Impacts of Spatıotemporal Dependency and Asymmetrıc Informatıon on The Analysıs of Optımal Crop Yıeld Insurance

Abstract

Having the most indispensable role in agricultural production, farmers need to protect them- selves against the risks arising from agricultural production in order to continue producing. It is very essential to provide an insurance policy aimed at meeting the farmer’s coverage need. Yield insurance is our specific interest since one of the most vulnerable segments of the society is farmers due to the fact that the farmers’ wealth is highly exposed to environmental risks. In order to discuss the equilibrium in an insurance market, we study the insurance demand in the frame of principal-agent problem which examines the optimal contract between the principal and the agent. The principal-agent problem that is widely investigated in economics and finance could be also used for insurance market to analyze the relationship between the insurer (prin- cipal) and the insured (agent). In this context, we mathematically deal with adverse selection which is one of the most essential concepts in insurance. Since individuals who need insurance have different expectations, preferences and justifications, it is very complicated to produce an insurance product which would satisfy all needs and cover several people. Therefore, it is very significant to analyze the demand for insurance accurately. In this thesis, we model the insur- ance demand in order to examine the equilibrium in insurance market. Having discussed the equilibrium for an insurance product generally, we study the asymmetric information concept in yield insurance specifically. Yield insurance covers the yield loss occuring when the yield does not exceed a specified yield level. This insurance product has a special place among agricultural insurance types because the aim of keeping the agricultural production at a determined level provides sustainability in the ecosystem. In this thesis, we examine the effects of moral hazard in yield insurance in order to optimize the farmer’s effort for both loss-prevention and loss-reduction. The impacts of moral hazard is also analyzed graphically and numerically by using certainty equivalent approach. Since the agricultural insurance is a government-backed system in Turkey, reducing risks related to asymmetric information leads the farmer to make an effort to search for more efficient ways of agricultural production. Climate change, which is the biggest challenge for agricultural insurance, is also investigated through spatiotemporal modelling of the crop yield in the thesis. We consider changes in extreme weather events such as flood, drought and hail by taking into account of location and time related effects on agricultural production. In this thesis, we use the hierarchical Bayesian approach for the conditional distribution of the yield to reflect the spatiotemporal dependency. In order to examine the district-based dependency, we consider spatial, temporal and spatiotem- poral effects on the estimation of the crop yield by using hiearchical Bayesian structure. After we select the best model by using performance criteria, we compute the premium rates related to various coverage levels.