A. Introduction

In 2001/02 the US grew 5.7 million hectares of cotton with a lint yield of 790 kg/hectare for a total production of 4.5 million MT. The US is by far the largest exporter of cotton (2.4 million MT) representing almost 15% of global exports of 19.9 million MT. The US has the second largest area of cotton in the world (5.6 million hectares) after India (8.7 million hectares) and is the second largest producer (4.5 million MT) after China (5.3 million MT).

Cotton is the fifth largest crop in the US by area and in recent years the value of the crop has been about $4 billion. Cotton is grown in the south and in the west in 16 states. Texas is the largest producer of cotton (30%), followed by Georgia. Cotton production systems range from low input rainfed cotton in Texas to the very intensive systems of Arizona, California and New Mexico. There is an extensive literature on Bt cotton in the US that includes several comprehensive reviews (Gianessi et al 2002, Carpenter et al 2002, Benedict and Altman 2001, Edge et al 2001, Carpenter and Gianessi 2001). The aim in this case study, is to provide a brief overview of Bt cotton in the USA.

B. Cotton Insect Pests and Crop Losses

Of the lepidopteran pests, the cotton bollworm (Helicoverpa zea), pink bollworm (Pectinophora gossypiella) and tobacco budworm (Heliothis viriscens) are the major pests of economic importance. These three pests are often called the bollworm/budworm complex. Boll weevil, Anthonomus grandis, is also important but a rigorous scheme is underway to eradicate this pest. Other secondary lepidopteran pests include beet and fall armyworms, loopers and cutworms.

The potential losses due to insects, in the absence of any form of control in the US is estimated at 38% and the actual losses that occur despite the application of control is estimated at 11% (Oerke 2002). More detailed surveys of losses due to cotton insects in the US indicate that crop losses range from 4.5% in 2001 to 11% in 1995 (www. msstate.edu/entomology/cotton). The average loss reported for the 18 year period 1979 to 1996 was 7.5% (Williams 1997b) with the bollworm/budworm complex reported to be the most important pest in 13 out of the 18 years.

For the period 1994 to 2001 cotton insect losses in the US have ranged from 4.5% in 2001 to 11.1% (Table 17) with the value of loss per hectare ranging from $65/hectare to $145/hectare with an average of approximately $100 per hectare. This translates to an annual loss of approximately $500 million at the national level in the US. In addition to the $ value of crop loss, the US data in Table 11 also includes the cost of control, the major portion of which is insecticides and their application. During the period 1994 to 2001, control cost ranged from $113/hectare to $158/hectare. Taking into account both crop loss and the cost of control, which is the total cost to US farmers associated with cotton insect pests, this ranged from $187/hectare to $293/hectare which is substantial; these translate to annual losses in the US due to insect pests of $1.0 billion to $1.7 billion.

C. Adoption of Bt Cotton

Bt cotton (Cry1Ac), developed by Monsanto and Delta Pine Land, was introduced in 1996, principally to control the three major pests: tobacco budworm, cotton bollworm and pink bollworm. In the US, in the mid south and south east, cotton bollworm and tobacco budworm are the most prevalent pests, whereas pink bollworm is the most prevalent in the western states. Before the introduction of Bt cotton in 1996, 75% of the cotton area was treated with insecticides and an average of 2.4 sprays were specifically applied to control the bollworm/budworm complex which was estimated to cause a loss of 4%, despite the application of insecticides (Carpenter and Gianessi 2001). In 1995, the year prior to the introduction of Bt cotton, tobacco budworm infestations were particularly high causing estimated losses of 29% in Alabama (Williams 1996). This was due to the development of resistance to the insecticides used.

The increase in adoption, depicted in Table 18, indicates a high rate of adoption starting from 14% in 1996 to 34% in 2001 (Edge et al 2001); these estimates are based on USDA/NASS data, whereas USDA.AMS data indicate that the percentage Bt in 2001 was 39%. Over 2 million hectares of Bt cotton were grown in the US in 2001; they include varieties with the single Cry1Ac Bt gene (10% of all Bt cotton) and varieties with the stacked genes of Bt and herbicide tolerance (90%).

D. Yield Advantage of Bt Cotton

Extensive field trials in the US report a range of results indicating that on average, Bt cotton will yield 10% or more than conventional varieties (Perlak et al 2001). In a 55 field comparison Kerby (1996) reports an average increase of 18% with a range of 15 to 21% (Table 19). Benedict and Altman (2001) report an average yield increase of approximately 14%, equivalent to 175 kg/hectare of lint.

E. Reduction of Insecticides

Bt cotton has led to a consistent decrease in the number of insecticide sprays required. Benedict and Altman (2001) estimate that the overall average reduction is 2.2. This is consistent with other estimates which calculate a reduction of approximately 2. In terms of active ingredients, Benedict and Altman calculated that in 1998 this translated to a saving of 1.09 million kg (a.i.) of insecticide equivalent to 1,090 MT on 1.1 million hectares of Bt cotton. The estimate of Carpenter and Gianessi (2001) for 1998 is of the same order as Benedict and Altman (2001) at 907 MT. Savings of insecticides for the three years 1998, 1999 and 2001 are detailed in Table 20 (2000 data not available). The data indicate that on average about 1,000 MT of insecticide (a.i.) was saved annually. For 2001, when infestation of lepidopteran pests was the lowest in recent years, 848 metric tons of insecticide (a.i.), equivalent to 0.45 kg/hectare was saved (Gianessi et al 2002, Carpenter and Gianessi 2001).

There are secondary benefits associated with the deployment of Bt cotton, requiring less insecticide, which include the following: more favorable environment for non-target predatory insects that act as biological control agents for secondary pests; less insecticide pollution of soil; less contamination of water sources and aquifers with insecticide run off; more favorable environment for wild life, e.g. birds that depend on insects for food; less packaging for insecticides reducing waste; lower application costs; more flexible insect control programs, etc.

F. Overall Economic Advantage of Bt Cotton

The yield increases and decreased costs of insect control associated with Bt cotton are partially offset by higher costs of seed for Bt cotton. Taking these factors into account, overall economic benefits for farmers in 1998 were estimated at $84/hectare equivalent to a national benefit of $92 million. Similarly, for 1999 the economic advantage of Bt cotton was $52/hectare for a national benefit of $99 million (Table 20) and for 2001 it was $50/hectare for a national benefit of $103 million (Carpenter and Gianessi 2001;Gianessi et al 2002). These estimates are of the same order as EPA estimates that range from $60 to $ 126 million annually. Thus, in summary the overall economic benefit for Bt cotton growers in USA is $50/hectare to $85/hectare, (after deducting additional costs related to seed and insect protection), which translated to approximately $100 million/year for the 2 million hectares of Bt cotton in the US in 2001 at current world prices of cotton.

G. Distribution of Benefits

One of the “corporate concerns” often voiced by the critics of biotechnology relates to their perception that the developers of GM crops, usually transnationals, are the major or sole beneficiaries from GM crops. On the contrary, analyses of Bt cotton in the US over the 3 year period 1996 to 1998, consistently show that farmers have been the major beneficiaries. Data in Table 21 indicate that farmers gained 59%, 42% and 46% of the total surplus in 1996, 1997 and 1998 respectively, compared with 21%, 35% and 34% for the technology developers and only 5%, 9% and 9% for the seed supplier. Thus on average farmers capture approximately 50% of the surplus compared with 30% for the technology developers. The share of benefit to consumers can be expected to increase, as the higher Bt cotton yields will increase supply and decrease prices, thereby providing consumers with more affordable cotton.

H. Highlights