Few would argue with the notion that an improvement in the material well-being of the poor will produce better health outcomes for that group, said Milyo. But the inequality hypothesis says something very different: that all people, regardless of their wealth or social position, run a greater risk of illness if they live in a society that is more hierarchical than egalitarian. Thus, while the poor may suffer the most from inequality, the rich would suffer as well.
Proponents of the hypothesis have offered reasons behind such a correlation. Hierarchies may increase the stress levels of individuals, they propose, which can lead to heart disease or self-destructive behavior. Stress would most greatly affect the poor, but those consequences would affect others in society, as well. Another explanation for a correlation might be that inequality causes people to view their neighbors as more alien than would be the case in an egalitarian society. As a result, citizens would be less concerned for the well-being of their neighbors.
Milyo disagreed with the hypothesis, however. He and his colleague Jennifer M. Mellor have found that, on closer inspection, the basic arguments and evidence that inequality and poor health are closely correlated are wanting in many respects.
At the national level, said Milyo, the association between inequality and public health is not robust when different measures of inequality or health are used, different periods are examined, or other control variables are included. The evidence behind the inequality hypothesis relies mainly on data from around 1970 using few control variables. He and Mellor conducted a similar study based on data from a longer period to account for lags in health effects. They found that the correlation between inequality and health is sensitive to both the time period and the number of countries examined. When they controlled for variables among nations such as education level, the correlation became insignificant. Milyo stressed, however, that because of the dubious quality of international income-inequality data, little weight should be placed on any of the findings, one way or the other.
In the past, studies of income distribution among U.S. states showed a strong link between income inequality and health outcomes. As in their other studies, Milyo and Mellor reviewed data from a longer period of time, but this time at the state level. With no control variables, the outcome confirmed the inequality hypothesis, but when they controlled for variables among states such as racial composition and education levels, the correlation became insignificant and in some cases inverse–suggesting that higher inequality within a state leads to positive aggregate health outcomes.
Finally, proponents of the inequality hypothesis have observed that, at the individual level, the wealthier a person is, the better his health. Although such studies offer fairly convincing evidence that rank in a social hierarchy is correlated with poorer individual health, Milyo has found that they do not demonstrate that either rank in the hierarchy or the existence of a social hierarchy is a cause of poorer health.
"Given some of the contrary findings in recent work, one could argue that inequality is as likely to improve people's health as harm it," concluded Milyo. He added that the income inequality hypothesis is particularly worrisome because a number of scholars have jumped to the policy implications.
Discussant Nicholas Eberstadt of AEI has studied statistical data on the relationship between international economic data and health outcomes for developing countries. At the conference he confirmed Milyo's finding that the data does not show a consistent correlation. AEI's Newt Gingrich proposed that politics and culture in a society shape health more than income inequality does. He worried that those behind the inequality hypothesis may be yielding to a temptation to find data to support increased income redistribution by central governments.