Project, Intervention Models and Cities

Interventions I, Rehousing with displacement - The PER

Shanty settlement in Algés, Oeiras, c. 1993. Source: Câmara Municipal de Oeiras

Shanty settlement in Algés, Oeiras, c. 1993. Source: Câmara Municipal de Oeiras

Slum clearance and rehousing with displacement


Slum clearance has its roots in Paris’ ‘Haussmanisation’ and in the late 19th century/early 20th century American and British clearance of tenement buildings that were part of broader normative programs for housing. It was the beginning of modern urbanism, where technical-scientifically defined ‘dwelling principles’ were materialised in the idea of universal ‘housing standards’ (Rowe 1993). It grew together with the ‘hygienist’ discourse on the city (Gandy 2004), giving birth to the ‘modern infrastructural ideal’ (Graham and Marvin 2001). State-imposed or state-led regeneration of cities was underpinned by ideas of universality; and notions of efficiency, science and progress (Rabinow 1989) were used in combination with new technologies of governmentality, surveillance and social control. The policy and practices of slum clearance programs often involved the imposed rehousing of working-class populations in new locations, both in the vicinities of the original ‘slums’ or in urban peripheries. In the UK, postwar modernist housing programs such as the London City Council’s were often developed in privileged areas of the city (Gold 2007), while French HLMs were built in the banlieue.


In the Portuguese-speaking landscape, classic examples include the favela removals in Rio de Janeiro from the 1950s to the 1970s, from the privileged areas of Lagoa Rodrigo de Freitas (e.g. the Praia do Pinto and Catacumbas favelas) to either nearby housing complexes such as Cruzada São Sebastião or far away housing estates in the undeveloped West Zone of Rio such as Cidade de Deus (Gonçalves 2016, Valladares 2005). After four decades of very different types of intervention in Rio’s favelas, from the Favela Brás do Pina upgrade program in the 1960s to State Governor’s Leonel Brizola regularization (land titling) program in the 1980s or the Favela Bairro in the 1990s, clearance and displacement made a comeback with the preparation of the 2014 World Cup and the 2016 Olympics. The forced removal of favelas near the Olympic venues to housing estates in the West Zone (Faulhaber and Azevedo 2015) was also made possible by a perverse implementation of the Minha Casa Minha Vida two-million homes federal housing program, which was devised by the Lula administration to finance publically-subsidized housing but also as a windfall for the construction sector.


Every case obviously had its historical specificities regarding the role urban poor populations had for the political regimes in place at each time, and these specificities must always be at the forefront of analysis. If Cruzada São Sebastião was something of a nod to large modernist housing estates and aimed at the social improvement of workers’ housing conditions near the city, solutions like Cidade de Deus came with the return of the ‘favela residents as undesirables’ theme that resurfaces every other decade or so, thus attempted to expel them to the outer periphery.  Brizola’s regularization efforts were underpinned by a Socialist developmental mindset that saw the agency of the working classes as key to a future developed state, while later Favela Bairro was less ambitious politically but came to mean a re-appreciation of the cultures of favela residents and an acknowledgement that they were established communities. And so on.


In Portugal, after a period in the mid-1970s when the urban poor were considered the recipients and makers of a revolution, and a program of social architecture was enacted for the areas where they lived (see Intervention Model III), there was a slightly contradictory period in the 1990s in which immigrant populations living in shanties were for the first time deemed worthy of attention from the state, though rehousing was carried out through the anachronic tenets of slum clearance and dispersion to peripheral ghettos.


Below I turn attention to this period and observe with more detail the implementation of the PER rehousing program in the Lisbon Metropolitan Area from 1993 to 2017, from a very specific angle, that of the location of rehousing within the city.

Bairro do Pombal housing estate, Oeiras, c. 1995. Source: Câmara Municipal de Oeiras

Bairro do Pombal housing estate, Oeiras, c. 1995. Source: Câmara Municipal de Oeiras

The PER in Lisbon


In the late 1980s, it was estimated that around 200,000 people lived in informal settlements in Portugal (AML, 1997; Ascensão 2015: 52). This was the peak of a long process of internal migration to the Porto and Lisbon metropolitan areas since the 1960s, and immigration to Lisbon since the mid-1970s from Portuguese-speaking African countries (former colonies) such as Cape Verde, Guinea-Bissau, Angola, Mozambique and São Tomé and Príncipe. The most vulnerable populations had been priced out of the housing market and left out of the eligible pool for the diminutive public housing system; in effect they had been ‘led’ to the interstices of the urban fabric to look for or build the accommodation they could not find or afford in the regular city. Informal settlements were then tacitly accepted by the state because of its inability to provide decent housing for everyone. (Salgueiro, 1977; Rodrigues, 1989; Malheiros and Vala, 2004; Nunes and Serra, 2004; Pinto 2009).


In 1993, the National Housing Institute initiated a slum eradication program for these areas the Plano Especial de Realojamento (Special Rehousing Program, from now on PER), in what was to become a significant process of urban restructuring. In Greater Lisbon, for instance, the PER involved the transfer of the population residing in 939 small and medium settlements to 220 dedicated public housing estates (IHRU 2015). In total, the program built around 35,000 new dwelling units (Santos et al 2015). However, in many cases the implementation of the PER was beset by the inexistence of public land near the original settlements needed to accommodate in situ rehousing. Coupled with municipal authorities’ financial constraints (despite central government covering 50% of the costs of building new housing estates), this resulted in the displacement of populations to far-away sites and the disruption of existing social and labour networks. Below I provide some glimpses of this process.


Scholarship on the PER


At the beginning of the program, a number of academic publications coming from LNEC’s Social Ecology Group focused on the social needs of households to-be-rehoused (Pinto 1994, Freitas 1994) but they were mostly preoccupied with guidelines for good rehousing and included elements of socio-ecological determinism (e.g. advising certain architectural features for particular populations such as Gypsies) that rendered them quite dated as the years passed. Immediately after, academic research centers such as CET-Centro de Estudos Territoriais or SociNova were contracted by different municipal governments to help (as consultants or prime movers) to survey the settlements, in order for implementation to be a precise administrative procedure. These collaborations ranged from sociological studies to rigid census-like surveying (e.g. Marques et al 2001, Dinamia CET 1992). At approximately the same time, rich urban histories of some of the oldest settlements with very particular histories, such as the then ‘drug quarter’ Casal Ventoso in Lisbon, were produced by young social science researchers (Chaves 1999). The PER also appeared as an explanatory factor in research on residential segregation.  For instance Malheiros and Vala (2004: 1079-80) noted a reduction in the segregation levels of immigrants from Portuguese Speaking African Countries in the 2001 Census and pointed to the PER as a possible explanation, but curiously their work also showed such population group was still over-represented in the ‘inhabiting slums or ‘inhabiting overcrowded dwellings’ categories, even after a good part of the PER estates had been built, leaving open the question of whether the program had in fact helped reduce segregation or, alternatively, reinforce it. Later, in the 2010s, critical studies pointing to the way the type of governmentality the program instituted (with non-participated decisions, the atomization of communities and even households, eligibility controversies, forced evictions, displacement to far-away sites, loss of labour and social networks) was experienced by target populations started to appear (Cachado 2013, Pozzi 2017), but by their own nature they focused on a limited number of sites and could thus be put aside as negative cases not illustrative of the whole program. Interestingly, no complete survey or record of where all the 938 informal settlements stood in the LMA, nor of the 220 housing estates built in order to rehouse people, was produced. In the following paragraphs I review some preliminary data from a research project that is currently developing such cartography.[1] The data regards just two municipalities but helps to shed an initial light on the way the PER can be argued to have been an integral part of a broader process of uneven urban restructuring.


The displacing effects of the PER


The PER involved both in situ rehousing (with apartment blocks built on the sites of the previous informal settlements) and clearance and re-housing in more peripheral, segregated and underserviced housing estates across the metropolitan area. For instance the Marianas-Adroana (in the municipality of Cascais), Azinhaga dos Besouros-Casal da Mira (Amadora) or Pedreira dos Húngaros-Moinho das Rolas (Oeiras) transfers were among the latter cases. A third set of controversial transfers involved the even more complicated cases where the squatted land was privately-owned and local governments failed to expropriate it in timely fashion: the original privileged locations were earmarked for private residential development for the white middle classes and the mainly black immigrant poor populations were displaced to scattered destinations, through the PER-Famílias or ad hoc arrangements.


[1] The data was collected and reworked for the research project ‘exPERts’, developed at the University of Lisbon, by Alessandro Colombo, Ana Ferreira, Ana Estevens and Cinzia Sofia. Many thanks to them. See

The Marianas settlement in 1995 (left), the Adroana peripheral housing estate (top right) and the redevelopment of Marianas-Carcavelos for the better off (bottom right) in 2017. Source: LNEC 1995.

The Marianas settlement in 1995 (left), the Adroana peripheral housing estate (top right) and the redevelopment of Marianas-Carcavelos for the better off (bottom right) in 2017. Source: LNEC 1995.

The question is to what degree this happened across the metropolitan scale and within municipal boundaries. How many people stayed on site, how many moved, where they moved to and what was the average distance covered with the transfer? There are other relevant elements to analyse the PER, but for now I focus on the most simple of indicators, location. Location is a clear indicator of urban restructuring processes that require the uprooting of particular communities from their living space, and thus a crucial element to measure urban dispossession. Furthermore, the centrality of location in terms of urban social justice was not lost on the experts and the housing milieu involved in the PER’s design as a public policy, and that was one of the reasons the legal figure which enacted the PER stated very clearly in its article 5, c) that City Councils had to:


“Assure that land currently occupied by shanty settlements to be demolished which is the property of, or is in possession of, municipal authorities and is zoned as residential [housing] should be directed to the program’s implementation or to the provision of controlled cost housing.” (Decree-Law 163/93)


It was a clear option for in situ solutions and a nod to the Portuguese derivative of the right to the city dating back to the post-revolutionary housing program SAAL, the right to the place (direito ao lugar) [See more, Intervention Models III]. Dislocation vs the right to stay put is not a perfect indicator to measure an entire process of urban restructuring. In abstract terms, spatial dislocation does not equate purely with dispossession, for instance in the cases where a) the second location is also a privileged one; b) has at least the potential for people to re-enact their original job and social networks in the new place of inhabitation; or finally c) if improved tenure status can balance the negative effects of displacement. However, monographic cases have established that for the PER such cases are but a minority.  Thus we take dislocation as a good proxy indicator for a complex process that stretches beyond mere travels across metropolitan space.


Many Xs, many Ys: dispersing communities to heterogeneous estates in Oeiras


The first important thing the cartography illuminates is a configuration of displacement patterns clearly contained in discrete geographical boundaries, those of the municipalities. Although obvious as the program was delegated to City Councils for its implementation, it still gives a remarkable image of how administrative and planning boundaries shape metropolitan-wide programs. The second element that stands out from the Oeiras case is how data, in GIS terms, amounts to a large number of Xs (origin) and a large number of Ys (destination).

Rehousing paths in the Oeiras municipality, more than 10 households. Source: Câmara Municipal de Oeiras. Map by Alessandro Colombo.

Rehousing paths in the Oeiras municipality, more than 10 households. Source: Câmara Municipal de Oeiras. Map by Alessandro Colombo.

What this means is that there was a break-up of established communities in a large majority of the original informal settlements (48% of settlements have 4 or more destinations, increasing to 80 % for those with more than 10 households and 93 % for those with more than 20 households – the exceptions being Leceia and Parque das Merendas), but also that housing estates have very mixed populations in terms of settlement of origin (out of the 24 housing estates, 14 (58%) have more than 10 origins, but the remaining ones are very small in number of dwellings). Among the largest estates, it is common for them to have population coming from 34 (Pombal), 33 (Páteo dos Cavaleiros), 25 (Moinho das Rolas), 25 (São Marçal), 22 (Encosta da Portela) or 21 (Francisco Sá Carneiro) different origins. It explains why for 71 informal settlements and 24 housing estates, 345 different transfers existed within the municipality (average of 4.9 destinations for each settlement).


In its simplest form, such type of process can be represented this way:


a                           A

b                           B


where ‘a’ and ‘b’ stand for informal settlements and capital letters for new estates. Instead of a and b going to A, or each going to their corresponding estate, both were dispersed; and both estates have a mixed population. This tallies with testimonies from municipal workers familiar with the PER in Oeiras, which admitted that ‘there was a deliberate policy of dispersal, as it was judged to be a positive feature’ (municipal officer, Jan. 2017). So what we have is a process of deep restructuring within municipal boundaries, combining displacement with dispersal. Take one of the largest settlements, Pedreira dos Húngaros. Its 602 households were rehoused in 17 different destinations across the entire municipality, with the three largest contingencies going to Navegadores (158 households), Páteo dos Cavaleiros (141), both within the deposit area for public housing in the municipality, Portela-Outurela; and the third (79 households) travelling 7 km West to the Moinho das Rolas estate, next to office park Lagoas Park.

The Pedreira dos Húngaros settlement and Miraflores (background), c. 1993. Source: CMO.

The Pedreira dos Húngaros settlement and Miraflores (background), c. 1993. Source: CMO.

The argument can be made that the displacement incidentally benefitted those households coming from Pedreira dos Húngaros, in the sense that the majority had working careers in the industries around Algés, such as fishing and dock work at Docapesca, and as these industries rapidly declined, they may have been moved to the middle of nowhere but it happened to be where they could find the only type of jobs their now obsolete skills allowed for – as security or cleaning personnel in the office park. However, it also illustrates a symptomatic solution that urban restructuring of this kind has for urban poor populations and it raises questions regarding the future labor trajectories of these populations, in the sense that being caught in a segregated, distant urban trap may equate to being caught in a poverty trap. Or put another way, how many of the children in the estate will ever work inside the offices of Lagoas Park, not cleaning and guarding them?

Buildings at Encosta da Portela, Outurela-Portela, Oeiras. Source: CMO

Buildings at Encosta da Portela, Outurela-Portela, Oeiras. Source: CMO

This last principle can also be applied to the two other larger contingencies of Pedreira dos Húngaros inhabitants, which were moved to Outurela-Portela, the deposit area for public housing in Oeiras. Although nearer the original settlement at the top of the privileged area of Miraflores (with seaview), Outurela-Portela features six different public housing estates from different eras (from the SAAL to municipal CDHs to the PER) amidst several industrial and warehouse facilities and, since the late 2000s, a large shopping mall and an Ikea store, all circumscribed by motorways and access ways, making it a confined and socially segregated area with difficult public transportation to the rest of the city. Yet in a socially selective way, the population of these housing estates can find low-qualified jobs in the retail and automobile compounds, and does.


To conclude, the Oeiras municipality, although often cited as a success case of the PER and having been quick to fully implement it, instituted a complex movement of people across its space, combining dispersion with displacement to new estates with a high concentration of poverty. From the viewpoint of populations, it involved three different things they had to adapt to quickly, without much of a say so: the uprooting from one location to another; the splintering of their established community; and insertion into a new setting with other poor contingencies coming from very different places.

All the 37 rehousing paths in the Almada municipality. Source: Almada City Council (CMA – Câmara Municipal de Almada). [manual georeferencing by Ana Estevens]

All the 37 rehousing paths in the Almada municipality. Source: Almada City Council (CMA – Câmara Municipal de Almada). [manual georeferencing by Ana Estevens]

A gradual approach: shorter displacement paths and keeping communities together in Almada


The case of Almada contrasts with Oeiras regarding its implementation model. In the municipality and according to current data, for 29 informal settlements and 12 housing estates, only 37 different transfers existed within the municipality, producing a much smaller average of 1.2 destinations for each settlement.


This means that Almada opted, with minor variations, to preserve the same populations upon rehousing. For instance the 114 households living in the Valdeão settlement near what today is a large hospital were all rehoused to the Monte da Caparica estate 500 metres away. Likewise, the 179 household living in Asilo 28 de Maio were all rehoused in Urbanização Nossa Senhora da Conceição, although this scheme was implemented by Casa Pia, a centenary welfare association. Other paths are longer, but overall there is a sense that displacement and especially dispersion were minimized. Furthermore, many estates are located amidst Pragal, Almada and Cova da Piedade’s functional areas, so people stayed near jobs and services.


In conclusion, based on this ongoing research the argument can be sketched that the urban restructuring of the LMA in the past two decades was for the urban poor populations affected by the PER an experience of major disruption, involving a) the uprooting and splintering of communities, b) the displacement to sites several kilometres away, and c) a re-insertion into areas with fellow previous slum dwellers from many different origins, essentially areas with a high concentration of poverty.


I will get back to the PER on this page at a later date.



AML – Área Metropolitana de Lisboa (1997), Caracterização do Programa Especial de Realojamento na Área Metropolitana de Lisboa, Lisboa, AML.

Ascensão, E. (2015). Slum gentrification in Lisbon, Portugal: displacement and the imagined futures of an informal settlement. In Lees, L., Shin, H.B. and López-Morales, E. (eds.) Global Gentrifications: Uneven development and displacement. Bristol: Policy Press, pp. 37-58.

Cachado, R. Á. (2013) O registo escondido num bairro em processo de realojamento: o caso dos hindus da Quinta da Vitória. Etnográfica17/3, 477-499.

CET (1992) Levantamento e Caracterização Urbanística dos Bairros de Barracas do Concelho de Loures (CET report n. 12). Lisboa: CET.

Chaves, M. (1999) Casal Ventoso: da gandaia ao narcotráfico – marginalidade económica e dominação simbólica em Lisboa, Lisboa, Instituto de Ciências Sociais.

Faulhaber, L., and Azevedo, L. (2015). SMH 2016: remoções no Rio de Janeiro Olímpico. Mórula Editorial.

Freitas, M. J. (1994) Os paradoxos do realojamento, Sociedade e território, 20, 26-35.

Gold, J. R. (2007). The practice of modernism: Modern architects and urban transformation, 1954–1972. Routledge.

Gonçalves, R. S. (2016). Favelas do Rio de Janeiro: história e direito. Rio de Janeiro, Editora PUC.

Gandy, M. (2004). Rethinking urban metabolism: Water, space and the modern city, City, 8/3, 363–379.

Graham, S. and Marvin, S. (2001) Splintering urbanism: networked infrastructures, technological mobilities and the urban condition. London, Routledge.

IHRU (2015b). Information request to national housing institute (Instituto da Habitação e da Reabilitação Urbana), July 2015.

LNEC (1995) Bairro das Marianas. Pré-Diagnóstico. Relatório 354/95. Lisboa: LNEC.

Malheiros, J. M. and Vala, F. (2004) Immigration and city change: the region of Lisbon in the turn of the 20th century, Journal of Ethnic and Migration Studies, 30/6, 1065-1086.

Marques, M. M., Santos, R., Araújo, F. and Nóbrega, S. (2001) Realojamento no Concelho de Oeiras: um estudo sociológico. Oeiras: CMO.

Nunes, J. A. and Serra, N. (2004), Decent housing for the people: urban movements and emancipation in Portugal, South European Society & Politics, 9-2, 46-76.

Pinto, P. R. (2009), Housing and citizenship: building social rights in twentieth century Portugal, Contemporary European History, 18-2, 199-215.

Pinto, T. C. (1994) A Apropriação do Espaço em Bairros Sociais: o gosto pela casa e o desgosto pelo bairro, Sociedade e território, 20, 36-43.

Pozzi, G. (2017) Cronache dell’abitare. Pratiche di costruzione informale e rialloggiamento forzato nel quartiere Santa Filomena (Lisbona). Antropologia, 4/1, 49-69.

Rabinow, Paul (1989). French modern: norms and forms of the social environment, Cambridge, MIT Press.

Rodrigues, C. M. (1989), Eficiência e equidade na produção de espaço clandestino, in Gaspar, J. (ed.), Clandestinos em Portugal – Leituras, Lisboa, Livros Horizonte.

Rowe, P. (1993) Modernity and Housing, Cambridge, MIT Press.

Salgueiro, T. B. (1977), Bairros clandestinos na periferia de Lisboa, Finisterra - Revista portuguesa de geografia, XII – 23, 28-55.

Santos, A. C., Teles, N. and Serra, N. (2014). Finança e Habitação em Portugal. Coimbra, Centro de Estudos Sociais.

Valladares, L.P. (2005). A invenção da favela. Rio de Janeiro, Editora FGV.