No abstract

The results of a survey of 100 smallholders in the Makhathini Flats region of KwaZulu-Natal, South Africa give cause for cautious optimism regarding the impact of Bt cotton. The farmers who adopted the Bt cotton variety in the 1998 and 1999 seasons benefited from the new technology, according to all the measures used. Average yield per hectare and per kilogram of seed was higher for adopters than for non-adopters. The increase in yields and reduction in chemical application costs outweighed the higher seed cost, so that gross margins were also considerably higher for adopters in the second season. This was a bad year, due to unusually heavy rainfall, and the Bt adopters suffered far less fall in yields than those who did not adopt. Both yields and gross margins are partial measures of efficiency, which fail to take account of major inputs such as labour. Thus, they are supplemented by deterministic and stochastic efficiency frontiers, which consider the efficiency with which all inputs are converted into outputs. These methods use only the more reliable input and output quantity data and avoid prices, which are less well recorded or simply non-existent. Both methods confirm the farm accounting results, showing that the Bt cotton adopters were considerably more efficient than those who used the non-Bt varieties. For 1998, the stochastic frontier results showed that the adopters averaged 88% efficiency, as compared with 66% for the non-adopters. In 1999, the equivalent figures were 74 and 48%. Similarly, the determinist frontier results for both years show that the adopters were over 62% efficient, while the non-adopters averaged only 46%. Finally, there is no evidence that the better-off farmers gained more than the less well off: indeed, income inequality was slightly reduced.

The results of this survey of 100 smallholders in the Makhathini Flats, KwaZulu-Natal give cause for cautious optimism regarding the impacts of Bt cotton. Farmers who adopted Bt cotton benefited from the new technology, according to all the measures used. Average yield per hectare and per kilogram of seed was higher for adopters than for non-adopters. The increase in yields and the reduction in chemical application costs outweighed the higher seed cost, even in a poor production season due to unusually heavy rainfall. Bt adopters suffered far less of a fall in yields than those who did not adopt. As yields and gross margins are partial measures of efficiency, deterministic and stochastic efficiency frontiers were measured. Both methods confirm the farm accounting results, showing that Bt cotton adopters were more efficient. For 1998, the results showed that adopters averaged 88% efficiency, as compared with 66% for non-adopters. In 1999, the equivalent figures were 74% and 48%. Similarly, the determinist frontier results for both years show that adopters were over 62% efficient, while non-adopters averaged only 46%.