Essays on crop insurance yield guarantee and product choice decisions



Journal Title

Journal ISSN

Volume Title



The U.S. Federal Crop Insurance Program has been continuously reviewed for issues relating to proper insurance guarantees and actuarially sound premium rates. Moral hazard and adverse selection were among the major identified issues considered in the past, while other issues include the yield guarantee level, rate making, and crop insurance participation. Two issues in existing crop yield insurance were identified for this dissertation research and are discussed and examined separately in chapters 3 and 4. The first issue is variability in the yield guarantee created by using a short, four to ten year yield history to establish the guarantee. Policy mechanisms that have been developed to mitigate this small sample problem are incorporated into the analysis. The second issue examined is spatial heterogeneity in crop insurance product choice decisions. A review of the literature in chapter 2 summarizes prior studies and clarifies the gap filled by the present work. Chapter 5 summarizes our findings and provides direction for future research.
The first essay (chapter 3) examines the issue of sampling variability in yield guarantees applicable for the APH-based crop insurance products offered under the Federal Crop Insurance Program. Crop yield and revenue insurance products with coverage based on actual production history (APH) yields dominate the program in terms of liability insured and premiums collected. The APH yield, which plays a critical role in determining the coverage offered to producers, is based on a small sample of historical yields for the insured unit. The properties of this yield measure are critical in determining the value of the insurance to producers. An APH yield based on a small sample of historical yields has the potential to lead to over-insurance in some years and under-insurance in other years. Premiums, which are in part determined by the ratio of the APH yield to the county reference yield, are also affected by variations in APH yields. Congress has enacted two measures, yield substitution and yield floors, which are intended to limit the degree to which small samples can reduce the insurance guarantee and producer welfare. We examine the impact of the small sample problem and related policy provisions for Texas cotton, Kansas wheat, and Illinois corn. The analysis is conducted using county level yield data from the National Agricultural Statistics Service (NASS) and individual insured-unit-level yield data obtained from the Risk Management Agency (RMA) insurance database. Our findings indicate that variation in insurance guarantees due to small samples in APH yields has the potential to reduce producer welfare and that the magnitude of this effect differs substantially across crops. The yield substitution and yield floor provisions mitigate the negative impact of small samples but also bias insurance guarantees upward, leading to increased government cost of the insurance programs.
The second essay (chapter 4) examines the spatial components of producer heterogeneity in crop insurance product selection among U.S. corn producers and identifies neighborhood spillover or agent marketing effects in these decisions. County-level insurance and yield data are used to demonstrate that a gradual shift from yield-based insurance to revenue-based insurance has spatial patterns. Conventional risk variables such as yield variability, price variability, prevalence of irrigation, other crops, and yield-price relationships play an important role in this shift and are consistently estimated only when spatial components are included. A spatial random effects model is used to identify the impact of spatial lag effects, which include neighborhood spillover, and potentially, agent marketing effects, on the share of corn acres insured with revenue-based versus yield-based crop insurance plans. Variables associated with risk are found to significantly influence the choice between crop revenue and yield insurance. Nonlinear parameters identify the region-specific effects on insurance decisions of factors including changes in percentage of irrigated acres, yield price correlation, and the percentage of corn acres to total acres. In addition, spatial components that might include decisions made by nearby producers or agent marketing drives are also found to influence decisions. These results may demonstrate the relative influence of trusted sources, such as nearby producers and insurance agents, on insurance decisions. Traditional risk variables are consistently estimated only by controlling for spatial heterogeneity.



US crop insurance, APH insurance, Small sample problem, Spatial modeling, Crop insurance decisions