Addressing Practical Issues in Designing Weather Insurance Contracts for Risk Management Applications in Developing Countries

Abstract

In this dissertation we address practical issues in designing weather insurance contracts for risk management in developing countries in three different scenarios. First, we develop an innovative contract design strategy based on agronomic considerations that can be implemented in situations where only short and/or aggregate data series are available. We attempt to mitigate both the aggregate nature of yield data and the need for data-demanding analysis by looking at areas sharing the same growing conditions and using agronomic requirements to specify contract parameters. We find that the proposed contracts do not achieve the same degree of risk reduction as the contracts that can be constructed using no data limitations, but they do provide meaningful risk protection and typically at lower premiums. The implication is that the proposed methodology can be used to design weather derivatives for developing countries, where paucity of data often renders the conventional design approaches unworkable.

The second essay aims to derive a general-form optimal payoff of an index contract that takes into account potentially nonlinear dependence between the index underlying the contract and the loss that is insured. We find that the quasi-linear contract payoff structure may not be the optimal choice if the dependence between the index and the yield/revenue is nonlinear. The implication is that the proposed methodology can help to improve risk-reducing capabilities of weather derivatives particularly in situations where the effect of weather on yield is complex and not obvious.

The third essay analyzes the use of weather derivatives in managing water supply risk arising in making water allocation decisions. The specific application is developed for the Alto Rio Lerma Irrigation District (ARLID) in the state of Guanajuato in Mexico. We argue that incorporation of weather derivatives in water allocation decisions can improve overall well-being of producers and allow shift water allocations from the wet to the dry season with the assumption that the wet season farmers can cope with the risk of water shortages by using weather derivatives. We find that use of weather derivatives does lead to better water allocation policies that allow the representative farmer to reach higher levels of utility. The implication is that introduction of weather derivatives can help to improve water management decisions in developing countries where agriculture heavily depends on irrigation and can be severely affected by extreme weather events.