We estimate the impact of genetically modified (GM) crops on countrywide yields, harvested area, and trade using a triple-differences rollout design that exploits variation in the availability of GM seeds across crops, countries, and time.
We find positive impacts on yields, especially in poor countries. Our estimates imply that without GM crops, the world would have needed 3.4% additional cropland to keep global agricultural output at its 2019 level. We also find that bans on GM cultivation have limited the global gain from GM adoption to 1⁄3rd of its potential.
Poor countries would benefit most from lifting such bans.
…We bring new evidence to this debate. At the core of our analysis is a triple-differences (DDD) rollout design in which we exploit that GM varieties of cotton, maize, rapeseed, and soybean became commercially available in 1996, whereas GM varieties of rice, wheat, and other important crops have yet to be commercialized. Countries that ban GM cultivation constitute another natural control group. We use the two control groups and the staggered national approval of commercial GM cultivation to identify causal effects of GM adoption in a 3-dimensional panel in which the unit of observation is a crop in a given country in a given year. We only include field crops in our analysis, whereby we exclude GM varieties of a few specialty crops, such as eggplant and papaya. This is no severe limitation, as cotton, maize, soybean, and rapeseed account for more than 98% of global GM production (ISAAA 2020).
…We find that cultivation of GM varieties statistically-significantly increases yields, particularly cotton yields. The yield gains are larger in countries with low incomes and many frost-free days, as warmer climates make pests and weeds more prevalent and poorer farmers have less resources to keep them in check (Oerke et al 199430ya; 2003). Like NASEM 2016, we find no effect of GM adoption on maize and soybean yields in countries with climates and incomes similar to that of the United States, but the null finding cannot be extrapolated to poorer countries with warmer climates. In a country like India, we estimate that nationwide maize yields could increase by as much as 64% if cultivation of GM maize was allowed. Soybean yields could increase by almost as much.
…Aggregating to the global level, we find that GM varieties increased the value of global agricultural production by about $39 billion in 2019, the last year in our sample. Without GM crops, the world would have needed 3.4% additional cropland to produce the same amount of output as in 2019, corresponding to an area the size of Spain. Only 1⁄3rd of the potential of the currently available GM varieties has been achieved, however. We find that without any bans on GM cultivation, the value of global agricultural production could have been a further $69 billion higher in 2019. Poorer countries, notably African ones, would have benefited the most. Not only have they most to gain in terms of yields, they also have large agricultural sectors. Lifting current GM bans could consequently support economic development of the poorest places on our planet while increasing agricultural production at a time when food security is a growing concern…A comparison to the realized gains reported above shows that while most of GM cotton’s potential has been realized, the opposite is true for GM maize. The difference comes down to regulation. Cotton is not used for feed or food, and growers can freely export GM cotton fibers to other countries without special permission.
