Premium estimation inaccuracy and the actuarial performance of the US crop insurance program

The purpose of this paper is to explore the impact of the levels of inaccuracy associated with three different premium estimation methods, one of which attempts to mimic the protocol currently used by the Risk Management Agency (RMA), on the actuarial performance of the US crop insurance program.

The analyses are conducted using empirically‐grounded simulation and other computational methods, under various plausible assumptions about the producer's risk aversion behavior and knowledge of his/her actuarially fair premium.

Regardless of the assumed producer knowledge and behavior, it is concluded that the persistently high government subsidy levels required to keep the program solvent could be solely explained by the inaccuracy in the RMA's premium estimates. In other words, the observed need for large subsidies does not necessarily imply that the program is systematically favoring less efficient farmers or particular crops or production areas. Also, contrary to the commonly accepted “adverse selection” argument, it is shown that farmers having more information about their actuarially fair premiums than the insurer is not the reason why high subsidies are needed. Actuarial performance, however, could be improved by using the more elaborate methods exemplified in the paper, as well as larger sample sizes for premium estimation.

The paper provides conclusions and recommendations that could substantially reduce the amount of public subsidies needed to keep the US crop insurance program solvent.