Are Women Taking Over the Farm in China?

Development practitioners in the West have proclaimed that an increasing “feminization of agriculture” is occurring in the developing world. In this paper, I use household survey data collected in rural China to empirically test whether or not women have been doing an increasing amount of farmwork. I find exactly the opposite: if anything, the proportion of farmwork being done by women is declining over the late 1990s. Furthermore, I find that the future feminization of agriculture in China is unlikely. ∗Alan de Brauw is assistant professor, Department of Economics and Center for Development Economics, Williams College. I am grateful to Betsy Brainerd, John Giles, Songqing Jin, Bryan Lohmar, Scott Rozelle, Steve Sheppard, and Anand Swamy for useful suggestions. Are Women Taking Over the Farm in China? Development practitioners in the West have proclaimed that an increasing “feminization of agriculture” is occurring in the developing world. Publications by the FAO, the World Bank, the US Agency for International Development, and the Consultative Group on International Agricultural Research (CGIAR) have all recently proclaimed that women are doing an increasing proportion of farmwork in the developing world, and that policies should be designed to account for the feminization of agriculture (FAO, 2003; Buvinić, Gwin, and Bates, 1996; World Bank, 2002; Simmons, 2002; CGIAR, 2003). Yet, as some of these agencies or authors note, there is little empirical support for a feminization of agriculture hypothesis (e.g. Buvinić et al., 1996). These studies and others (e.g. Mehra and Gammage, 1999) base statements about the feminization of agriculture largely on aggregated data published by the United Nations that indicate women’s participation in agriculture and presence in rural areas fell more slowly than men in China, sub-Sarahan Africa, and South Asia between 1970 and 1990 (UN, 1995). However, these data do not prove a feminization of agriculture, as the men left in rural areas could work increasingly longer hours than women. Surprisingly, to my knowledge no one has used microeconomic data to test whether or not women are doing an increasing amount of farmwork anywhere in the developing world. From a theoretical perspective, two stories are consistent with women doing an increasing proportion of farmwork. If one assumes that separate off-farm labor markets exist for men and women, then an increasing wage for men relative to women would drive up the proportion of agricultural labor done by women; hence, a feminization of agriculture (e.g. Boserup, 1970). Alternatively, barriers to off-farm entry could be different for men and women, and loosening barriers for men relative to women would also increase the proportion of farmwork done by women. In West Africa, where men and women in rural areas farm separate plots (e.g. Doss, 1996; Goldstein and Udry, 1999) one could potentially test for the feminization of agriculture by testing whether net transfers of land from men to women were occurring over time. Elsewhere, one would ideally find a place with functioning off-farm labor markets in order to test whether or not agriculture is being feminized. In this paper, I will use household survey data to empirically test whether or not agriculture is being feminized in rural China. China makes a nice case study in which to test for the feminization of agriculture for several reasons. First, there is anecdotal and descriptive evidence that agriculture is being feminized in China.