Definitions of variables used in the models.

The study explores gender participation in agricultural value chains, showing how roles are differentiated across genders and regions. The study fills an important gap in the literature which takes gender disparities as uniform and ignores subnational differences in farming and trading activities, with a focus on Africa’s most populous country, Nigeria. We estimated gender differences in participation, and compared welfare outcomes, using a combination of logit regression models, and instrumental variable estimation, that control for unobserved factors in the model. Household consumption expenditure is used as a proxy for welfare in the data, the most recent General Household Survey, covering more than 32,000 individuals. We find that men dominate farming, while women tend to dominate retail and wholesale trade, but with large regional differences that blur these distinctions. Being a woman implies a lower probability of farming, but almost doubles the likelihood of being engaged in wholesale and about triples that of retailing compared to being a man. The regional differences matter. In the southern regions of the country, women are more diversified across farming and trade activities relative to the north. In assessing the impact of the allocation of activities by gender, we show that farming households spend the least, while households with wholesalers have significantly higher expenditures.