Environmental and cost efficiency of pesticide use in transgenic and conventional cotton production

This study focuses on the quantification of pesticide use efficiency for producers of transgenic cotton versus conventional cotton in order to test for the improvement promised by the genetically engineered crop. The environmental and cost efficiency of pesticide use is assessed by means of data envelopment analysis (DEA) and the external effects of pesticide are quantified by means of the pesticide leaching potential. To account for the fact that conditions other than the 'treatment" (seed type) are not equal in farm surveys the study employs a second step Tobit regression. The data are from a survey of cotton growers in North Carolina, USA. Differences in environmental efficiency are found to be significant between herbicide tolerant and stacked gene (herbicide tolerant and insect resistant) cotton and between stacked gene and conventional cotton. In contrast, no statistically significant differences are found for efficiency of pest control cost. In the follow-up Tobit regression, differences in production environment and in farm, farmer and field characteristics are accounted for so that the contribution of seed type to efficiency can be observed. The regression results confirm the importance of stacked gene cotton for improving the environmental efficiency of pesticide use in cotton. In contrast, seed type was not significant in explaining differences in cost efficiency among North Carolina cotton growers. The new technology reduces pesticide application but these benefits are curbed by the high price of cotton seed (technology fee). (c) 2006 Elsevier Ltd. All rights reserved.