A discrete choice analysis of smallholder farmers' preferences and willingness to pay for weather derivatives: evidence from Ethiopia

The ability of Ethiopian farmers to deal with rainfall risk is scanty due to the extension of land plots and incomplete and inefficient financial markets. Traditional drought insurance is flawed by information asymmetries, high administrative costs, and non-diversifiable risks. Insurance based on indexes is a promising alternative. Working on 120 rural households, we estimate the willingness to pay for a drought weather derivative through a mixed logit model allowing for random preferences. The results suggest that the premium, indemnity, and perceived frequency of drought are important determinants of the take-up. Apparent inconsistencies in behavior can be interpreted as rational choices.