Rainfall insurance and agricultural and financial decisions: evidence from a series of experiments in India
Speaker: Shawn A.Cole, Associate Professor, Harvard Business School
Date: 17 December, 3.30-5 pm
Discussions at this seminar centered around potential implications for the broader Indian agricultural insurance landscape through two research projects on agricultural risk management and decision-making. The studies look at the determinants of adoption of weather index insurance; the dynamics of demand for the product; and the impact of adoption on agricultural investment decisions. The research results are based on a series of randomized evaluations conducted in Andhra Pradesh (in cooperation with BASIX) and Gujarat (in cooperation with SEWA) over the past five years.
The study based in Andhra Pradesh looks at how the availability of risk management instruments designed to hedge rainfall risk affects investment and production decisions of small- and medium-scale Indian farmers. To overcome identification challenges, the study randomises the provision of rainfall insurance to farmers. While little effect on total expenditures is evident, increased insurance induces farmers to substitute production activities towards high-return but higher-risk cash crops, consistent with theoretical predictions. The results support the view that financial innovation may help ameliorate costs associated with weather variability and other types of risk.
In a follow-up experiment with the same farmers, the study elicits willingness to pay for both standard insurance policies, as policies whose features vary, using the Becker Degroot Marshak mechanism. Demand is high enough that a commercially-priced policy would be viable. Farmers’ valuations increase (decrease) when the insurance contracts increase (decrease) the expected payout of the policy, but the change in willingness to pay is much smaller than the change in the actuarial value of the contract, suggesting farmers have difficulties valuing these products.
The Gujarat study document the dynamics of demand for rainfall insurance for households for a period of five years. It demonstrates how past purchase, rainfall realisations, and insurance payouts affect subsequent demand for the policies.