The Impact of Index-based Insurance on Informal Risk-sharing Networks under Moral Hazard

Moral hazard and adverse selection tend to impede the development of insurance markets, particularly in developing countries. Besides, it is largely acknowledged that informal risk-sharing mechanisms only offer a very imperfect substitute to formal markets. Indeed, on the one hand, local risk-sharing cannot absorb shocks of a common nature, on the other hand, they are themselves limited in scope by limited commitment and moral hazard. In this context, index-based insurance is seen as a very promising scheme. By construction, it is immune to moral hazard and adverse selection. However, the main drawback of index-insurance as compared to classical damage insurance is basis risk, namely a lack of (negative) correlation between payouts and the farmers’ income. This is precisely to cover this residual risk that a complementarity with informal risk-sharing is expected to operate. While this statement is technically valid, it ignores the potential effects on incentives and behaviors that the interaction between both schemes may generate. This paper explores the interaction between index-insurance and informal risk-sharing in a model with moral hazard and highlight that the formal contract may crowd out informal risk-sharing if it is offered to individuals rather than groups. More precisely, we find that the rate of informal risk-sharing is reduced as soon as individual demand for index-insurance is interior. Second, we highlight a reduction in risk-taking and welfare if the premium is too high, but demand still interior. This adverse effect is due to the worsening of the moral hazard issue that the formal contract produces. In the case of group subscription, the impact on moral hazard is internalized by the group. This result suggests that group subscription should be favored. Future research should however assess how robust this conclusion is to the introduction of group heterogeneity. Also, the case where the imperfection of informal risk-sharing is mainly due to limited commitment (instead of moral hazard) should be carefully considered. Indeed, intuition suggests that, by reducing the probability of catastrophic events, index-insurance may relax the incentive compatibility condition and actually improve informal risk-sharing in those cases, therefore leading to opposite predictions. Policy recommendations in terms of contract design should then be based on a careful empirical assessment of the functioning of pre-existing risk-sharing networks