Polygyny is still practiced throughout much of sub-Saharan Africa, with important social consequences. This paper makes the first attempt to link African polygyny directly to the productivity of women in agriculture using micro data. I develop a structural model of the demand for wives that disentangles wealth and substitution effects. Using a large household survey from Côte d'Ivoire, I find that marked geographic diversity in cropping patterns leads to regional variation in female labor productivity. I also find that, conditional on wealth, men do have more wives when women are more productive, that is, cheaper. This substitution effect may explain why polygyny declined in rural areas of Côte d'Ivoire during agricultural development

Polygyny is still practiced throughout much of sub-Saharan Africa, with important social consequences. This paper makes the first attempt to link African polygyny directly to the productivity of women in agriculture using micro data. I develop a structural model of the demand for wives that disentangles wealth and substitution effects. Using a large household survey from Côte d'Ivoire, I find that marked geographic diversity in cropping patterns leads to regional variation in female labor productivity. I also find that, conditional on wealth, men do have more wives when women are more productive, that is, cheaper. This substitution effect may explain why polygyny declined in rural areas of Côte d'Ivoire during agricultural development