The main motivation for this paper is the recognition of the fact that asymmetric information is the form of moral hazard and adverse selection results in sizeable efficiency losses. These costs are passed back to producers in the form of excessively high premium rates and also passed back to the government via the crop insurance subsidy program. A secondary motivation stems from a recent debate in the literature regarding the specific effects of moral hazard on agricultural input use. Conventional wisdom suggests that moral hazard will induce producers to reduce input usage. A competing hypothesis has emerged which suggests that moral hazard may induce producers to increase their usage or risk increasing inputs. The main objective of this paper was to develop a model of agricultural insurance to understand why asymmetric information problems might exist and to compute and evaluate the relative program costs of agricultural insurance that can be attributed to moral hazard and adverse selection. These objectives are achieved by developing a theoretical model of agricultural insurance, and by conducting numerical simulations of the model. Simulation results indicated that insured farmers use less agricultural inputs than uninsured farmers in an attempt to maximize expected indemnities. Moral hazard was fould to be a significant problem only at higher coverage levels. Expected returns (in term of expected indemnities) to agricultural insurance were found to vary substantially between productivity (i.e., risk) types, and farmers were shown to recognize and respond to these differences. These results suggest that crop insurance is confronted with an adverse selection problem. Simulation results further indicated that program costs to a myopic insurer attributed to moral hazard and adverse selection could be substantial.