Dynamic Optimization of Area Revenue Insurance and Consumption Smoothing

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