The Impact of Size on Yield Variability and Crop Insurance Premiums

Abstract

tions of enterprises, the benefits of diversification, and farmer response to different farm programs. Studies comparing the risk of different cropping practices have required data on farm yield variability (Williams, Llewelyn, and Mikesell; Mikesell, Williams, and Long). Crop insurance studies have needed the data for developing appropriate crop insurance rates (Carriker et al.). In some of these studies, availability of farm-level yield variability data has been a major problem. Frequently, other data have been used because they were the only data available. In particular, county-level data or experiment station data have been used as substitutes for actual farm-level data (Rowell, Williams, and Hickman; Williams, Llewelyn and Barnaby; Marra and Carlson). The accuracy of these data sources in representing farm-level data is not known. Frequently, county-level variability in yield is assumed to be less than farm-level variability, because it is more aggregated. On the other hand, variability in yield from experiment station data is sometimes assumed to be greater than farm-level yield variability, because the plots are usually small. The issues of aggregation level (Eisgruber and Schuman) and the impact of aggregation on variability are implicit in these assumptions about county and experiment station variability relative to farm-level variability. Marra and Schurle have recently addressed the possibility that aggregation on a farm can also influence yield variability on the farm. The implications of yield variability on the farm being related to the number of acres of the crop, and thus in many cases to the size of the farm, are far reaching. Pope and Prescott addressed the issue of diversification and farm size, and identified the trade-off between risk reduction and possible economies of size in a particular activity. Robison and Barry also addressed diversification and specialization as strategies for managing risk. If yield variability is less for larger acreage, and thus larger farms, then large farms may have a business-risk advantage (ceterius paribus) over small farms; i.e., business-risk economies of size. Then, it would also follow that farms with larger acreage should pay lower premiums per acre for crop insurance, other things being equal. In addition, model modifications may need to be developed that recognize the impact of enterprise size on yield variability. Current risk models typically consider variability and enterprise size to be independent. The Federal Crop Insurance Corporation (FCIC) has recognized a need to offer different premiums for different size acreage. They began offering a financial incentive to combine smaller insurance units into larger insurance units in the mid to late 1980's. A crop insurance premium reduction of 10 percent has been allowed if an acreage of one crop owned by one entity in one county is insured as a total unit rather than as smaller segments. Unit subdivision is allowed by section or by irrigated or dryland practice with supporting historical yield records of planted Bryan Schurle is a Professor in the Department of Agricultural Economics at Kansas State University, Manhattan, Kansas. This work was supported in part by the Federal Crop Insurance Corporation under agreement No. 92-EXCA-30203. The author thanks Jong-I Perng for her extensive computer work and patience throughout this project, Ken Harrison at FCIC for his provision of information and help, Art Barnaby and Barry Goodwin for help and review of previous manuscripts, three anonymous reviewers, and Ted Schroeder for his suggestions that improved the manuscript. Contribution No. 95-48-J from the Kansas