Association between gender equity in the family and fertility intentions: an analysis of the Brazilian context using machine learning

Abstract

The literature on gender equity and fertility proposes a U-shaped curve to explain the relationship between these variables. As women achieve greater equity in the public sphere (e.g., through education and employment), their opportunity costs increase. Combined with their almost full responsibility for unpaid domestic work, this results in reduced reproductive intentions and, consequently, lower fertility. Reversing this downward trend would require greater equity in the private sphere, with increased male participation in domestic tasks. Several studies have examined the positive association between gender equity in the family and fertility or fertility intentions in high-income, low-fertility countries - the primary context for this model. However, few studies explore this relationship in lower-income, low-fertility countries like Brazil. This study is pioneering in investigating the association between gender equity in the family and fertility intentions in Brazil, at the micro level. Given the lack of a nationally representative data source that includes both variables of gender equity in the family and fertility intentions, our analytical approach combines data from the 2006 PNDS and the 2006 and 2015 PNADs, applying Machine Learning methods to calculate the predicted probability of intending to have the second child. Using this proxy variable, we adjusted a series of models, and, overall, the results indicate a negative relationship, contrary to the theoretical framework. This outcome may stem from how variables were operationalized. Alternatively, Brazil’s lag in the first phase of the Gender Revolution - advancing public sphere gender equity - might also explain the negative association observed in this context.

Keywords:
Intentions; Fertility; Gender; Equity; Family; Brazil; Machine learning