Can Openness to Trade Reduce Income Volatility? Evidence from Colonial India’s Famine Era∗ Robin Burgess LSE and NBER

Whether openness to trade can be expected to reduce or exacerbate the equilibrium exposure of real income to productivity shocks remains theoretically ambiguous and empirically unclear. In this paper we exploit the expansion of railroads across India between 1861 to 1930—a setting in which agricultural technologies were rain-fed and risky, and regional famines were commonplace—to examine whether real incomes became more or less sensitive to rainfall shocks as India’s district economies were opened up to domestic and international trade. Consistent with the predictions of a Ricardian trade model with multiple regions we find that the expansion of railroads made local prices less responsive, local nominal incomes more responsive, and local real incomes less responsive to local productivity shocks. This suggests that the lowering of transportation costs via investments in transportation infrastructure played a key role in raising welfare by lessening the degree to which productivity shocks translated into real income volatility. We also find that mortality rates became significantly less responsive to rainfall shocks as districts were penetrated by railroads. This finding bolsters the view that growing trade openness helped protect Indian citizens from the negative impacts of productivity shocks and in reducing the incidence of famines. ∗Correspondence: [email protected] and [email protected] We thank Richard Blundell, Chang-Tai Hsieh, and seminar participants at Bocconi University and the 2012 Nemmers Prize Conference (at Northwestern) for helpful comments. This document is an output from research funding by the UK Department for International Development (DFID) as part of the iiG, a research programme to study how to improve institutions for pro-poor growth in Africa and South-Asia. The views expressed are not necessarily those of DFID.