Dynamic Optimization of Area Revenue Insurance and Consumption Smoothing

This study examines the interaction between insurance, credit and liquidityconstraints using a stochastic dynamic model. A risk averse farmer whoseobjective is to manage both production and market risk is assumed tomaximize the expected utility of life-time consumption by using both arearevenue (AR) insurance and consumption smoothing subject to a creditconstraint. The dynamic programming is coded in MATLAB usingDDPSOLVE algorithm (Appendix a). DSSAT crop simulation model is usedto determine optimal irrigation strategy and to simulate farm level yield. ABayesian Model is used to estimate expected county-level revenue. Theapproach and results are illustrated via a numerical example using countylevel data from Mitchell County in Georgia. Results support the hypothesisthat liquidity constraints can have a large impact on optimal insurancedecisions, and that different levels of premium load, risk aversion level, andtime-preferences can also affect optimal insurance decisions under liquidityconstraints.