Over the last decade, the proposition that inequality in income—and not just poverty per se—leads to detrimental health consequences has gained increasing currency in public health circles. This belief has become a popular topic within schools of public health, scholarly journals and health philanthropies, and has surfaced in political discussions and in the popular media both in the United States and abroad.
This newly fashionable “inequality hypothesis” predicts that the risk of illness rises importantly in a society with income disparities. Indeed: it implies that redistributing wealth from the rich to the poor (and thereby reducing measured income inequality) should in itself produce an improvement in national health.
Not surprisingly, many progressives have reacted enthusiastically to a thesis that points to the redistribution of wealth as a solution to the nation’s health problems. But in an age of “evidence-based medicine,” how solid is the evidence that is being used to promote this bold and dramatic public health claim? In Health and the Income Inequality Hypothesis (AEI Press, January 2004), two American Enterprise Institute scholars, Nicholas Eberstadt, holder of the Henry Wendt Chair in Political Economy, and Sally Satel, a physician, probe for answers to this question. They find that the available data do not support a causal relationship between distribution of wealth and population health.
Instead, Eberstadt and Satel find that:
- The income inequality hypothesis is seemingly supported by country or state-level comparisons between patterns of income inequality and health. But such comparisons are misleading: statistical constructions are being mistaken for real effect. The relationship between per-capita income and health is not linear—a simple and well-known fact, but one nevertheless overlooked in almost all “inequality hypothesis” studies. (Doubling your income, for example, does not necessarily double your life expectancy.) Neglecting this simple truth biases results.
- Variables other than income distribution can explain observed differences in health between groups. After social and behavioral factors (education, smoking, diet, etc.) are taken into account, in fact, the additional explanatory power of income distribution and other measures of inequality seems to be minimal.
- “Sociobiology” experiments with animals provide some of the basis for “inequality hypothesis” speculations about the stress of hierarchies and adverse health for human beings. (If rank plays a significant role in shaping an animal’s physiological profile, could literally moving a social monkey from a lower to a higher tier in its social arrangement make him healthier?) Studies of macaque monkeys, however, suggest that when predetermined social arrangements are disrupted in the laboratory, the reconfigurations do not improve the overall health of the group.
Eberstadt and Satel conclude that the ambitious claims of the “inequality hypothesis” school rely all too often upon limited or unrepresentative data sets, hazily expounded causality, econometric fallacies, and results that cannot be replicated. The widespread popularity of the income inequality hypothesis must be explained in terms of non-empirical appeal: namely, the age-old siren song that refashioning society in the name of redistributive justice can improve life prospects for us all.