June 10, 2013
FAI Video: Shawn Cole Discusses Microinsurance & Factors Affecting Take Up (Part 2)By FAI
Part of our series, In Conversation with FAI
Economist Shawn Cole discusses findings from several experiments on rainfall insurance in India and index insurance products. In Part 2 of this FAI video conversation, Cole discusses an experiment that was conducted in Gujarat, India. Here the researchers were interested in understanding the barriers to adoption of a particular product. The goal of the study is to help farmers manage risks due to failed monsoons when making production decisions. Read his paper with Xavier Giné, Jeremy Tobacman, Petia Topalova, Robert Townsend, and James Vickery, "Barriers to Household Risk Management: Evidence from India."
Read the transcript:
Shawn Cole: The second experiment was conducted a couple of years earlier in Gujarat and there we were interested in understanding if this product is great what were the barriers to adoption. And so the project was motivated by a fact that when we introduced the product with a substantial substantial discount, take up was actually very low. So all insurance policies have a negative expected value—that is, on average, you will get paid less by an insurance company than the amount of money you pay to the insurance company. But we all buy car insurance or home owner’s insurance because if that very bad state of the world happens—our car gets totaled or our house burns down, it’s very important that we get a payout. What we did is that we introduced primers to the product and we offered them products with positive expected value. So for every one dollar of insurance policy they bought, they would expect on average to get more than one dollar paid out at the end of the term. And this in theory should be a very attractive product for the farmers. In one of the districts the return was as high as 80% over a three month period. If you annualize that that’s all of a sudden in the hundreds of percent return. And if you risk adjust, it’s even more attractive. Even when the policy was so attractive take up was less than 50%. And so we conducted a range of experiments to try to understand what limited demand for insurance policies. And you could certainly hypothesize a range of things: So this, is a new insurance policy for many people, they are not familiar with it; they are not familiar with the formal financial contract, they have never entered in a formal contract in life, so maybe they just don’t understand it. Perhaps trust is important. Perhaps liquidity is a concern. So we randomly assigned a range of treatments to try to unpack these hypotheses. To everybody to whom we sold insurance we provided sufficient education to explain what the product was, but we randomly assigned a fraction of households to get an additional education module which related millimeters rainfall to moisture in the soil. Read the whole transcript: Transcript_Shawn Cole 2_FNL.pdf