Poverty and Health in Developing Countries

Featured in print Bulletin on Aging & Health

In recent years, global health issues have received a great deal of attention and funding from charitable foundations and governments around the world. Much of the attention has been focused on specific diseases, such as AIDS and malaria, which present grave challenges to many developing countries.

More generally, what is known about the health of people in developing countries, particularly the very poor? Re-searchers have long been aware of large disparities in health outcomes by income and other measures of socioeconomic status in the U.S. and other developed countries. Do these disparities exist in developing countries as well? If so, programs to alleviate poverty could have beneficial effects on health.

In "Aging and Death Under a Dollar a Day" (NBER Working Paper 13683), researchers Abhijit Banerjee and Esther Duflo examine the relationship between income and adult mortality in developing countries.

The authors' first approach to explore whether the poor have excess mortality in adulthood is to construct age pyramids by income level to see if the number of poor older people in the population is unusually low. While the authors find evidence that it is in many of the countries they study, they also point out that differences by income level in fertility rates and family structure make it difficult to interpret the results of this exercise.

As an alternative, the authors use information on whether the parents of adults sampled in the surveys are alive. As these parents would likely be over age 50 (if they are alive), this data can be used to see how the population of those aged 50 and above changes across the income spectrum. The authors find that the probability that parents are alive is similar for those with DPCE of less than $4 and rises with income thereafter. The effects can be very large - for example, pooling rural households in all countries, the probability that the mother is alive is 36 percentage points higher if the family has a DPCE of $6 to $10 versus a DCPE of $1 to $2. The effects for fathers and urban households are smaller but also significant.

The third approach the authors employ is to use panel data to compare mortality in later waves of the survey by poverty status at the first wave. The data to conduct such an analysis exist only for Indonesia, Vietnam, and India. Once again, the results are striking - in rural areas, adults over 50 living on less than $1 or $2 a day are at least three times as likely to die over the next five to seven years than those living on $6 to $10 a day.

Finally, the authors explore the relationship between morbidity and income. Here, the findings are more mixed - health deteriorates more rapidly with age for the poor than the non-poor in rural India, but this is not the case in rural Indonesia.

Overall, the results indicate that the poor have a lower chance of survival than those who are somewhat more well-off. The direction of causality is not immediately obvious - adults could be poor because they are in poor health (and subsequently end up dying), or being poor could make them more likely to die. As the authors point out, however, most old people in developing countries live with other younger adults, a fact that weakens the case for the causality to run solely from the health of the old person to the poverty status of the household. The authors conclude "on balance, we are tempted to interpret the evidence accumulated in this paper as revealing, as least in part, that poverty does kill."

The authors warmly thank the National Institute on Aging (grant P01-AG005842), the Center for Health and Wellbeing at Princeton University, and the World Bank for funding.