The impact of weather on crop yield distribution in Taiwan: some new evidence from panel data models and implications for crop insurance

This study examines the impact of weather on the yields of seven major crops in Taiwan based on pooled panel data for 15 prefectures over the 1977-1996 period. The unit root tests and maximum likelihood methods involving a panel data model are explored to obtain reliable estimates. The uncertain yield outcome is incorporated into a discrete stochastic programming model to address a comparison between sector analysis with and without considerations of a crop insurance policy under different climate change scenarios. Simulation results suggest that crop insurance may stabilize revenues and protect farmers from exposures to increasing weather-related risk.