54000 Finance & PSD Impact MARCH 2009 The Lessons from DECRG-FP Impact Evaluations ISSUE 3 The third in our new series of impact notes highlights an example of where randomized experiments have been used to help understand the impact of a new insurance product first developed by CRMG at the World Bank and various public and private partners in several client countries. The Promise of Index Randomizing access to insurance Insurance Recent randomized experiments in Malawi Xavier Gine and India were used to try and answer these questions. Index-based insurance is an innovative financial product, which has been introduced In Malawi, smallholders were offered credit in recent years in countries as diverse as to purchase high-yielding seed varieties. To India, Mongolia, Malawi and Thailand. It understand whether adoption was allows individual smallholder farmers to discouraged by its inherent riskiness, hedge against agricultural production risk, farmers in some localities were randomly such as drought or flood. The product pays selected to be just offered credit, while those out in events that are triggered by a publicly in other localities were offered a bundle of observable index, such as rainfall recorded credit and insurance. on a local rain gauge. Advocates argue that index insurance is transparent and inexpensive to administer, enables quick In India, smallholders were offered a stand payouts, and minimizes moral hazard and alone insurance product whose price adverse selection problems associated with elasticity was estimated by randomly other risk-coping mechanisms and varying the price of the policy. To traditional insurance programs. understand the role of credit constraints, certain households were given a positive This financial innovation holds significant liquidity shock. To measure the importance promise for rural households. Weather of trust, some households received a product shocks to agricultural income generate endorsement by a trusted local agent. To fluctuations in household consumption that understand whether limited financial are not perfectly insured; at the extreme they education about the product limits adoption, may lead to famine or death. This suggests additional information was provided to a that households in developing countries are subset of households relating the unfamiliar only partially insured and as a result, may concept of rainfall in millimeters to the pass on more profitable but riskier familiar concept of soil moisture. Finally, to investments. understand the effect of product framing, information was presented in subtle different An important set of interrelated questions ways. about these micro-insurance products are (i) what types of households buy index insurance, (ii) what are the factors which Low uptake of insurance prevent the remaining households from participating and (iii) does purchase of index In Malawi, take-up of the credit was 33 insurance result in more efficient risk taking. percent for farmers offered the loan without insurance, and only 17.6 percent for farmers Do you have a project you want evaluated? DECRG-FP researchers are always looking for opportunities to work with colleagues in the Bank and IFC. If you would like to ask our experts for advice or to collaborate on an evaluation, contact us care of the Impact editor, David McKenzie (dmckenzie@worldbank.org) who were offered a loan bundled together can be a complicated function of the event with rainfall insurance. This suggests that being insured against. smallholders did not value insurance, 2. Mistrust can be overcome by designing perhaps because the limited liability clause a product that initially pays fairly often. in the loan contract was already providing implicit insurance. It is easier to sell insurance in villages where a positive past insurance payout has In India, demand was sensitive to the price occurred. Therefore, it would be useful to and to the endorsement from a third trusted modify the contracts, at least in the party. But uptake remained low, even when beginning to ensure that they payout a the purchase of insurance had positive net positive return with sufficient frequency as present value given the high subsidy. These to engender trust in the population. Current results are consistent with the view that in contracts resemble catastrophic insurance, addition to price and liquidity, trust and paying large sums only in rare events. financial literacy influence take-up to a significant degree. 3. Because liquidity constraints matter, lenders could offer a loan to pay for the Besides the wealthier, the more educated premia. and financially literate were also more likely to purchase insurance, suggesting that these In this case, lenders would have to be products are complex. explicit about the events that trigger a payout, so that the culture of repayment is not undermined. Policy Implications 4. The focus of research has been on the 1. Implementation and Marketing of farmer or borrower as beneficiaries of index insurance should be carefully insurance. When considering agro- designed. businesses or lenders the demand for The findings on trust and financial literacy insurance should be a no-brainer. suggest scope for modifying implementation From the agro-business or the lender's and marketing in a way which will boost standpoint, weather insurance is an attractive demand. To the extent that poor farmers are way to mitigate default risk and thus, it can unable to understand complicated insurance become an effective risk management tool products, a trade-off arises since the product with the potential of increasing access to has to be simple and yet has to pay out in the credit in agriculture at lower prices. events the smallholder cares about, which For further reading see: Gine, Xavier, Robert M. Townsend, and James Vickery. 2007. "Statistical Analysis of Rainfall Insurance Payouts in Southern India." American Journal of Agricultural Economics 89(5): 1248-54. Gine, Xavier, Robert Townsend, and James Vickery. 2007. "Patterns of Rainfall Insurance Participation in Rural India." World Bank Economic Review 22(3) (2008):539-566. Gine, Xavier, and Dean Yang. 2007. "Insurance, Credit, and Technology Adoption: Field Experimental Evidence from Malawi." Journal of Development Economics 89(1) (2009): 1-11. Cole, Shawn, Xavier Gine, Jeremy Tobacman, Petia Topalova, Robert Townsend and James Vickery "Barriers to Household Risk Management: Evidence from India". Working Paper, World Bank, Washington, DC. Recent impact notes are available on our website: http://econ.worldbank.org/programs/finance/impact Impact Note 1: Which Microenterprises have high returns to capital? ­ David McKenzie Impact Note 2: Does Business Registration Reform increase entrepreneurial activity? ­ Miriam Bruhn