This is the old United Nations University website. Visit the new site at http://unu.edu
9. Rio de Janeiro: Urban expansion and structural change
Introduction
Population
growth
Employment
Poverty and the distribution of income
Housing
and infrastructure
Health
and education
Transport
Pollution and environmental policies
Crime
Emerging issues for the coming decade
Acknowledgements
Notes
References
Hamilton Tolosa
Until the mid-1970s, the Brazilian economy grew consistently at an annual rate of slightly over 6 per cent. At the beginning of the 1980s, economic conditions changed radically. Falling rates of investment led to a widespread recession, and a distorted price structure completely upset business expectations and disorganized national production. A series of unsuccessful stabilization experiments distorted prices further and increased social inequalities.
National economic problems had a significant impact on the urban system. During the boom, the major cities were expanding but industrial deconcentration in the hinterland of these cities led to the rapid growth of a number of secondary cities in the south and south-east of the country. During the 1980s, urban poverty increased markedly, both in the largest cities and in small cities based in backward agricultural regions or dependent upon consumer-goods industries. With the recession, many small industrial companies simply vanished.
This chapter considers how the Rio de Janeiro metropolitan area (RJMA) fared during the recession of the 1980s. How did its national urban role change and how was the quality of life in the city affected?
These questions are considered through an examination of the dynamics of population change, employment, poverty and income distribution, housing, supply of services, transport provision, pollution and environmental policies, and crime. The final section considers the city's future prospects.
Rio de Janeiro was formally constituted into a metropolitan area (RJMA) in 1973. The RJMA contained fourteen counties: Rio de Janeiro, Duque de Caxias, Itaborai, Itaguaí, Magé, Mangaratiba, Maracá, Nilópolis, Niterói, Nova Iguaçu, Paracambi, Petrópolis, São Gonçalo and São João de Meriti.1 With around 10 million inhabitants, the metropolitan area is Brazil's second largest city and its second most important port. It is located some 400 kilometres north of its greatest national rival, São Paulo. Together, these two cities contain over 25 million people and almost half of Brazil's manufacturing activity. They dominate the south-east, the most prosperous region in Brazil, and in recent years have maintained their combined share of the national population.
In contrast to São Paulo, however, Rio's economic situation has been in decline for some years. Its economic future was damaged when the federal capital was moved to Brasília in 1960, along with a huge amount of public investment. Rio has also suffered badly from the Brazilian economic recession and has been losing out in the struggle with São Paulo for commercial and industrial dominance. In 1985, São Paulo accounted for 26 per cent of the country's manufacturing production compared to Rio's share of 7 per cent. Many of Rio's leading banks, industries, and research and development companies either relocated or moved their headquarters to São Paulo. Earnings from tourism also declined as the media drew international attention to Rio's escalating crime rate. As a result, an increasing gap opened up between the two largest metropolitan areas. Since 1970, the population of São Paulo has grown nearly twice as fast as that of Rio. In 1988, average household earnings per capita were 22 per cent higher in São Paulo than in Rio; in 1970 the difference had been only 10 per cent; in 1976, 18 per cent. This was not just a relative decline. As a result of the national recession, Rio's population became much poorer. Between 1976 and 1988 real earnings in Rio de Janeiro fell by 29 per cent.
Over the last thirty years, Brazil's urban population has grown at an average annual rate of just over 5 per cent. In the 1960s and early 1970s, the fastest rates of growth were recorded by cities with between 100,000 and 250,000 inhabitants. By the 1980s, although the economies of many of these secondary centres were continuing to prosper, their population growth rates slowed. The fastest urban growth rates were now to be found among the large metropolitan centres, particularly those in the north-east of the country.
Table 9.1 Metropolitan growth in Brazil, 1970-1991
Population (000s) |
Annual growth (%) |
||||
Metropolitan Area | 1970 | 1980 | 1991a | 1970/80 | 1980/91 |
Rio de Janeiro | 6,891 | 8,872 | 9,600 | 2.4 | 0.8 |
Capital city | 4,252 | 5,091 | 5,336 | 1.8 | 0.4 |
Periphery | 2,639 | 3,681 | 4,264 | 3.4 | 1.3 |
São Paulo | 8,139 | 12,588 | 15,199 | 4.4 | 1.7 |
Capital City | 5,924 | 8,493 | 9,480 | 3.6 | 1.0 |
Periphery | 2,215 | 4,095 | 5,719 | 6.3 | 3.1 |
Southern citiesb | 4,053 | 6,334 | 8,451 | 3.8 | 2.7 |
Capital Cities | 2,729 | 3,929 | 4,600 | 3.7 | 1.4 |
Peripheries | 1,324 | 2,405 | 3,851 | 6.2 | 4.4 |
Northern citiesc | 4,629 | 6,692 | 8,959 | 3.8 | 2.7 |
Capital Cities | 3,558 | 4,942 | 6,350 | 3.3 | 2.3 |
Peripheries | 1,071 | 1,750 | 2,609 | 5.0 | 3.7 |
Brazil | 93,165 | 119,002 | 146,154 | 2.5 | 1.9 |
Source: Demographic Census, 1970,1980, and 1991.
a. Preliminary results.
b. Includes the following metropolitan areas: Belo Horizonte,
Curitiba, and Pôrto Alegre.
c. Includes the following metropolitan areas: Belém, Fortaleza,
Recife, and Salvador.
Note:
Statistical information throughout this paper was obtained from two major sources: the population censuses for 1970, 1980, and 1991 (preliminary results) and the National Household Survey, which is conducted annually. The National Household Survey covers the nine metropolitan areas and the federal capital, Brasilia. Unfortunately, the sampling design prevents use of National Household Survey data at the sub-metropolitan scale. Comparison of the municipality of Rio and the surrounding counties has been based on either census information or independent surveys.
Table 9.1 compares the rates of population growth in Rio and São Paulo between 1970 and 1991 with those in Brazil's other seven metropolitan areas.2 It shows that the pace of population growth slowed markedly in the 1980s in all nine areas, partly the result of a slowing of natural increase in Brazil as a whole and partly the result of economic recession. However, the falls in the growth rates of Rio de Janeiro and São Paulo were much more marked than those of the other cities. During the 1980s, Rio's population grew at 0.8 per cent per annum, São Paulo's at 1.7 per cent.
Table 9.2 Metropolitan Rio de Janeiro: Migrants and natives, 1980
RJMA | Rio de Janeiro | Periphery |
|||
Mean | Maximum | Minimum | |||
Migrants/total population (%) | 42.2 | 35.2 | 51.5 | 59.1 | 9.4 |
Recent migrants/total migrants (%) | 14.8 | 10.0 | 21.0 | 31.1 | 10.6 |
Source: State of Rio de Janeiro Yearbook, 1988.
Table 9.1 also shows that the population of the central city has been expanding less quickly than the periphery in every metropolitan area of Brazil (see chapter 4). However, this process is much more advanced in Rio and São Paulo and is reflected in the large differential in those cities between the central and peripheral growth rates.
Over the years, migration has been a significant factor in Rio's growth and, in 1980, migrants made up more than 40 per cent of the total population. Research has shown that 70 per cent of migrants to metropolitan centres in Brazil originate from urban areas (Pastore and Haller, 1993; Tolosa, 1976). Most arrivals are first absorbed into either the construction industry or the service sector and eventually move to the suburbs. Table 9.2 shows the location of migrants to Rio de Janeiro in 1980. Most of these migrants were living in municipalities outside but relatively close to the central area: Nilópolis, São João de Meriti, Nova Iguaçu, São Gonçalo, and Caxias (figure 9.1). More distant towns, such as Petrópolis, with its pleasant site in the mountains more than 80 kilometres from downtown Rio, its economic base centred upon tourism and fairly sophisticated, clean industries, and its reliance on skilled labour, attracted fewer migrants. Among recent migrants there has been a stronger tendency to move into municipalities on the eastern banks of Guanabara Bay, notably Itaborai (31 per cent) and Magé (26 per cent).
During the 1980s, Rio's population growth slowed right down. The decline is explained by Brazil's economic recession and the slowing of metropolitan growth throughout the country. Fewer migrants moved to the major cities, a trend particularly marked in Rio owing to the latter's especially serious economic problems (see next section). But the slower pace of growth was also due to longer-term demographic trends. As table 9.3 shows, fertility rates plummeted in Rio between 1970 and 1988. And, while life expectancy increased, the rate of change was far less marked. As a result, there was a substantial fall in the rate of natural increase.
Figure 9.1 Rio de Janeiro: Metropolitan area
Table 9.3 Metropolitan Rio de Janeiro: Major demographic characteristics, 1970-1988
Indicators | 1970 |
1980 |
1988 |
||||
Rio de Janeiro | National | RJMA | Rio de Janeiro | National | RJMA | National | |
Fertility | 3.5 | 5.7 | 2.8 | 2.4 | 4.3 | 2.1 | 3.5 |
Infant mortality | - | 117 | 77 | 68 | 88 | 36 | 63 |
Life expectancy | 62.1 | 53.4 | 63.2 | 65.2 | 60.1 | 65.6 | 64.8 |
Household size | 4.4 | 5.2 | 3.9 | 4.5 | 3.9 | 4.3 |
Sources: Demographic Census, 1970 and 1980; National Household Survey, 1988.
Notes: Fertility rate per 1000 women; infant mortality rate per 1000 live births; life expectancy rate in years; density in persons per square kilometre; household size = number of persons.
Between 1970 and 1988, employment in the Rio metropolitan area increased annually at 3.8 per cent. This was well above the 2.9 per cent annual growth in the economically-active age group (over 10 years old), so that participation rates rose markedly, from 45 per cent in 1970 to 52 per cent in 1988. Nevertheless, participation rates in Rio remain lower than the national urban average of 55 per cent because participation has been rising quickly in most metropolitan areas.
Despite the slackening pace of Rio's economic growth, the pace of employment growth hardly changed. Between 1970 and 1980, the latter grew annually at 4.0 per cent; between 1980 and 1988 by 3.9 per cent.
Table 9.4 shows most workers to be employed in the service sector and indicates how the dominance of this sector has increased during the years of decline. The national recession has brought a slight decline in employment in both construction and manufacturing.
The increase in service activity signifies a decline in the proportion of the population with a formal contract. Few workers in personal or domestic services have social security coverage or, for that matter, any sort of legal protection or unemployment insurance. This group constitutes a substantial proportion of the "unprotected" segment of the metropolitan labour force. In 1988, about one-third of Rio's 3.5 million workers lacked a formal contract, with 800,000 lacking any kind of social security cover. These figures compare very unfavourably with the situation in São Paulo, where unprotected workers accounted for only 22 per cent of the labour force. Unprotected and selfemployed workers have much lower average earnings than wage earners, the latter earning 4.5 times the minimum wage on average, unprotected workers only 1.7 times the minimum, and the self-employed 3.7 times the minimum.
Table 9.4 Rio de Janeiro: Economically active population by productive sector, 19821988 (percentage of population)
Sector | 1982 | 1985 | 1988 |
Industry | 16.4 | 16.0 | 15.4 |
Construction | 8.9 | 7.6 | 7.4 |
Services | 67.5 | 70.0 | 75.0 |
Personal and domestic | (23.6) | (25.0) | (25.3) |
Business services | (3.8) | (4.5) | (4.7) |
Public administration | (11.5) | (11.4) | (12.5) |
Other | 7.2 | 6.4 | 2.2 |
Total | 100.0 | 100.0 | 100.0 |
Source: National Household Survey, various years.
During the 1970s, although the age of the average worker remained constant at around 35 years, the labour force recruited increasing numbers of people at both top and bottom of the age pyramid. First, more young workers entered the job market as large numbers of young in-migrants and babyboomers looked for work. Second, with falling death rates, rising numbers of elderly people were also looking for employment. As education levels in Rio rose, however, the labour force became better educated, the average number of years of formal education rising from 5.9 in 1970 to 7.3 in 1988.
Perhaps the most dramatic change in Rio's labour force, however, has been its increasing feminization. The female participation rate rose from 27 per cent in 1970 to 38 per cent in 1988. The scattered empirical evidence suggests that female employment rose during the "miracle" years, as employers sought increasing numbers of cheap workers, and continued during the recession of the 1980s, as women and children were called upon to offset reductions in family income due to unemployment and falling real wages.
Table 9.5 Employment characteristics of the poora (percentages)
Informal labour | Under-employment | Open unemployment | Participation | |||||
Indicatorsb | 1981 | 1985 | 1981 | 1985 | 1981 | 1985 | 1981 | 1985 |
Rio de Janeiro | 46.5 | 54.3 | 20.6 | 18.6 | 16.9 | 9.2 | 41.2 | 48.6 |
São Paulo | 39.8 | 46.2 | 14.4 | 14.7 | 19.5 | 14.0 | 44.1 | 48.6 |
Nine metropolitanareas | 45.7 | 55.9 | 18.1 | 20.0 | 9.6 | 6.8 | 42.3 | 47.3 |
Sources: National Household Survey, venous years; Rocha, 1990.
a. Refers to the population below the poverty line.
b. Informal labour includes those workers and wage earners not
contributing to social security; underemployment is defined as
those working less than 40 hours per week; open unemployment
measures the proportion of poor workers looking for a job; and
participation rates are calculated as a percentage of the poor
population over 10 years of age.
Table 9,5 implicitly supports this interpretation by showing employment trends during the early 1980s among workers living below the poverty line. It reveals how unemployment fell during a period of rapidly rising labour participation in both Rio and São Paulo. Most of the rise in employment was absorbed by the informal sector.
The rise in both participation rates and the size of the informal labour force is a strong sign of deteriorating employment conditions. Combined with evidence of increasing levels of female and child labour, it is a clear sign of the terrible price paid by poor families during the economic crisis throughout metropolitan Brazil (table 9.5).
Poverty and the distribution of income
Both absolute and relative poverty got worse in Brazil during the 1980s as a result of the economic recession and the state's unsuccessful efforts to tackle inflation. Relative inequality improved slightly during the 1970s and deteriorated during the 1980s. Table 9.6 shows that inequality worsened during the 1980s in Brazil's three largest metropolitan areas. It also shows that inequality was greater in Rio than in either São Paulo or Belo Horizonte in both 1981 and 1989. Indeed, the situation in Rio has deteriorated faster than in any other metropolitan area. By 1989, Rio had the most unequal distribution of income of any metropolitan area in Brazil; during the 1980s it had displaced even the cities of the north-east from first place.3
Table 9.6 Relative poverty in metropolitan areas of south-east Brazil, 1981 and 1989 (Gini coefficients)
City | 1981 | 1989 |
Rio de Janeiro | 0.58 | 0.67 |
São Paulo | 0.52 | 0.57 |
Belo Horizonte | 0.57 | 0.62 |
Source: Rocha, 1991.
Note: The Gini coefficients were calculated on the basis of per capita family income.
Table 9.7 Absolute poverty in metropolitan areas of south-east Brazil, 1981 and 1989
Metropolitan Areas | Poverty lines (US$) |
Poverty shares (%) |
||
1981 | 1989 | 1981 | 1989 | |
Rio de Janeiro | 44.8 | 44.8 | 27.2 | 32.5 |
São Paulo | 52.6 | 53.5 | 22.0 | 20.9 |
Belo Horizonte | 40.7 | 38.8 | 31.1 | 27.2 |
Source: Rocha, 1991.
Notes: Per capita monthly family income converted into dollars using average annual exchange rate. The methodology employed in calculating the poverty lines is described fully in Rocha, 1988.
Absolute poverty also increased dramatically during the 1980s and, in 1991, some 3.5 million people were living below the poverty line. According to table 9.7, however, the increase in poverty was reflected more in the rising proportion of poor families than in the level of immiseration. The average income of families living below the poverty line in Rio remained constant between 1981 and 1989. The pattern was very similar in São Paulo and Belo Horizonte. Like relative poverty, however, absolute poverty appears to be far worse in Rio than in the other two cities. São Paulo has a far lower share of its people living below the poverty line.4 In fact, Rio had a higher proportion of its population living in absolute poverty than the Brazilian metropolitan average of 28.1 per cent.