LAUSR

Abstract This paper examines the causal effect of public investment on food output by exploiting a panel dataset covering a large central government program (Hundred Billion Plan) in China from 2007 to 2013. Based on an event study methodology, we find that public investment significantly increases food output by 4.34%. We find that the sources of the policy effect come mainly through improving productivity, which is driven by increases in fertilizer inputs by households and by the crowding in local government agricultural investments. Further, an ad-hoc benefit-cost analysis shows that the internal return rate (IRR) is 0.23% over a 20-year period. Our results are robust in terms of the assumption of parallel trends, the confounding effect from an alternative policy, the nonrandom selection issue of policy implementation, the inclusion of alternative sets of control variables, the treatment of outliers and the omitted variable biases.