Journal Issue
Share
Article

Bringing the poor into the growth process: the case of Brazil - Making the poor more productive makes economic sense

Author(s):
International Monetary Fund. External Relations Dept.
Published Date:
December 1981
Share
  • ShareShare
Show Summary Details

Petehr T. Knight and Ricardo Morán

Up to now, Brazil’s extraordinary economic progress—which led to real growth rates of 8.5 per cent a year between 1965 and 1980—has sprung from a strategy to build an economy modeled on those of the Western industrial nations. This approach, which has been followed by many other middle-income countries, usually with less success, has brought material well-being to a considerable portion of the population. But serious social problems persist. Malnutrition, high infant mortality, and lack of access to basic public services are still experienced by a large proportion of Brazilians.

Since 1974, the Brazilian Government has been investing more in basic public services and has initiated programs that both alleviate material deprivation among the poor and increase their productivity. This article discusses the nature of that deprivation and how the effort to reduce it may be seen as part of a coordinated development strategy that makes good economic as well as social sense. The underlying rationale is that a program of reform designed to satisfy the latent demand of Brazil’s millions of poor families for basic public services and wage goods could expand the internal market, cushion the adjustment of the economy to higher energy prices and lower growth, and still create more jobs. Meeting the basic needs of Brazil’s entire population by the end of this century is not essentially an economic or financial problem. The binding constraints are more likely to be organizational, staffing, and logistical bottlenecks that could retard the pace of extending services, especially into the rural areas.

The strategy is relevant for other medium-income countries with market income inequalities and relatively underdeveloped human resources—such as Algeria, Mexico, Peru, and Turkey, and a number of other Latin American countries. Such countries have the financial capabilities to invest more in the poor and have the organizational skills necessary to carry out such programs, while at the same time expanding the internal market for wage goods, most of which can be produced domestically.

Pattern of growth

Despite Brazil’s rapid rate of population increase, per capita gross domestic product (GDP) growth over the past 15 years has been impressive, averaging 5.7 per cent. With a gross national product (GNP) equivalent to over US$230 billion (roughly $1,940 per capita) in 1980, Brazil’s economy is now the world’s tenth largest, roughly on a par with that of the People’s Republic of China or Canada. Brazil is now the third largest agricultural exporter and the tenth largest producer of automotive vehicles.

A hint of vulnerability in the country’s economic performance has been its lagging primary energy production, which has grown at a rate one fourth less than that of energy consumption between 1967 and 1980 (see Table 1). The shortfall has been made up by increasing imports of petroleum and, to a much lesser extent, coal. Since 1973, petroleum-fueled growth has been maintained only at the cost of massive foreign borrowing and accelerating inflation. By 1980, petroleum imports, accounting for 83 per cent of domestic petroleum consumption, absorbed almost 50 per cent of total export earnings. Servicing the foreign debt required another 60 per cent. Meanwhile domestic savings rates fell in both the private and public sectors. Incentives to save were eroded as interest rates increasingly lagged behind inflation over the past five years. Thus Brazil has become highly dependent on continued capital inflows and petroleum imports.

The style of Brazilian growth was an immediate outcome of the major national goal to build in Brazil a “modern” economy. Development projects often utilized the most up-to-date technologies as soon as they could be brought to Brazil. Exchange rate policy, fiscal incentives, and subsidized credit reduced the cost of capital goods, while the use of labor was taxed. With its large domestic market, Brazil was able to follow a strategy of import-substituting industrialization longer and more successfully than most other countries that have taken this route and has even been able to expand rapidly its exports of manufactures.

But this development strategy has required increasing amounts of investment to produce an additional unit of output. The incremental capital output ratio averaged 2.9 over the period 1965-70 and 3.0 for 1970-75. But by 1975-79 it had risen to an average of 3.9. Plant expansion often involved a high degree of automation. Capital-intensive and high-technology industries—such as petrochemicals, aircraft, and nuclear power—absorbed large amounts of capital and led to rapid increases in the demand for skilled labor that helped raise the salaries of skilled technical and managerial personnel to levels at or above those paid in much richer countries. They have not, however, created as many jobs for the unskilled.

Distribution; employment

The available evidence suggests that to some degree growth benefited the poor. Gains in absolute income levels appear to have been widespread. The distribution of consumer durables illustrates the extent of trickle-down, which has been facilitated by a rapid expansion of consumer credit. In 1976 almost 76 per cent of Brazilian homes had a radio, 47 per cent had a television, and 42 per cent had a refrigerator. Even among the poorest third of the rural households, located mainly in the rural Northeast, Brazil’s poorest region, 52 per cent had a radio (although very few had either a television or a refrigerator).

But the gains of the rich have been immeasurably larger than those of the poor, reflecting persistently high levels of income inequality. Available estimates suggest that, for the last 20 years or so, the richest 10 per cent of families have been receiving over 50 per cent of the income, while the poorest 40 per cent have received well under 10 per cent. Not only is the income received by the poorest 40 per cent of families a very small share of the total but also the absolute amounts of income received by the households are meager. Data from the 1978 National Household Sample Survey show that in 1978 about 43 per cent of Brazilian families had a total income (including income in kind) equivalent to less than about $216 per month in today’s prices (or about $47 monthly per family member). Such low incomes received by so many households largely explain why life expectancy and infant mortality in Brazil are comparable to those in countries with much lower average incomes.

The large differences in the incomes earned by different population groups are reflected in correspondingly large differences in the amenities available to them in different parts of the country. In 1970, the latest year for which sufficiently detailed data are available, life expectancy at birth in the industrialized southern regions of Brazil was roughly comparable to that in many developed countries; the average life expectancy of the lowest income group in the urban areas of the central northeast states, however, at 40 years, was about the same as Ethiopia’s average—a country with a per capita GNP about one eighth of Brazil’s. While average life expectancy in the Northeast increased 26 per cent between 1960 and 1976, faster than for Brazil as a whole, a Northeastern child born in the latter year could still expect to live eight years less than the average Brazilian child. Data on health, nutrition, water supply, sanitation, and education tell a similar story, though rates of improvement have been quite fast in some areas, particularly water supply. Despite substantial progress over the past two decades and despite high numbers of physicians, hospital beds, educational opportunities, and so on, relative to the population—which allow Brazil to compare favorably with other countries in its income class—the quality and distribution of these services are uneven. Residents of the Southeast are generally much better supplied than those of other regions—particularly of the poverty-ridden Northeast; urban areas are much better supplied than rural areas.

To accelerate improvement in the economic welfare of the neglected groups in Brazil will require a concerted effort by the authorities to extend basic services and to redistribute incomes. The surest way for the authorities to redistribute incomes is to make more jobs available to the poor and underskilled. Rapid population growth during the past decades assures that the potential labor force will continue to grow between 2.5 and 3.0 per cent until the end of the century. If the trend toward in-creasing labor force participation rates continues, the rate of job creation will have to increase even faster.

Over the period 1960-78, rapid economic growth expanded the number of jobs paying more than one 1970 "minimum wage" (or about $76 per month in today’s prices) almost twice as fast as the working age population but roughly 30 per cent slower than GDP. By 1978 such jobs were held by about half the employed labor force compared with one third in 1960. Jobs paying more than one minimum wage held by low-skilled workers (those with four or less years of schooling), however, increased about one third as fast as those held by workers with five or more years of formal education, 40 per cent as fast as GDP, and only slightly faster than the working age population. Between 1973 and 1976, total employment of adults (aged 20 and above) outside the agricultural sector increased at 4.4 per cent per year while GDP growth averaged 8.1 per cent. All this information suggests that, unless Brazil’s economic development strategy is altered, minimum growth rates of GDP on the order of 5 to 7 per cent will be necessary in the 1980s to avoid growing social tension due to inadequate creation of jobs paying more than the minimum wage, a rather low standard that is not enough even to feed a family of five adequately except in a few rural areas of Brazil (given prevailing dietary habits), much less meet other needs.

Table 1.Growth of the Brazilian economy, 1965-80
Annual average
rate of change
(In per cent)
196519801965-80
Population (in millions)811192.6
Real GDP (1970 prices)1003388.5
Real GDP per capita1002315.7
General price index, domestic supply100733933.2
Total exports (1970 prices)1003909.5
Total imports (1970 prices)10043210.2
Agricultural production1001723.7
Industrial production1004009.7
Automotive vehicle production185116513.0
(in thousands)
1967198011967-80
In million metric tons
petroleum equivalent
Primary energy production40775.2
Primary energy consumption511226.8
Primary energy imports114411.0
Sources: Fundacáo Getúlio Vargas; Conjuntura Económica, December 1979 and February 1980; Fundacáo Instituto Brasileiro de Geografía e Estatistica, Censo Demográfico do Brasil 1980, Resultados Preliminares; Instituto Brasileiro de Geografía e Estatistica, Anuario Estatistico do Brasil, 1967; Ministry of Mines and Energy. National Energy Balance, 1978.

1980 figures are estimates based on projections for 1980 in National Energy Balance, 1978 and adjusted using 1980 trade data.

Sources: Fundacáo Getúlio Vargas; Conjuntura Económica, December 1979 and February 1980; Fundacáo Instituto Brasileiro de Geografía e Estatistica, Censo Demográfico do Brasil 1980, Resultados Preliminares; Instituto Brasileiro de Geografía e Estatistica, Anuario Estatistico do Brasil, 1967; Ministry of Mines and Energy. National Energy Balance, 1978.

1980 figures are estimates based on projections for 1980 in National Energy Balance, 1978 and adjusted using 1980 trade data.

Brazil’s serious balance of payments problems, a less favorable international environment, rising energy costs, and the need to bring down the rate of inflation suggest that the sustainable rate of GDP growth in the 1980s will be in the lower half of this range. Unexpected difficulties could reduce it further. If Brazil is to continue or accelerate its social progress under these difficult conditions, a development strategy is required that increases employment generated per unit of new investment, reduces imports per unit of additional output, increases the proportion of children from low-income families obtaining the higher skills that more than four years of schooling provide, and raises the productivity and well-being of the poor through improving their health. Since decreased infant mortality, higher educational status of mothers, and increased participation of women in the labor force are known to reduce fertility, such a strategy could also help slow population growth even in the absence of an official family planning program.

A new strategy

Many elements of such a strategy are contained in Brazil’s Third National Development Plan for the years 1980-85, major objectives of which are to improve income distribution and accelerate the growth of income and employment while fighting inflation and achieving equilibrium in the balance of payments. The idea is to alter the profiles of investment and consumption, to slow imports and energy use, while creating more jobs and giving more people the opportunity to do them. The essence of the strategy is to accelerate the expansion of basic public services (through the state sector) and of basic wage goods (through the private sector). The former would be financed by a combination of taxes and user charges. Production of the latter would be stimulated by tax and credit reforms designed to increase employment and reduce the market prices of such goods, thus expanding effective demand for them. A land reform, not mentioned in the Government’s plan, could also increase employment, the production of foodstuffs, and effective demand for food and other wage goods among the rural poor. Brazil in the 1980s would invest more in education and basic health services, water supply, and sewerage. The pattern of production for the domestic market would shift toward low-cost housing, food, clothing, bicycles, and buses; and away from luxury housing, color television sets, cosmetic surgery, private automobiles and the infrastructure they require, and the like. But there would be no relaxation of Brazil’s successful drive to expand exports, which in the 1970s have been, on average, more labor intensive than import substitutes.

The new growth strategy would also require heavy investments in energy production. The era of cheap energy is over. While there are clearly ways that are less costly than others—in both economic and social terms—to meet Brazil’s increasing energy demands, there are no miraculous solutions in sight. All forms of commercial energy, particularly liquid fuels, have higher real costs today. Given the vulnerability of imported petroleum supplies as well as the high probability of further relative price rises, planners have persuasive arguments to consider investments in domestic substitutes for imported petroleum, even when costs are well above current import prices. Equity and environmental considerations also need to be factored into the planning process.

Difficult choices are inevitable. A striking example is a potential conflict between the production of basic wage goods and domestic energy inherent in the fuel alcohol program launched in 1975. In 1980, alcohol production was 3.4 billion liters (0.9 billion gallons), accounting for 19 per cent of combined alcohol and gasoline consumption. The target for alcohol production in 1985 is 10.7 billion liters (2.8 billion gallons), more than triple the 1980 level, to be achieved almost entirely by expanding sugarcane production, which requires good agricultural land.

Table 2.Selected social indicators, 1960 and 1978
19601978
BrazilNortheastBrazil1Northeast
Years
Life expectancy at birth252426153
Deaths of infants aged 0-12 months per thousand live births
Infant mortality212316989122
In per cent of total population
Population with access to
general water networks2155228
Population living in homes
with any sanitary device
(latrine, septic tank, flush
toilet, and so on)49247542
Adult literacy rate60407656
In per cent of population aged 7-14 years
Students in grades 1-857379479
In per cent of all children aged 0-17 years
Children with second or third
degree malnutrition3n.a.n.a.2130
Sources: Knight. Moran. et al. (1979); Fundagao Instituto Brasileiro de Geografia e Estatistica, Pesquisa Nacional por Amostra de Domicilios 1978; Instituto Brasileiro de Geografia e Estatistica, Censo Demografico do Brasil. 1960; and United States National Academy of Sciences Committee on Population and Demography, Panel on Brazil. Preliminary Report. 1979.

Brazil data for 1978 excludes the rural areas of the North and Center West regions, roughly 5 per cent of total population.

Estimates centered on the end of 1957 and of 1973. derived from data in the 1960 demographic census and 1976 National Household Sample Survey, the latter excluding rural areas of the North and Center West regions.

75 per cent or less of normal body weight.

Sources: Knight. Moran. et al. (1979); Fundagao Instituto Brasileiro de Geografia e Estatistica, Pesquisa Nacional por Amostra de Domicilios 1978; Instituto Brasileiro de Geografia e Estatistica, Censo Demografico do Brasil. 1960; and United States National Academy of Sciences Committee on Population and Demography, Panel on Brazil. Preliminary Report. 1979.

Brazil data for 1978 excludes the rural areas of the North and Center West regions, roughly 5 per cent of total population.

Estimates centered on the end of 1957 and of 1973. derived from data in the 1960 demographic census and 1976 National Household Sample Survey, the latter excluding rural areas of the North and Center West regions.

75 per cent or less of normal body weight.

Although Brazil has the potential to feed its people well, expand agricultural exports rapidly, and have sufficient land left to produce a substantial portion of its liquid fuel requirements from biomass, the fact is that, between 1966 and 1979, the output of domestic food crops has hardly kept up with population. On the other hand, export and industrial crops, dominated by soybeans, have increased over seven times as fast as population. Estimates suggest that, if recent trends in agricultural productivity persist, Brazil would need roughly to double the annual rate of increase in cultivated area if it is to produce enough for domestic consumption, exports, and energy production between 1978 and 1985. This would require substantial new investments in economic and social infrastructure. This competition for good agricultural land contains the danger that foodcrop production will be displaced from prime lands, particularly toward the end of the period and in the Northeast.

There are three ways to reduce the potential for food-fuel conflict: by increasing yields of both food and fuel crops; by developing fuel crops, including wood, that can be grown economically on land of little or no food-producing potential; and by converting agricultural wastes into alcohol. All of these approaches require a major research effort to develop the technology—a task on which Brazil has embarked but which could be speeded up.

On the other hand, the production of energy from biomass is relatively labor intensive compared to electric power or petroleum production and, if carefully planned and supported by improved technology, could offer new sources of agricultural incomes and employment. The potential for exploiting biomass energy is not limited to alcohol, whether from conventional crops or wood, but also includes direct production of fuelwood and vegetable oils, which are alternatives to diesel fuel being studied.

Feasibility of the strategy

Brazil, with its high per capita GNP and good prospects for growth, has the resources to meet the basic needs of all its citizens within 20 years without any major reduction in living standards for the better off.

The existing government programs in health, nutrition, education, housing, water supply, and sewerage, as well as the kinds of reforms necessary to improve their efficiency and extend basic coverage to the entire Brazilian population in need of these services, are analyzed in considerable depth in the Brazil: Human Resources Special Report published by the World Bank. Estimates presented there suggest that virtually all the main needs of the population could be met between the years 1990 and 2000 by the Government spending on the order of 5 to 6 per cent of GDP annually between now and then. The objective could be achieved, moreover, with maximum additional taxation (beyond taxes or other charges already on the books) of approximately 2 per cent of GDP.

Thus achieving this objective is not, at its root, a major economic or financial problem. The binding constraints are more likely to be organizational, staffing, and logistical bottlenecks, which could retard the pace of extending services, especially in rural areas. These are serious obstacles. Much remains to be learned about how to deal with them effectively. But if priority is given to meeting basic needs, the obstacles can be overcome.

Tax reform could increase the number of jobs available by removing existing subsidies on capital use—fiscal incentives having this effect cost the Treasury an estimated 1.2 per cent of GDP in 1978—and changing the taxes and quasi-taxes that fall on labor so that instead they fall on value added. This would change the relative price of labor and capital thereby encouraging the use of more labor-intensive techniques of production for any given product and the production of goods that use less capital per unit of labor. It would also stimulate research and development on more labor-intensive techniques. By changing the structure of indirect taxes so as to reduce or eliminate the taxes on basic wage goods and increase them for luxury goods, the lower prices to consumers of the former (generally produced under competitive conditions) would expand their market and encourage production. The reverse could happen for luxury goods. Though research is needed to prove this hypothesis, we believe that changing the structure of indirect taxes in this way would reduce imports and energy use while expanding employment, compared with what would happen in the absence of such a reform.

By eliminating the fiscal incentives, establishing a capital gains tax, and increasing the very low inheritance tax, it should be possible to mobilize at least an additional 4 per cent of GDP for the Treasury, or twice the incremental taxation required for meeting basic needs by the year 2000. This should provide an ample margin for applying higher standards in providing services to the poor or for accelerating the rate of implementation. These reforms could be phased in over a period of several years. If GDP growth can be maintained at 5 or 6 per cent, these tax reforms would not require any significant decrease in the disposable incomes of the well-to-do.

In recent years massive credit subsidies and a system of selective credit distribution have caused widely diverging interest rates paid by firms of different economic sectors, locations, and sizes. Though causality is difficult to prove, economic theory and the available evidence suggest that these factors have stimulated consumption at the expense of saving, hampered employment growth, and furthered geographical and personal concentration of income and wealth. They have certainly made rational economic calculation extremely difficult and have been very expensive to administer. Reducing these distortions in credit markets would reinforce the allocative impact of the tax reforms.

The combination of tax and financial reforms would help finance the extension of basic public services and encourage the production of basic wage goods by the private sector. By increasing employment, these reforms would also help to create demand for the increased supply of wage goods. They might also help reduce the rate of increase in energy consumption and save on imports, thereby contributing to easing Brazil’s balance of payments problems. By investing in the human resources of the poor, the flexibility and productive capacity of Brazil’s labor force would be enhanced, allowing the economy to respond more effectively to future challenges and opportunities presented by the world economy. F&D

Related reading

Other Resources Citing This Publication