Simultaneous Adoption of Herbicide-Resistant and Conservation-Tillage Cotton Technologies

Journal of Agricultural and Applied Economics, Dec 2006 by Roberts, Roland K, English, Burton C, Gao, Qi, Larson, James A

If adoption of herbicide-resistant seed and adoption of conservation-tillage practices are determined simultaneously, adoption of herbicide-resistant seed could indirectly reduce soil erosion and adoption of conservation-tillage practices could indirectly reduce residual herbicide use and increase farm profits. Our objective was to evaluate the relationship between these two technologies for Tennessee cotton production. Evidence from Bayes' theorem and a two-equation logit model suggested a simultaneous relationship. Mean elasticities for acres in herbicide-resistant seed with respect to the probability of adopting conservation-tillage practices and acres in conservation-tillage practices with respect to the probability of adopting herbicide-resistant seed were 1.74 and 0.24, respectively.

Key Words: Bayes' theorem, conservation tillage, cotton, genetically modified crops, herbicide-resistant crops, simultaneous logit model, technology adoption

JEL Classifications: Q12, Q16, Q24, O33

Herbicide-resistant BXN (Buctril-resistant) cotton seed was first introduced in 1995 by the Stoneville Pedigreed Seed Company (Ward et al. 1995) and Roundup-Ready cotton seed became commercially available in 1996 (Johnson 1996). The adoption of herbicide-resistant seed by farmers has dramatically changed cotton production practices with potential consequences for the environment. Monsanto claims that adoption of herbicide-resistant seed facilitates adoption of conservation tillage, which "sustains the environment." Soule, Tegene, and Wiebe used data from the 1996 USDA Agricultural Resource Management Survey (ARMS) and with logit analysis evaluated the effects of land tenure on adoption of conservation-tillage practices. Although data from the 1996 ARMS were available for adoption of herbicide-resistant crops (Fernandez-Cornejo and McBride 2002), adoption was low, and Soule, Tegene, and Wiebe were not intent on evaluating the synergy between adoption of herbide-resistant seed and conservation-tillage practices. Fernandez-Cornejo and McBride (2002) used 1997 ARMS data and a two-equation simultaneous probit model to evaluate this potential synergistic relationship. Contrary to Monsanto's claim, they found no evidence that soybean farmers who had adopted herbicide-resistant seed had a higher probability of adopting no-tillage practices than farmers who had not adopted herbicide-resistant seed. They found evidence supporting the converse, however; farmers who had adopted no-tillage practices had a higher probability of adopting herbicide-resistant soybean seed than farmers who had not adopted no-tillage practices. Lack of simultaneity most likely resulted from using cross sectional data for the year after herbicide-resistant soybean seed was first introduced, leaving little time for adjustment in tillage practices. Using data from a 1999 survey of cotton farmers conducted in South Georgia, Ward et al. (2002) found evidence based on efficiency measures that farmers may have incentive to simultaneously adopt herbicide-resistant seed and conservation-tillage practices. Marra, Piggott, and Sydorovych found that 76% of North Carolina corn, soybean, and cotton acreage in herbicide-resistant seed was produced with conservation-tillage practices in 2001, while only 64% of corn, soybean, and cotton acreage in conventional seed was produced with conservation-tillage practices. Their specific results for cotton were different, with these two percentages being about the same at close to 73%.

Findings from the aforementioned cross-sectional analyses suggest a simultaneous relationship may exist between adoption of herbicide-resistant seed and adoption of conservation-tillage practices, but the evidence is inconclusive, especially for cotton. Sufficient annual time-series data are now available to investigate the relationship of the adoption of these two technologies over time. The Conservation Tillage Information Center (Fawcett and Towery) used a limited time-series sample of percentages of acres in glyphosate-resistant crops by tillage method for 1998 through 2000 and a 2001 survey by the American Soybean Association to suggest a simultaneous relationship between adoption of glyphosate-resistant crops and conservation-tillage practices in the United States. The information for Tennessee cotton acreage in Figure 1 (Doane Marketing Research, Inc.; Monsanto; Tennessee Department of Agriculture, 1996-2003, 2004) also suggests a relationship between adoption of herbicide-resistant seed and conservation-tillage practices. From 1992 through 1998, the share of Tennessee cotton acreage in conservation-tillage practices averaged 38% with no discernable trend. In 1999 when adoption of herbicide-resistant cotton seed started in earnest, the share of conservation-tillage acreage began a slight upturn to 40% with a dramatic increase thereafter, averaging 76% between 2000 and 2004 and reaching almost 100% in 2003 and 2004. During the 1992 through 2004 period, cotton acreage in Tennessee showed no perceptible trend, except during the mid-to-late 1990s when the Boll Weevil Eradication Program was active in middle and southwestern Tennessee (Suarez, Larson, and English 2000). Because of eradication program costs, farmers had an incentive to switch to other crops during the active phase of the program. Cotton acreage was relatively stable after 1998, when herbicide-resistant seed and conservation-tillage practices were being rapidly adopted.