Assessing the Viability of Index Insurance as an Adaptation Tool in a Changing Climate Context: Case Study in the West African Sahel.

摘要

This dissertation contributes to the literatures on climate extremes, climate change, and financial adaptation. In a developing world context, weather based index insurance is emerging as a potential financial adaptation for poor populations that have historically been excluded from financial markets. Index insurance has potential drawbacks as well as benefits and there are a number of practical implementation challenges. However, as index insurance is sensitive to threshold crossing extreme event (TCE) frequency and as climate change renders the climate system non-stationary, there is a need to assess how the frequency of extreme events may change as the climate system evolves.;In an effort to address this research question for both hydroclimatological extremes (floods and droughts) and their associated risks, this dissertation explores the potential long-term viability of drought index insurance contracts for subsistence millet farmers and flood index insurance contracts for irrigated rice farmers in the West African Sahel nations of Mali, Burkina Faso and Niger. Potential hypothetical contracts were chosen on the basis of correlation analysis and Gerrity skill analysis between multiple potential geophysical indices and national crop production data.;Monte Carlo statistical methods are used extensively to simulate future streamflow and precipitation scenarios and are integrated with global climate model projections of precipitation. TCE frequency is found to be particularly sensitive to multi-decadal variability (MDV) and to changes in the mean (and less sensitive to changes in the variability). Moderate changes in the mean can have a more than two fold impact on the TCE frequency and multi-decadal variability typical of the region is shown to have a significant effect on the likelihood of a large number of TCEs in a specified time window. These changes to TCE frequency have important implications for both the actuarial and uncertainty related costs of index insurance over time; thereby creating challenges for long term index insurance viability. While specific results are simulation dependent, the more extreme scenarios indicate that there may be important limitations to the viability of index insurance in the future.