Alternative Insurance Indexes for Drought Risk in Developing Countries

The paper compares the risk coping potential of insurances that are based on indices derivedfrom weather (rainfall and temperature) data as well as from crop model and remote sensinganalyses. Corresponding indices were computed for the case of wheat production in the Alepporegion of northern Syria, representative for agricultural production systems in many developingcountries.The results demonstrate that weather derivatives such as the rainfall sum index (RSI) and therainfall deficit index (RDI) have a very good potential for coping with risk in semiarid areas.Crop simulation model index (CSI) on the other hand could serve as an alternative to RSI andRDI when historical farm yield data is not available or not reliable. In such cases we simulatedhistorical yields using the CropSyst cropping system simulation model.Remote sensing data could be used to establish index insurances where weather stations aresparsely located and (daily time step) weather data thus not available. The study analyzes twoindexes estimated from the Normalized Differential Vegetation Index (NDVI): (1.) the farm levelNDVI (FNDVI) and (2.) the area level NDVI (ANDVI). FNDVI may have a very high potential forsecuring farm revenues, but may be prone to moral hazard since farm management changes andsubsequent gains or losses in crop production are directly revealed by the NDVI when highresolution images are used. Therefore, we recommend ANDVI for developing countries since theindex is estimated for the whole agricultural zone similar to traditional area-yield insurances.