The Demand Side of Africa’s Demographic Transition: Desired Fertility, Wealth, and Jobs

Sub-Saharan Africa (SSA)’s fertility transition is considered “unique” by demographers, as the number of children born per woman has been decreasing at a slower pace than in any other world region historically (Bongaarts 2017). This paper presents stylised facts, constructed from the Demographic and Health Surveys, on the role of demand for children in the region’s fertility decline. I find that fertility desires are substantially higher in SSA, particularly among poorer women: the socioeconomic gradient in ideal family size is twice as negative as in other low-and-middle-income regions. The slope of this gradient correlates with features of the occupational structure: poorer women are less likely to work for a wage in SSA, where wage work prevalence is negatively correlated with desired fertility. Finally, within SSA, an increase in the share of women working for a wage at the sub-national level flattens this gradient over time. Taken together, this set of facts raise a new question: to what extent do women’s fertility choices respond to the type of work opportunities available to them in SSA? To date, we know little about whether and how new types of jobs (e.g. through the modernisation of agriculture) shift demand for female labour, and how this affects female employment and fertility. In particular, salaried employment, which is arguably less compatible with looking after young children, might reduce fertility by raising the time cost associated with having children. To test this hypothesis, in an ongoing project co-authored with Niclas Moneke (Oxford), I exploit the emergence of the cut-flowers export industry – which disproportionately employs women, mostly on wage contracts – since the late 1990s in Kenya. To do this, we first geo-locate all existing flower-processing plants in rural Kenya and time their entry. Second, we assemble and geo-locate rich micro-data to track employment and fertility across women in affected locations. Combining both data sources will enable us to assess fertility responses to changes in demand for female labour generated by the opening of new flower-processing plants differentially across locations and over time. The second part of the seminar will introduce the data and identification challenges involved in answering this question in this setting.

Speaker: Céline Zipfel (Assistant Professor of Economics at the Stockholm School of Economics)