located along the reeking banks of the heavily polluted Río de la Reconquista and Río de la Matanza. “Stagnant water and untreated sewage,” writes geographer David Keeling of a visit to a typical villa along the Río Reconquista, “created an overpowering stench, and the entire area was overrun with rats, mosquitos, flies, and other insects.” The villas are tolerated only because such brownfield sites are temporarily worthless in a depressed economy.73 Likewise, in Caracas precarious squatter ranchos continue to inch their way up rugged and landslide-prone mountain slopes that no sane developer would ever consider to be marketable real estate. Squatting has become a wager against inevitable disaster.
But flat peripheral land, even desert, has market value, and today most low-income settlement on the urban edge, although often characterized as squatting, actually operates through an invisible real estate market.74 This “pirate urbanization” was carefully studied for the first time by the World Bank’s Rakesh Mohan and his research team in Bogotá at the end of the 1970s:
… these pirata subdivision settlements did not result from land invasions: the land has actually changed hands through legal purchases. It is the subdivision itself that is usually illegal. But these settlements are better described as extralegal rather than illegal. Low-, lower-middle-, and middle-income families, having been shut out of the formal housing market, buy lots from entrepreneurs who acquire tracts of undeveloped land and subdivide them without conforming to zoning laws, subdivision regulations, or service provision standards. The lots sold usually provide only a bare minimum of services, often nothing more than some streets and water standposts. Typically, this rudimentary infrastructure is incrementally upgraded after initial settlement has taken place.75
Pirate urbanization is, in effect, the privatization of squatting. In an important 1990 study, housing experts Paul Baross and Jan van der Linden characterized pirate settlements or “substandard commercial residential subdivisions” (SCRSs) as the new norm in poor people’s housing. In contrast to true squatters, the residents of a pirate subdivision have obtained either a legal or de facto title to their plot. In the case of a legal title, the subdivider is usually a speculator, a latifundista or large farmer, a rural commune (for example, a Mexican ejido), or customary entity (such as a Bedouin tribe or village council). The landowners – as in the case of an asentamiento in suburban Buenos Aires discussed by David Keeling – may even encourage residents to organize themselves as a land invasion in the shrewd expectation that the state will be forced to guarantee eventual compensation as well as infrastructural development.76
In the second case of de facto tenure, the land is usually state-owned but settlers have purchased a guarantee of tenure from powerful politicians, tribal leaders, or criminal cartels (for example, the Triads, who are the major informal property developers in Hong Kong).77 Another notorious example are Karachi’s dalals, whom Akhtar Hameed Khan, the founder of the famed Orangi Pilot Project, describes as “private entrepreneurs who have learnt the art of collaborating with and manipulating our greedy politicians and bureaucrats. With their costly patronage, the dalals secure possession of tracts of [public] land, buy protection against eviction, and obtain water and transport facilities.”78 The dalals (the word can mean “pimp” as well as “middleman”) dominate the katchi abadis – the pirate subdivisions like Orangi – that house almost half of Karachi’s population.79
Although the actual houses are almost always formally unauthorized by local government, pirate subdivisions, unlike many squatter camps, are generally subdivided into uniform lots with conventional street grids; services are rudimentary or non-existent, however, and the selling price is based on residents’ ability to bootleg or negotiate their own infrastructural improvements. “In short,” write Baross and van der Linden, “planned layouts, low service levels, suburban locations, high-tenure security, non-conformity with urban development plans, and self-help housing are the generic features of SCRSs.”80 With appropriate local wrinkles, this definition characterizes edge development in Mexico City, Bogotá, São Paulo, Cairo, Tunis, Harare, Karachi, Manila, and hundreds of other cities – including, in the OECD bloc, the clandestinos around Lisbon and Naples as well as the recent colonias outside El Paso and Palm Springs.
In some countries the commercialization of peripheral slum development has existed for decades. “By the mid-1960s,” explains urban planner Ayse Yonder, “squatting in the traditional sense of the term had disappeared in Istanbul. Settlers had to pay local strong men for the right to occupy even public land. In the mid-1970s, entrepreneurs with underground connections started controlling public lands in certain districts of Istanbul, selling land and monopolizing all construction activity.”81 In Nairobi – today a city of rack-rented poor tenants – full-fledged commercialization took off in the early 1970s as wealthy outsiders discovered that squatting was creating a new land market with huge windfalls from legalization. Landowners (often successors to the original Asian owners) began to peddle unauthorized subdivisions. According to poverty researcher Philip Amis, “they in effect invaded their own land, building unauthorized housing according to their own plans … the risk paid off handsomely. No demolition orders were issued and returns on investment were very high.”82
3. Invisible Renters
As a rule of thumb, both the popular and scholarly literatures on informal housing tend to romanticize squatters while ignoring renters. As World Bank researchers recently acknowledged, “remarkably little research has been done on low-income rental markets.”83 Landlordism is in fact a fundamental and divisive social relation in slum life worldwide. It is the principal way in which urban poor people can monetize their equity (formal or informal), but often in an exploitative relationship to even poorer people. The commodification of informal housing has included the rapid growth of distinctive rental subsectors: infill developments in older shantytowns, or multifamily constructions in pirate subdivisions. To be sure, most of the urban poor in West Africa have always rented from landlords, as have a majority of residents in Dhaka and some other Asian cities (in Bangkok two-thirds of “squatters” actually rent the land they build their shacks upon).84 Renting has also become far more common than usually recognized in the peripheries of Latin American, Middle Eastern and South African cities. In Cairo, for example, the more advantaged poor buy pirated land from farmers, while the less advantaged squat on municipal land; the poorest of the poor, however, rent from the squatters.85 Likewise, as urban geographer Alan Gilbert observed of Latin America in 1993, the “vast majority of new rental housing is located in the consolidated self-help periphery rather than in the centre of the city.”86
Mexico City is an important case in point. Despite a Model Law of the colonias proletarias which sought to ban absentee ownership, “poaching,” and speculation in low-income housing, the Lopez Portillo government (1976–82) allowed slum-dwellers to sell their property at market rates. One result of this reform has been the middle-class gentrification of some formerly poor colonias in good locations; another has been the proliferation of petty landlordism. As sociologist Susan Eckstein discovered in her 1987 return to the colonia that she had first studied fifteen years earlier, some 25 to 50 percent of the original squatters had built small, 2-to-15-family vecindades which they then rented to poorer newcomers. “There is, in essence,” she wrote, “a twotiered housing market, reflecting socioeconomic differences among colonos.” She also found “a ‘downward’ socioeconomic leveling of the population since I was last there…. The poorer tenant stratum has increased in size.” Although some older residents had thrived as landlords, the newer renters had far less hope of socioeconomic mobility than the earlier generation, and the colonia as a whole was no longer a “slum of hope.”87
Renters, indeed, are usually the most invisible and powerless of slum-dwellers. In the face of redevelopment and eviction, they are typically ineligible for compensation or resettlement. Unlike tenement-dwellers in early-twentieth-century Berlin or New York, moreover, who shared a closeknit
