Area Yield Crop Insurance and Diversification in Ghana: An Agricultural Household Programming Model

Although diversification and insurance are acknowledged as important in ameliorating risk, to our knowledge, no empirical study has investigated how the two approaches work in concert in developing countries. Given increasing climate variability and the risks it poses to rain-fed agriculture, this study uses the Expected Utility Framework to develop two non-separable household mathematical programming and simulation models to examine the relationship between crop-diversification and index-based (area-yield) insurance for a representative agricultural household. The representative household is constructed using the IMPACT Lite household survey on Ghana, collected by the Consultative Group on International Agricultural Research Program on Climate Change, Agriculture and Food Security. In the base model, the household manages risk by diversifying their activities and allocating resources among six crop activities without the option of insurance. In the second model, the household has the option of both diversification and area yield insurance. Also, we assess how insurance coverage level and premium subsidies influence household participation and overall welfare benefits from insurance using sensitivity analysis. The results point to a highly risk averse representative household with a coefficient of absolute risk aversion of 0.016. The level of diversification increases with risk aversion in both the base and insurance models. Although insurance does not completely substitute for crop diversification, it reduces the degree of diversification. The degree of diversification for low and very high risk averse households is 23% and 35% more in the base model than in the insurance model, respectively. At 70% insurance coverage, the representative household insures all of its available land, regardless of risk aversion and premium subsidies. However, it only insures part of its available land at 90% coverage, due to higher premiums, and the portion under insurance decreases as risk aversion rises. For the household to insure all of its available land at the 90% coverage level, 80% of its premiums must be subsidised. Regardless of risk aversion, coverage level or subsidy, the availability of insurance increases the household’s expected value of consumption and thus increases welfare. Comparing welfare (as measured by the expected value of consumption) in the base and baseline insurance models (i.e., no subsidy and 70% coverage level), our results show that insurance increases welfare by approximate 45% and 55% for low and very high risk averse households, respectively. For policy makers, this study provides empirical evidence suggesting that even in the presence of diversification, introducing area yield insurance would help reduce the perceived riskiness of rain-fed agriculture and increase agricultural households’ welfare.