Our systems are now restored following recent technical disruption, and we’re working hard to catch up on publishing. We apologise for the inconvenience caused. Find out more: https://www.cambridge.org/universitypress/about-us/news-and-blogs/cambridge-university-press-publishing-update-following-technical-disruption
We use cookies to distinguish you from other users and to provide you with a better experience on our websites. Close this message to accept cookies or find out how to manage your cookie settings.
Published online by Cambridge University Press: 03 March 2009
Brazil is often cited as a developing country that has exemplified the trade-off between economic growth and equity. The basis for the claim was the rising inequality in the income distribution between the 1960 and 1970 censuses, combined with an increase in per capita income over the decade. Numerous scholars have analyzed the period, yet there is little or no agreement in the literature regarding either the causes or the extent of this apparent increase in inequality. Using a recently released sample of the 1970 demographic census, the empirical basis for the claim of deteriorating income distribution over the decade is examined here. Applying poverty lines defined for both 1960 and 1970 data sets, the correlates of income variance and poverty are analyzed. The results show that under most reasonable assumptions regarding data deficiencies and appropriate deflators, the fraction of the population in absolute poverty declined only slightly over the decade. Although the poorest households were found among the small farmers and sharecroppers in the rural areas in both 1960 and 1970, a large increase in the incidence of poverty in urban areas occurred over the decade.
1 Dervis, Kemal, Melo, Jaime de, and Robinson, Sherman, General Equilibrium Models for Development Policy (New York, 1982).Google Scholar
2 Some have estimated as much as 40 percent of Brazilian national income is not accounted for by the 1970 census. For a detailed discussion of the weaknesses of the census data, see Pfeffermann, Guy and Webb, Richard, “The Distribution of Income in Brazil,” World Bank Staff Working Paper No. 356 (Washington, D.C., 09 1979).Google Scholar
3 Pfeffermann and Webb (ibid.) hold this position.
See Bacha, Edmar and Taylor, Lance, “Brazilian Income Distribution in the 1960's: Facts, Model Results and the Controversy,” Journal of Development Studies, 14 (04 1978), for a discussion of mutations of this position and a critical assessment.Google Scholar
4 Bacha and Taylor, “Brazilian Income Distribution in the 1960's.”Google Scholar
5 Results of this tabulation have been published in two places: in Fishlow, Albert, “Brazilian Size Distribution of Income,” American Economic Review, 70 (05 1972),Google Scholar
and in Meesook, Oey Astra, “Income Distribution in Brazil,” (Ph.D. diss., University of California, 1972). The presentation differed somewhat in each work, but the underlying data base is the same—the adjusted 1960 census sample. I have used Fishlow's table of the distribution of income for my relative inequality comparison and Meesook's table of the characteristics of the poor for my poverty line tabulations, since Meesook provides a much clearer discussion of her methodology than does Fishlow. I refer to the joint methodology as the Fishlow-Meesook method.Google Scholar
6 Results of this tabulation and extended documentation on the methodology used in the analyses described here are provided in Fox, M. Louise, “Income Distribution in Brazil: Better Numbers and New Findings,” (Ph.D. diss., Vanderbilt University, 1982).Google Scholar
7 See ibid., chap. 2 for a complete discussion of the issues that precipitated this difference in methodology.
8 Meesook, “Income Distribution in Brazil.”Google Scholar
9 Fox, “Income Distribution in Brazil: Better Numbers and New Findings,” for other poverty line calculations using price indices derived from the 1974 expenditure survey (or ENDEF).Google Scholar
10 Williamson, Denise, “Food Prices and Consumption Comparisons—Brazil 1975,” mimeo (April 1981).Google Scholar
11 Fox, “Income Distribution in Brazil: Better Numbers and New Findings,” chap. 3.Google Scholar
12 Beckerman, Paul and Coes, Donald, “Who Benefits from Economic Development?: Comment,” American Economic Review, 70 (03 1980).Google Scholar
13 The ideal price index for my use would be one that measured the changes in the price of a basket of commodities consumed by the poor. This index is not available. Estimates using the Guanabara Cost of Living Index, which is disaggregated in seven separate indices, and a commodity basket (computed using the ENDEF data) that represented the distribution of purchases of goods and services by the poorest quintile, suggest that the use of Sāo Paulo Cost of Living understates slightly the changes in the national price level as they affected the poorest segment of the population.Google Scholar
14 Using a lower poverty line in Fox, “Income Distribution in Brazil: Better Numbers and New Findings,” chap. 3, I found a much higher relative incidence of poverty in rural areas. As noted above, the Fishlow-Meesook method of adjusting the data and drawing the poverty line has an urban bias, and therefore I find lower relative incidence of poverty in rural areas using their method.Google Scholar
15 This observation ignores the large increase in access to public services that occurs for all income groups with increasing urbanization. Given a choice, many migrants might rather be poor in the city than in rural areas, but my data do not allow this measurement. For evidence on service delivery, see Knight, Peter and Moran, Ricardo, “Brazil: Human Resources Special Report, Summary Report,” World Bank Report No. 2604-BR (July 1979).Google Scholar
16 For quantitative estimates of the contribution of the head to total household income, see Fox, “Income Distribution in Brazil: Better Numbers and New Findings.”Google Scholar
17 Fishlow, “Brazilian Size Distribution of Income.”Google Scholar
18 This decline in the portion of zero income earners could simply represent the better calibration of the 1970 census instrument, since the 1960 census only includes nine income classes. It should be noted that the figures in Table 3 for 1960 are very rough estimates.Google Scholar
19 Fox, “Income Distribution in Brazil: Better Numbers and New Findings,” chap. 5.Google Scholar
20 Knight and Moran (“Brazil: Human Resources”) provide data that suggest this to be true.Google Scholar
