Health and Wealth among the Poor: India and South Africa Compared

摘要

Health and wealth are the two most important components of well-being. Rankings of well-being based on income will differ from more comprehensive rankings, depending on the way that income and health are related. There are strong bidirectional causal links between income and health so that we cannot understand either without understanding both. What we call the "wealthier is healthier" hypothesis asserts both that income is the main determinant of health and that the international correlation between income and health is sufficiently tight for income rankings to indicate well-being more broadly (see e.g., Richard G. A. Feachem [2001] who, like many researchers, cites Lant Pritchett and Lawrence H. Summers [1996] on the causal role of income in protecting health). Comparisons between India and South Africa provide a good illustration of both the useful-ness and the complexities of working with both health and wealth. In 2000, India's per capita income was $2,045 in (chained) purchasing power parity (PPP) dollars, less than a third of South Africa's, $7,409. Even if we adjust the latter figure for the distribution of income be-tween whites and Africans in South Africa, Af-ricans in South Africa are at least 50 percent better off than Indians in terms of GDP per capita. Yet life expectancy in South Africa, which in 1990 was three years longer than in India, by 2000 was 14 years shorter; and the reversal is even more dramatic if South African whites are excluded. In this paper, we report direct comparisons of a number of objective and subjective measures of economic and health status in two sites, one in the district of Udaipur in rural Rajasthan, and one in the shack township of Khayelitsha near Cape Town. We are interested in an assessment of the "wealthier is healthier" hypothesis and, more generally, in the feasibility of making international comparisons of well-being using individual-level data, and especially the use of subjective measures of health and living standards (i.e., whether they reflect objective measures, or are fully adapted to local expectations). Our results show that the economically better-off South Africans are healthier in some respects, but not in others. They are taller and heavier, but their self-assessed health is no better; they suffer from depression and anxiety to about the same degree; they have a remarkably similar pattern of prevalence of various health conditions; and both adults and children in South Africa are more likely to go without food for lack of money. Even if some of the self-reported deprivations, such as hunger, are assessed relative to different local expectations, the effects on anxiety and mental health are absolute and absolutely comparable. Because health, like well-being, is multidimensional, and because the components of health do not correlate perfectly with one another, nor with income-based measures, income on its own is likely to be misleading as a shortcut measure of international health. Even within places, such as the two examined here, the links between health and wealth are far from universally strong.