In India, Muslim children are far more likely than Hindu infants to survive past their first birthday even though Muslims tend to have lower consumption, wealth, educational attainment, and worse access to state services.

This pattern has confounded economists. In the April issue of American Economic Journal: Applied Economics, authors Michael Geruso and Dean Spears attempt to answer this puzzle by looking beyond individual households to the actual neighborhood environment.

They find two interesting facts that may help explain the difference in mortality rates. First, Muslim infants typically live in neighborhoods where a larger share of their neighbors are Muslim. And second, both Hindu and Muslim infants are more likely to survive in neighborhoods with high shares of Muslim neighbors.

Figure 2 from Geruso et al. (2018)

The figure above shows how infant mortality declines as the share of Muslims in a neighborhood increases. The mortality rate declines for both Hindus and Muslims.

But why would neighborhoods with more Muslims be better places for health, especially when those households have so many seeming disadvantages?

The authors say it comes down to sanitation.

Hindus are 25 percentage points more likely to defecate in the open — in fields, bushes, or near roads — than the poorer Muslim population. This behavioral difference implies that the fraction of a household’s neighbors who are Muslim has a strong correlation with the cleanliness of the local environment.