Strategic self-employment and family formation

Aperçu

Résumé (français)

The recent literature has highlighted the lasting impact of family formatio nfor a woman’s labourforce participation and earnings (Angelov et al., 2016; Kleven et al., 2019a). Even twenty years after a woman’s first child is born she is significantly less likely to participate in the labour market, while men show no evidence of a short- or long-run parent penalty on either the earnings or participation dimension. Kleven et al. (2009a, 2009b) argue that family formation explains what is left of the gender wage gap once you account for human capital differences, and differences in the shape and size of the parent penalty appear to be highly correlated with regional gender norms. These recent studies have been made possible through increased access to large administrative panel datasets. In this paper I provide similar evidence for Canada using the Longitudinal Administrative Dataset (1983-2016). Using information on the age of the seven youngest children in the family I identify the year that a woman has her first child and estimate changes in labour market outcomes in relation to this event. After corroborating the findings of Kleven et al. (2009a) I shift focus to the choice between wage-and self-employment. This shift is motivated in the following way. The recent literature has emphasized “flexibility” as a key barrier to female participation in the labour market after childbirth (Goldin, 2014; Goldin & Katz, 2016). Kleven et al. (2019a) show that women select into more flexible forms of employmentand even switch to ‘family friendly’ firms. Certainly, one would expect self-employment to demonstrate the same qualities given a self-employed worker’s ability to more freely allocate their working hours and location. Indeed, I show that there is an increase female self-employment of ~3% points following initial childbirth; an over 50% increase from pre-childbirth levels.

Résumé (anglais)

The recent literature has highlighted the lasting impact of family formatio nfor a woman’s labourforce participation and earnings (Angelov et al., 2016; Kleven et al., 2019a). Even twenty years after a woman’s first child is born she is significantly less likely to participate in the labour market, while men show no evidence of a short- or long-run parent penalty on either the earnings or participation dimension. Kleven et al. (2009a, 2009b) argue that family formation explains what is left of the gender wage gap once you account for human capital differences, and differences in the shape and size of the parent penalty appear to be highly correlated with regional gender norms. These recent studies have been made possible through increased access to large administrative panel datasets. In this paper I provide similar evidence for Canada using the Longitudinal Administrative Dataset (1983-2016). Using information on the age of the seven youngest children in the family I identify the year that a woman has her first child and estimate changes in labour market outcomes in relation to this event. After corroborating the findings of Kleven et al. (2009a) I shift focus to the choice between wage-and self-employment. This shift is motivated in the following way. The recent literature has emphasized “flexibility” as a key barrier to female participation in the labour market after childbirth (Goldin, 2014; Goldin & Katz, 2016). Kleven et al. (2019a) show that women select into more flexible forms of employmentand even switch to ‘family friendly’ firms. Certainly, one would expect self-employment to demonstrate the same qualities given a self-employed worker’s ability to more freely allocate their working hours and location. Indeed, I show that there is an increase female self-employment of ~3% points following initial childbirth; an over 50% increase from pre-childbirth levels.