变革的种子：印度引进Bt棉花的影响

Seeds of Change The Effects of the Introduction ofBtCotton in India Yi Jie Gwee Development EconomicsDevelopment Research GroupDecember 2025 A verified reproducibility package for this paper isavailable athttp://reproducibility.worldbank.org,clickherefor direct access. Policy Research Working Paper11275 Abstract Although many studies have examined the factors affect-ing structural transformation, the role of relative factorendowments has not been extensively explored. This paperexamines the effects of the introduction of Bt cotton inIndia and finds substantial increases in cotton output andthe land allocated to growing it. The effects diminish as the ratio of labor-to-land increases. This suggests that wheninputs are complementary and technological change is laboraugmenting, the relative scarcity of other inputs can limitproductivity gains, shaping patterns of structural transfor-mation. Despite increases in cotton land and output, theshare of workers in agriculture remained unchanged. The Policy Research Working Paper Series disseminates the findings of work in progress to encourage the exchange of ideas about developmentissues. An objective of the series is to get the findings out quickly, even if the presentations are less than fully polished. The papers carry thenames of the authors and should be cited accordingly. The findings, interpretations, and conclusions expressed in this paper are entirely thoseof the authors. They do not necessarily represent the views of the International Bank for Reconstruction and Development/World Bank andits affiliated organizations, or those of the Executive Directors of the World Bank or the governments they represent. SeedsofChange:TheEffectsoftheIntroductionofB tCottoninIndia∗ YiJieGwee† 1Introduction Agriculture has long occupied a central place in theories of structural transformation. A commonlyheld view is that because genetically modified (GM) seeds improve agricultural productivity, la-bor will shift out of agriculture into manufacturing and services, transforming the structure of theeconomy. A vast literature has examined factors that influence these patterns of structural trans-formation but far less attention has been given to the role of relative factor endowments. In thispaper, I highlight that when technological progress is labor augmenting, the relative scarcity ofother inputs can limit the gains in productivity, shaping the direction of structural transformation.In addition to factor endowments, this paper also considers how the ability of labor markets toadjust to the new technology affects the process of structural transformation. The findings in thispaper suggest that technical change and structural transformation do not happen in a vacuum;they are shaped by factor endowments and labor market characteristics. The setting of this paper is the introduction ofBt cotton in India, one of the world’s largestproducers of cotton. With the spread of New World cotton hybrids that lacked resistance to Asianpests, farmers in India faced the prospect of falling cotton output. As a GM crop,Btcotton wasdesigned to combat pests, particularly, the American bollworm.It was developed by inserting agene from the soil bacteriumBacillus thuringiensis(hence “Bt”) into the cotton plant’s DNA,allowing it to produce proteins that are fatal when ingested by the bollworms. In 2002,Btcottonseeds were approved for sale in central as well as southern India and were subsequently approvedfor sale in northern India in 2005. The uptake of seeds rose dramatically and by 2013,Btcottonrepresented almost all the land used for the cultivation of cotton (Choudhary and Gaur, 2020). How do underlying endowments and labor market characteristics affect structural transformation?I present a conceptual framework to illustrate ideas and to introduce the mechanisms that willbe used to interpret the empirical results.While this paper treatsBt cotton as a form of laboraugmenting technical change, the mechanisms still hold if the technology is both labor and landaugmenting, as long as the labor augmenting component dominates. This stylized model highlightsthree insights. First, if land and labor tend to be complements in production and cotton is moreland-intensive than other crops, then labor augmenting technical change induces an expansion inthe (i) share of land allocated to growing cotton, (ii) cotton output, and the (iii) share of workersemployed in agriculture. Second, the effects vary by relative factor endowments.The intuition is that in districts withhigh labor-to-land ratios, land is the binding factor. Labor augmenting technical change effectivelymakes labor even more abundant relative to land.When inputs tend to be complements, thisamplifies the relative scarcity of land, leading to sharper diminishing returns, limiting the extent towhich cotton land share, cotton output, and the share of workers in agriculture are able to expand.In other words, these heterogeneous effects are