An analysis of regional and seasonal differences in the cotton basis

Cotton handlers can hedge their price risk using the cotton futures or options markets. These handlers then face basis risk. The objectives of this research are to determine if regional and seasonal differences in the basis exist, and to identify factors that may cause these differences.;Cash price data are available only by region. The monthly basis is calculated as the average monthly cash price in each U.S. cotton production region minus the average monthly settlement price for the New York Cotton Exchange July cotton futures contract. Data for August 1988 through July 1998 are used.;The non-parametric Friedman test is used to determine regional and seasonal differences in the cotton basis. As expected, the basis is found to be strongest in consumption regions, and weaker as the distance to consumption regions increases. The basis has a significant seasonal pattern in regions predominately supplying the relatively stable demand of domestic textile mills while no significant seasonality in the basis is found in regions that predominately supply the cyclical demand of foreign export. The statistical results indicate that the seasonal basis pattern becomes less pronounced as regions export larger percentages of cotton.;A theoretical basis model, consistent with the theory of storage, is developed to identify factors that cause the observed regional and seasonal differences in the basis. The theoretical model differs from other models because it covers multiple time periods and multiple regions and because it shows that current local and future national supply and demand shifters for consumption may explain the basis. The model bridges the gap between much theoretical and empirical basis work.;An empirical model, based on the theoretical model, is estimated using seemingly unrelated regressions. Large cotton stocks help explain the weaker basis in distant markets compared to consumption markets. Stocks explain the basis seasonality in storage markets. The basis is weakest at harvest when stocks are largest and strengthens as stocks decrease through the crop year. The empirical results also indicate that in storage (shipping) regions expected prices are more less important than current prices.