency to live outside public housing estates. This pattern is due to the availability of the housing stock, and the history of migration to Milan of each ethnic group. Limited knowledge of more recent migrants on how to access social housing, and the eligibility and selection mechanisms used by the public housing institutions help to account for the differences. Other factors include the limited number of dwellings available to newcomers. These dwellings are mainly residual parts of the stock with high levels of decay, and very frequently are those that have previously been turned down by Italian households.
Another important issue concerns the size of households that occupy large housing estates (Table 12.6). The smallest households constituted by one or two members count for three-quarters of the total stock while larger households (five and more persons) represent less than 5%. This is a reversal of the situation during the Fordist period, when medium and large-sized households represented the vast majority in large housing estates. A comparison of small households in public housing with those in the rest of the city shows that single-dweller households are less concentrated within public housing (41.9%) than in the rest of Milan (45%). This suggests that in public housing there is a dissymmetry between size of the dwelling and the size of the household. This is most probably the case for most recent estates where dwelling sizes are larger.
This issue is important to address in the context of rehabilitation that is urgently needed for the majority of public stock. Through a process of subdivision, a better correspondence between size of household and dwelling size could be achieved, which would also help to address the issue of overall housing shortage. Such a solution, however, would require a considerable increase in the level of funds available for the refurbishment of a rapidly decaying housing stock.

Physical Decay, Ageing and Social Segregation
According to the last census in 2011, a number of large housing estates show high levels of social deprivation on the ACE scale (Table 12.7). The estates of Gallaratese, Barona, and Quarto Oggiaro face a number of severe problems in terms of ageing structures (50 years old on average), unemployment, and youth unemployment. The most deprived areas are Quarto Oggiaro and Selinunte. The former has the highest rates of unemployment (12.5%) and young people out of work and out of study (27%), and also records the second highest rate of rental accommodation (53.5%). Selinunte has the second highest population (22,210), a concentration of ethnic minority groups (30%), and rented accommodation (55.4%). Gallaratese too is very deprived in terms of youth unemployment and ageing, but is the area with the lowest number of migrants (4%) and the highest level of home ownership (82%).
But there are also housing estates such as San Siro, Comasina and Baggio with relatively low indices of deprivation in terms of unemployment, youth unemployment, and very low shares of rental accommodation (20%, 18%, and 32% respectively). This picture shows heterogeneity of socio-economic characteristics of large housing estates, despite their frequent amalgamation and homogenisation in images projected by the media. It is important to underline that the concentration of social problems is also due to long-term decay and the absence of work opportunities. Figures 12.5 and 12.6 at the ACE scale based on 2011 census data give indirect supplementary information on the relationship between ethnic segregation and the location of large housing estates.
Non-Italian residents are mainly located in the north of the city and in particular between the edges of the nineteenth-century city and the most distant intra-urban ring, with location quotients up to three times the city average (Fig. 12.5). The highest concentrations are recorded in Selinunte, a zone in north-west (Accursio, Villapizzone, Dergano and Affori) and a sector in the north-east around via Padova (Turro Crescenzago). This representation hides a more complex ethnic division of space due to specific patterns of each ethnic group. Groups with high levels of relative concentration are: Pilipino residents located in Turro and Padova with concentrations 3.6 times greater than the city average; Egyptians in Selinunte (6.0), who are also highly concentrated in Bovisa; and Dergano and Turro (five times the city average) (Fig. 12.6). Chinese residents have a more complex pattern, forming two clusters with high location indices: their central original settlement around via Sarpi (5.3) and more peripheral relative concentrations including Villapizzone, Bovisa, Dergano, Affori and Comasina-Bovisasca with concentrations four times the city average. The category 'other African' are relatively concentrated in Selinunte, Giambellino and Corvetto, with concentrations up to 5.1. Egyptians and 'other Africans' present their highest concentrations in two big housing estates: Selinunte and Giambellino.
Groups with lower levels of segregation are: Romanians, who exhibit a relatively scattered pattern in the most peripheral areas, with relatively low concentrations inferior to 2.3; Ecuadorians are dispersed in the city, having their highest concentrations in zones of peripheral estates such as Giambellino, Selinunte, and Viale Umbria (2.5), which are characterised by social deprivation; 'EU (Italian excluded)' are mainly located in the residential areas of the centre (3.2), with concentrations decreasing towards the periphery (with an exception of Musocco); and Peruvians (up to 3.0) are mainly dispersed in the city with two important concentrations in Bisceglie and Baggio.
As we observed earlier in Table 12.1, there is a substantial improvement of the housing conditions in Italy and Milan in particular from the situation described at the end of the 1970s by Garzena and Petsimeris (1984). But this relief represents an average density that hides severe housing problems for many fragile groups of the population (Padovani 1996) in a context of a shrinking welfare state and the end of the production of housing by the public sector for low-income groups (Tosi 1994;CARITAS 2008;Federcasa 2015). This is also due to the change of tenure regimes. In Milan, in the 1950s, only 23.6% of the population were owner-occupiers while in the last census their share tripled (74.3). This caused a dramatic decrease of rental accommodations offered for groups of population that are excluded from the market of home ownership. The situation has become more critical since the 1990s with an accentuation of privatisation of social housing and a decline in new production (Table 12.8).
Between 1991 and 2011, there was a 30,000-unit increase in the number of dwellings in Milan. This was not equally distributed to various of housing actors (owners). Private individuals increased by 90,000 their part, the state by 1,000, and cooperatives decreased by 5,000. The most dramatic decline was recorded by the public housing sector, which lost 27,000 units in two decades, thereby reducing its share of the total housing stock in Milan from 11 to 6%. The biggest loss occurred in the 1990s due to the national act promoting the privatisation of 50% of social housing.
From this section, it emerges that even in a context of urban demographic decline, the absence of new public housing production and the erosion of public housing through privatisation, it becomes more difficult for the most fragile social groups to have access to affordable housing.

Critical Points and New Challenges
From this empirical analysis of large housing estates in Milan, a number of points emerge. It is clear that the number of households is increasing faster than the number of dwellings. At the same time, smaller household size indicates the need for an increase in the number of smaller dwellings. The need for housing by economic groups who are excluded from the market (ownership and rent) persists and is increasing. Other groups in acute need of housing are: the elderly, single-parent families, the disabled, and people with special needs. While many The fact that many tenants are not paying their rent and charges for electricity, water and/or gas has resulted in physical decay and social conflicts within buildings. Non-payment of charges by illegal occupants adds to the economic problems of ALER or the local authority, further compromising the financing of management and maintenance functions. In interviews in San Siro conducted by the author with older residents, they described their frustration with the continuous worsening of their living conditions, and their growing sense of vulnerability and isolation. One resident stated: "In these buildings we also experience vertical segregation particularly when the lift is broken for days or weeks. Particularly the old have no access to 'earth' in order to shop or walk or to see friends. My daughter comes from the city to bring me food one or two times per week, walking up all these stairs".
Despite the variety of social housing in terms of size, conception, typologies and the social and economic characteristics of inhabitants, the media produce stereotypes of 'othering'. Although activists on estates have been able to harness parts of the media in order to make their case, their use of social media faces the problem of representation of large, diverse groups of people. The wider context of denationalisation of social housing policies and the erosion-through privatisation-of public housing also needs to be acknowledged. In the case of Milan, the main actors are the city council and ALER (former IACP). These two actors have different budgets. ALER has a more problematic economic situation, such that the two sets of tenants have different opportunities in terms of management and maintenance of their buildings and public space. Future work on these issues will have to overcome the fact that data on large estates is incomplete. While it remains difficult to source data on housing conditions and the characteristics of the inhabitants of large estates, the PIM study and the San Siro Mapping Project offers avenues worthy of exploration.
Another issue relates to the ageing of the housing stock. Most of the housing is between 50 and 100 years old. Some estates-like Selinunte (known also as the Quadrilatero) in San Siro-are experiencing physical decay, and have problems with utility provision (water, electricity). Rehabilitation options must be explored in more detail from a range of perspectives: social, economic and physical. In order to do this, more data is required. There are also problems of ghettoization, particularly in areas characterised by high rates of vacancy and decay, such as Selinunte, Grattosoglio and Giambellino. This has occurred due to the dramatic decrease in resources for investment in social housing. Current levels of finance for renewal are little more than drops in the ocean of Milan's decaying stock of large housing estates.
These physical, data availability and financial problems occur in the context of the latest round of migration. This has been mainly of low-income groups originating from the Global South. The majority of these migrants have no access to social housing and those who have access are directed to the less desired areas or ones with quotas. Networks that operate illegal routes to occupation operate in a context of desperate housing need for a large number of migrants. The incidence of illegal occupation has grown since the 1990s and affects Milan more than other Italian cities. Illegal occupation of the ALER housing stock became a frequent phenomenon affecting the most vulnerable segments of the stock: empty dwellings for reasons of turnover or delays in refurbishment (due to the lack of resources and/ or serious structural problems of the buildings). The gravity of the problem varies from estate to estate. In some areas, it spreads very rapidly from the ground floor to the upper floors. One measure of the severity of this problem is the space that it takes on the front page of ALER's website, warning-in Italian, English, French, Spanish and Arabic-that occupation is a crime. A further measure is the fact that eligibility for social housing in Lombardy is denied if the applicant has a record of illegal occupation. Concerning the stock of social housing in Milan, the highest level of occupation occurred in 2014 (1,720 dwellings). In 2016, the new city government announced that the level had been brought down to 2010 levels (i.e. 1,000 dwellings).
There is a very low rate of residential mobility in the large housing estates of Milan's periphery in comparison with the rest of the city (Petsimeris et al. 2015). This is due to the small number of units available. In contrast to the period when a significant proportion of the population vacated the estates-in part as a function of social mobility through education-during the period of austerity, social and residential mobility through employment has declined. Coupled with increases in rents and reduced availability of units for rent relative to the 1960s and 1970s, for a growing proportion of the population, it is currently almost impossible to live in Milan. New migrants cannot move in and the old migrants are caught in place.
A vicious circle linking fragmentation, lack of investment, deterioration, and illegal occupation distresses a high proportion of Milan's large housing estates. The economic crisis increased the problems of both the major actors in public housing and the most vulnerable part of the population. Further privatisation of the housing stock as a means of raising capital for renovation and management is not the best solution, particularly in a period of increasing housing shortage that also affects the middle classes. The non-profit sector can be a solution and some developments within the areas of housing estates such as Barona Village can improve the balance in terms of housing, services and social mix. However, due to the increasing difficulty of attracting private investment, these are likely to be exceptions. In the wider context of neoliberal policies attempting to transfer from the state to the private sector the role of welfare state provisions, social housing in Italy has arrived at this current impasse. Reforms in a number of policy areas, some directly and others indirectly related to housing, will be required in order for progress to be made. Secchi B (1972) Il settore edilizio e fondiario in un processo di sviluppo economico. In: Indovina F (ed) Lo spreco edilizio. Marsilio Editori, Padova, pp 3-46 Tafuri M (1986) Storia dell 'architettura italiana 1944-1985. Einaudi, Torino Tosi A (1994  Open Access This chapter is licensed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made. The images or other third party material in this chapter are included in the chapter's Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the chapter's Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder.