Abstract:
How sensitive are married women to what they could potentially earn when deciding whether to enter the labour market? How sensitive are they to what their husbands earn? And, what do these sensitivities look like under prevailing conditions of high and persistent female unemployment? Answering these questions can potentially tell us something about female labour force participation as well as how conjugal couples divide up paid and unpaid work. A well-established result from the developed world is that strong own- and cross-price participation elasticities for married women's labour supply with respect to their husband's wages have attenuated towards zero over the past 50 years. Recent evidence from the US suggests the own-price effect in particular has then levelled out in the decade of the 2000s. We undertake this analysis for South Africa, an upper-middle-income country characterized by high levels of female unemployment and wage inequality. We focus on married (or cohabiting) women co-residing with their partner since this group has been the classic subject of study in this literature. Using 20-years-worth of timeseries cross-sectional household survey data, we find own- and cross-price elasticities shrank towards zero in the early 2000s before levelling out in the 2010s. The change in the direction of the trend in the 2010s is most prominent amongst groups with relatively more labour market advantage in South Africa (highly educated; Whites). A counterfactual simulation exercise suggests an important driver of this change appears to be increasing educational attainment in general.
