A. Introduction

Cotton area in South Africa fluctuates from year to year and can range from 40,000 hectares to 100,000 hectares or more. In 2001/02 South Africa grew 44,000 hectares of cotton, with an average yield of 430 kg lint/hectare for a total production of 19,000 metric tons (MT). In 2001/02 South Africa consumed 72,000 tons of lint and had to import 40,000 tons, twice the amount it produced (ICAC, 2002a). The cotton area in 2001/02 was lower than the previous year because low international prices of cotton led farmers to plant maize and sunflower, which were perceived to be more profitable.

Cotton is produced in South Africa under irrigated and rainfed agriculture. In 2001/02 there was a total of approximately 10,000 hectares of irrigated cotton and 22,000 hectares of rainfed (Table 45). Cotton is grown in 5 major regions: the largest irrigated areas are in Mpumalanga, Northern Province, Northern Cape and Orange River, whereas cotton is produced on rainfed/dry land in Northern Province, KwaZulu Natal and the North West.

Approximately 95% of cotton production is produced by approximately 400 large commercial farms and 5% by about 3,600 small farmers. In 2000/01 approximately 400 large commercial farmers produced 157,515 bales of cotton, whereas 3,300 small farmers in the Makhathini Flats and 300 small farmers in the Tonga region produced the balance of 7,300 bales. Several surveys have been conducted in the Makhathini Flats in KwaZulu Natal to characterize cotton production by small farmers who first adopted Bt cotton in 1998 (Ismael et al 2002, Kirsten et al 2002, Ismael et al 2001). Cotton production by small farmers in the Makhathini Flats has been promoted by the Vunisa cotton company which provides extension advice, inputs and manages credit. It is projected that eventually up to 4,500 farmers could produce cotton in the Makhathini Flats on up to 30,000 hectares of some of the best land in South Africa and contribute up to 30% of total cotton production in South Africa. The profile of a typical small scale farmer in the Makhathini Flats is a 40 plus year old man (52% of farmers) or woman (48% of farmers) farming 2.5 to 5.0 hectares of land, on which he/she grows less than 2 hectares of cotton as the principal crop (Ismael et al 2001), which is the major source of income for these resource-poor farmers. The bollworm complex is the major insect pest constraint in South Africa, which can cause severe damage and economic losses, for which Bt cotton offers good protection.

B. Adoption of Bt Cotton

The Bt cotton with the Cry1 A(c) gene developed by Monsanto/Delta Pine Land was field tested and introduced commercially to South Africa in 1998 with some pre-commercial areas planted in 1997 (Thomson 2002, James 1998). The area of Bt cotton has increased from 12,000 hectares in 1998 (10% adoption) to 20 to 30,000 hectares (45% adoption) by 2001/02. However, the striking feature about Bt cotton adoption in South Africa is the high adoption rate by small farmers in the Makhathini Flats and in the Tonga area. The number of farmers growing cotton in the Makhathini Flats increased from 75 in 1998 to 2,976 in 2001 (Table 46). Similarly, hectarage of Bt cotton increased from 80 hectares in 1998 to 5,670 hectares in 2001. The data in Table 46 for the Makhathini Flats confirm the very high rate of adoption from under 10% adoption of Bt cotton in 1998 to approximately 40% in 1999, 60% in 2000, and 92% in 2001. Adoption in the Tonga area was also at more than 95% in 2001(Kirsten et al 2002).

C. Yield Advantage of Bt Cotton

The yield advantage of Bt cotton for small farmers (49%), and large farmers under both irrigated (19%) and dry land (14%) conditions is consistent (Table 47). Surveys of small farmers (Ismael et al 2001) showed variable results in 1998/99 and 2000, with an average increase in yield of 24% over the two year period with a corresponding benefit of 29% in gross margins (Table 48). Another data set reports increases of 27 to 48% in yield of Bt cotton over non-Bt cotton (ISAAA 2002d). The yield advantage of Bt cotton will be mainly determined by the level of infestation of the bollworm complex; the higher the infestation, the greater the yield advantage with a gain of 25% to 50% being fairly representative over the last few years.

D. Reduction of Pesticide and Labor Requirements

The most recent and detailed data set for 32 small farmers in the Makhathini Flats in 2002 (Ismael et al 2002a) show that insecticide sprays were reduced by 7 sprays, from 11 to 4 sprays when using Bt cotton (Table 49). The corresponding cost savings from the reduced insecticide spray program was $45/hectare, with insecticide costs for non-Bt at $70/hectare and $25/hectare for Bt cotton. It is noteworthy that it takes one day for a person to spray one hectare of cotton, which involves 9 km of walking and the application of 120 liters of water which has to be carried over several kilometres. This saving of time for both men, women and children has important social implications. For example, half the farmers are women, who can use the time saved more profitably for household activities, caring for children, the sick, and other off-farm activities.

E. Health Implications - Reduced Insecticide Poisonings

Data on insecticide poisonings were obtained from hospitals in the Makhathini Flats where Bt cotton is grown, for a four year period starting in 1997/98 when Bt cotton was first introduced. The data covers the months of December to March, which coincide with insecticide spraying of cotton. The data in Table 50 indicate that there is a negative correlation between the incidence of insecticide poisonings and the percent adoption of Bt cotton. Thus, in 1997/98 when the % adoption of Bt cotton was only 0.1%, incidence of insecticide poisoning cases were 51. When Bt cotton adoption rates increased from 10% to 40% and 60% from 1998/99, to 1999/00, to 2000/01, the corresponding incidence for insecticide poisonings decreased from 30, to 14, to 6, respectively. Whereas this data provides an indication that there may be a relationship between the adoption of Bt cotton and insecticide poisonings, more research needs to be conducted to establish a causal relationship.

F. Overall Economic Advantage of Bt Cotton

Yield advantages for Bt cotton, coupled with insecticide and labor savings which are partly offset by higher seed costs, result in an overall economic advantage of the order of $50/hectare for small Bt cotton farmers in the Makhathini Flats (ISAAA 2002d). Increased seed costs for Bt cotton for small farmers can range from $8 to $15/hectare. The data of Ismael et al 2001, report Bt cotton seed at $20/hectare and non-Bt cotton at $12/hectare; a premium of $8 for Bt cotton seed. Data in Table 51 indicate that the premium for Bt cotton seed is $15/hectare for small farmers and can be as high as $54 for large farmers. The data in Table 51, which excludes savings in labor costs, show that Bt cotton provides an economic advantage of $35/hectare for small farmers, and ranging from $19 to $34/hectare for large farmers under rainfed and irrigated conditions respectively. Adjusting the data in Table 51 for savings in labor costs would result in an overall economic advantage of $40 to $50/hectare; the magnitude of the economic return will vary by year and be principally dependent on the infestation level of the bollworm complex.

G. Highlights

• South Africa grows 30,000 to 100,000 hectares of cotton per year. The planted area is influenced by the international price of cotton and planting conditions; cotton is grown under irrigated and dry land farming systems.

• About 95% of cotton production is produced by 400 commercial farmers, whereas the balance of 5% is produced by about 3,300 small farmers in the Makhathini Flats and 300 small farmers in the Tonga area.

• Bt cotton was first grown in 1998 and the hectarage is now 20,000 hectares or more, representing about 45% of the cotton area in 2001.

• Many surveys have been conducted in the Makhathini Flats of KwaZulu Natal to assess the impact of Bt cotton on small farmers. In 2001, 2,976 small farmers grew 5,670 hectares, representing 92% of the cotton area in the region.

• Small farmers growing Bt cotton in the Makhathini Flats gained through increased yields of 25% or more, decreased number of insecticide sprays (from 11 to 4 – a saving of 7 sprays), reduced pesticide costs ($45/hectare) and suffered less insecticide poisonings; the higher cost of Bt seed (up to $15/hectare for small farmers) resulted in an overall economic advantage of up to $50/hectare for Bt cotton. The time saving associated with fewer sprays has important social implications allowing women farmers (approximately 50% of farmers) to spend more time with family, caring for children, the sick and household activities.

• Bt cotton offers many advantages to the communal growers in the Makhathini Flats and Tonga regions of South Africa: less exposure and handling of insecticides; reduced probability of utilizing contaminated insecticide containers for drinking water; reduced contamination of wells and water sources with insecticides; lighter labor load for men, women and children involved in the arduous and hazardous task of insecticide spraying, leaving more time for important family household activities that are currently suffering from inadequate attention.