How Hard Did That Sting? Estimating the Economic Costs of Locust Attacks on Agricultural Production†

Estimates suggest that a swarm of desert locusts can consume food grains sufficient to feed up to 35,000 people. Despite the potential disaster that these migratory pests can bring about, there is no credible economic estimate of the actual damage caused by locust upsurges to agricultural production. This article takes advantage of plausibly exogenous variation generated by the timing of locust upsurges and geographic variation due to their migration patterns to estimate the impact of an upsurge on crop yields using a large district level panel dataset from India covering more than 200 districts over 45 years. I find that locust upsurges have no impact on rice yields but can lead to a decline of up to 0.2 tons of wheat production per hectare. Given that locusts attack India from the western side and majority of the rice is produced in the east and south, such findings are not surprising. There are two ways to interpret the estimated effect size in terms of how hard the locust upsurges can sting the economy. First, in terms of the economic costs, the magnitude of the effects are around 12% compared to mean per‐hectare production of wheat. Second, in terms of accounting costs, back of the envelope calculations suggest losses up to US$3 billion. Policy implications include recognizing locust upsurges as natural disasters and providing demand‐based market instruments such as locust insurance to help farmers hedge against this risk.