Contents - Previous - Next

This is the old United Nations University website. Visit the new site at

4. Land, housing, and infrastructure in Latin America's major cities

Housing conditions
The price of land
Residential segregation
Population density
The changing shape of the city

Alan Gilbert

Since 1950, Latin America's major cities have grown dramatically. The combined population of today's four mega-cities (Mexico City, São Paulo, Buenos Aires, and Rio de Janeiro) increased from some 13 million in 1950 to around 60 million in 1990 (UNDIESA, 1991). If we also include the three cities which have more than four million people (Bogotá, Lima, and Santiago), nearly 60 million additional people were added to Latin America's metropolitan population in forty years. This chapter is concerned with the effects of this massive population expansion on housing, servicing, and land use.

My general argument is that Latin America's major cities have coped with the pressures of rapid population growth extremely well. Housing and servicing provision has generally, if not universally, kept up with the pace of urban expansion. This does not mean that people live well, or that most people are well housed or serviced, or that the major cities are free of severe problems; it does mean that there is little sign of deterioration. Only in a couple of cities can a case really be made that living conditions are getting worse.

The problems facing Latin America's major cities are hardly peculiar to them. Indeed, they have begun to look more and more like the cities of the developed world. In many respects, "developed" and "less developed" cities have been converging in form. Many cities in Europe and North America now contain stereotypical features of the Latin American metropolis: homelessness, unemployment, and a burgeoning "informal sector." Similarly, Latin American cities have begun to look much more like "developed" cities, sprouting skyscrapers and adopting the car-based, privatized culture of the North American and Western European city. Latin America's major cities increasingly share the same urban problems as major cities in most developed countries. Such problems are the result of economic recession combined with the unfettered workings of capitalist land and housing markets, the rolling back of the state, and the unlimited use of the private car. Latin American cities are different only in so far as they are generally poorer than cities in the developed world and also because they suffered much more seriously from economic decline during the 1980s.

Table 4.1 The growth of self-help housing in selected Latin American cities

City Year City population Population in squatter settlements Percentage
(000s) (000s)
Rio de Janeiro 1947 2,050 400 20
1957 2,940 650 22
1961 3,326 900 27
1970 4,252 1,276 30
1991a 9,696 921 10
Mexico City 1952 2,372 330 14
1966 3,287 1,500 46
1970 7,314 3,438 47
1976 11,312 5,656 50
1990 15,783 9,470 60
Lima 1956 1,397 112 8
1961 1,846 347 17
1972 3,303 805 24
1981 4,608 1,455 32
1989 6,234 2,338 38
Buenos Aires 1956 6,054 109 2
1970 8,353 434 5
1980 9,766 957 10
1991b 10,911 659 5
Caracas 1961 1,330 280 21
1964 1,590 556 35
1971 2,200 867 39
1985 2,742 1,673 61
1991 2,966 1,238 42
São Paulo 1973 6,561 72 1
1980 8,493 321 5
1985 8,929 440 6
1987 9,109 813 8
1991 9,483 1,050 9
1989c 10,436 3,238 31
Bogotá 1955 917 367 40d
1965 1,782 766 43d
1975 3,069 921 30d
1985 4,123 1,278 31d
1991 4,824 1,254 26e

Sources: Gilbert, 1994: 82, which was based on a variety of primary and secondary sources, supplemented by Jacobi, 1990: 37; Kross, 1992: 154; IBGE, 1991; Molina et al., 1993: 153; UNCRD, 1994; and Villanueva and Baldó, 1994.

a. The figure for 1991 is based on the favela population. It therefore excludes consolidated seff-help housing.
b. Changed basis of calculation.
c. Except for the 1989 figure, the figures for São Paulo record the proportion of the population living in favelas. They therefore underestimate the total population living in self-help housing. The 1989 figure records the proportion living in favelas and in "precarious housing." Figures are for the municipality of São Paulo.
d. Estimates based on the area of land developed by decade since 1935. Population figures estimated by author.
e. Changed basis of calculation.

This chapter deals with a number of related issues. First, it discusses how effectively the huge growth in Latin America's metropolitan population has been accommodated: the way in which such vast numbers of people have been housed and the extent to which servicing has managed to keep up with demand. Second, it considers trends in residential structure: has segregation been increasing or decreasing and how has its form changed? Third, the chapter examines urban form: to what extent have Latin American cities developed polynuclear urban forms, a tendency that has been widespread among urban agglomerations in the developed world? What is happening to the inner city; is this expanding or contracting, commercializing or gentrifying?

The chapter includes discussion of four cities which have not so far achieved mega-city status: Bogotá, Caracas, Lima, and Santiago. Their inclusion has the advantage of showing how extreme size affects, and fails to affect, the process of urban growth. It also allows a wider breadth of experience to be addressed. Perhaps surprisingly, it also improves the data set, since some of these smaller cities have better information than some of the mega-cities.

Contents - Previous - Next