Crop Abandonment: Effects of Weather, Irrigation, and Prices

Abstract

Cotton is an important U.S. agricultural commodity, generating about 200,000 jobs among the various sectors from farm to textile mill and accounting for more than $25 billion in products and services annually. A double hurdle model is estimated to assess the effects of weather extremes, irrigation, crop variety choice, and changes in cotton prices on rates of county-level cotton acreage abandonment. Acres abandoned are those planted with cotton, but not subsequently harvested. The first step of the double hurdle model estimates factors affecting the probability that a county will have at least some acres abandoned. County level data were available for eight states and three crop types: upland irrigated cotton, upland non-irrigated cotton and irrigated Pima cotton. Abandonment rates were highest among upland non-irrigated acreage. Seasonal temperature and precipitation variables were significant predictors of abandonment behavior. A sub-sample of the data from Census of Agriculture years contained a variable for the number of cotton farms in a county. The probability that a county had some, positive amount of acres abandoned increased with the number of farms. The rate of abandonment for counties with abandoned acres, however, declined with the number of cotton farms. This results provides some justification of a more flexible double hurdle specification over a tobit specification to crop abandonment.