Impacts of copula modeling and parametric variation on revenue policy premium rates in multiple peril crop insurance

Abstract

Federally subsidized multiple peril crop insurance is the primary mechanism by which U.S. farmers receive public income. This study investigates the role of copula modeling in developing revenue product premium rates for multiple peril crop insurance. Simulation and empirical experiments are used to examine the viability of a ratemaking practice that relies on an assumed Normal copula. This study shows that the assumption of a copula cannot be statistically justified and that premium rates generated within copulas and between alternative copulas can diverge as a function of the marginal price and yield distributions, their relationship and the level of protection a producer elects. The current ratemaking practice does not account for the imprecision of premium rates implicit to a copula based approach. A copula selection method is proposed and examined in order to reduce premium rate imprecision resulting from copula misspecification. A non-copula based ratemaking method may better meet the overt policy objectives of multiple peril crop insurance.