Recently, the Federal Crop Insurance Program (FCIP) has come under fire from both popular press (e.g. Nixon 2012), the academic press (e.g. Hennessy 2011; Wright and Wimberly 2013), and this criticism is reflected in government publications (e.g. Shields 2012; US GAO 2007). The common argument is that subsidized crop insurance encourages expansion of cropping onto otherwise unsuitable land. In particular, the argument equates low productivity or high yield risk with environmental sensitivity (Nixon 2012; Hennessy 2011). In part, this conflation is due to the concern over land use change from the Conservation Reserve Program (CRP) to cropping. CRP was, by construction, designed to take low productivity land out of production. In addition to CRP conversion, concern has focused on grassland conversion to crop land (Archer 2011; Johnston 2011; Hennessy 2011; Wright and Wimberly 2013). The United States Government Accountability Office recognized the land use change and the possibility that federal programs were influencing land use choices and recommended that the executive branch investigate. The environmental effects of increased cropping use of environmentally sensitive land include erosion (which would lead to increased nutrient and sediment pollution), carbon release in conversion and reduction in carbon sequestration in future This paper will empirically test the assertion that crop insurance, broadly stated, is a causal factor in increased cropping land use. Data from The United States Department of Agriculture’s National Agricultural Statistics Service (NASS), The United States Department of Agriculture’s Risk Management Agency (RMA), The National Oceanic and Atmospheric Administration’s National Climatic Data Center, The Board of Governors of the Federal Reserve System, and The Federal Reserve Bank of Kansas City was compiled in Stata® for use in this analysis. This data was collected covering the period between 1981 and 2013 for Colorado, Kansas, Nebraska, Wyoming, Oklahoma, and Missouri. In 2013, these states grew twenty percent and eighteen percent of the corn and soybean, respectively, production in the US. Corn and Soybean production made up fifty and twenty-two percent of these states total crop production. These states are important to national production of corn and soybeans and these two crops are important to these states. If we accept that environmentally sensitive and marginal land are the same, as the press suggesting the existence of a link propose, one should expect that county yields would fall with higher levels of insurance participation. Further, if the effect is linear one would expect a statistically significant negative link between the proportion of land enrolled in crop insurance and county yields. The obvious dependent variable would look at acres, however with the length of CRP contracts, crop rotation, and the possibility of grassland or woodlot conversion to cropland suggested that the dependent variable would have to measure the quality of production rather than the acres carried out on. Supposing the relationship to be linear, a robust random effects generalized least squares regression model was built to test for the relationship between enrollment and yield. Results indicate that the link between crop insurance participation and yield is negative and statistically significant. It does appear that crop insurance participation has something to do with increased cropping of environmentally sensitive land, but further research is needed to confirm this result.
