Transportation infrastructure and child mortality: Subnational evidence for 22 developing countries

What prevents people from utilizing health services? This paper hypothesizes that transportation infrastructure, as measured by travel time to cities, improves population health. Combining data for travel times with information on child mortality for 290 sub-national regions in Sub-Saharan Africa and Asia for the years 2000 and 2015, we show that a 1 standard deviation reduction in travel times, within sub-national regions, is associated with 9.3 fewer child deaths per 1,000 live births. Using estimates from the literature on the statistical value of life in developing countries, a 1 standard deviation reduction in travel times generates gains equivalent to 1.8 – 4.4% of GDP. Our results are not driven by selection on unobservables or changes in economic development and population density. Instrumental variables support a causal interpretation of the results. The life-saving effects of transportation are larger where poverty is most dire and where political institutions are better functioning.