religion and health

I … find that, at least on average, over all [142] countries, and over countries sorted into income groups, religious people do better on a number of health and health-related indicators. These protective effects appear to be stronger the poorer is the country—… religion is a route to a better life in poor countries, but not in rich ones—and to protect men more than women, though this hypothesis requires more extensive investigation.

None of the results show that the health benefits of religion can be obtained simply by joining a church, or even by undertaking a serious conversion. People who are religious are almost certainly different from non-religious people in ways that go beyond their religiosity and beyond the basic educational and demographic controls that are used here. Even so, some of the correlations presented here are remarkably universal across the religions and countries of the world, and need to explained and better understood.

Angus S. Deaton, “Aging, religion, and health”, NBER Working Paper No. 15271, August 2009.

An ungated version can be downloaded here

Princeton economist Angus Deaton (1945-), current president of the American Economic Association, is well-known for his empirical studies of household behavior and for his research in health economics and economic development.

Professor Deaton in this paper assumes “a simple triangular causal structure, in which religiosity and religious practice are caused by income, education, age, and sex, and in which health is caused by religion, income, education, age and sex.”. He emphasises that he is actually measuring correlation, not causation: the latter is assumed, not proven. Many economists are not so careful, and claim too much for their econometric results.


Comments are closed.