This article revisits the issue of crop-yield distributions using improved model specifications, estimation, and testing procedures that address the concerns raised in recent literature, which could have invalidated previous findings of yield nonnormality. It concludes that some aggregate and farm-level yield distributions are nonnormal, kurtotic, and right or left skewed, depending on the circumstances. The advantages of utilizing nonnormal versus normal probability distribution function models, and the consequences of incorrectly assuming crop-yield normality are explored. Copyright 2003, Oxford University Press.

This paper develops and illustrates the application of a procedure to evaluate and compare the cost effectiveness of alternative crop insurance products for cotton in terms of their effect on expected producer net returns and the variation of net returns. Farm unit-level cotton yields and state-level price distributions are estimated by a multivariate nonnormal parametric modeling procedure and used to simulate the net returns to alternative crop insurance products over a 10-year planning horizon. The ranking of alternative insurance products using third-degree stochastic dominance is presented for Texas cotton producers.

No abstract