Três ensaios sobre trabalho reprodutivo, violência contra as mulheres e saúde mental

Abstract

The socially constructed differences between men and women manifest across various political, social, and economic spheres, leading to a range of gender inequalities. In this context, this Doctoral Thesis aims to investigate the interrelationship between the sexual division of labor, violence against women, and mental health. To this end, three essays were structured to progressively address these topics, demonstrating their interdependence. All studies used data from the 2019 National Health Survey (PNS). The first study examines the extent to which reproductive labor 3 understood here as the sum of household chores and caregiving 3 affects men’s and women’s labor market conditions, particularly their likelihood of holding a formal job, working part-time, or being self-employed. To address potential issues of endogeneity and selection bias, instrumental variable estimators and Probit models with selection were employed. The main findings indicate that both performing household chores and the time devoted to reproductive labor reduce the likelihood of formal employment and increase the probability of working part-time or being self-employed. These effects are observed for both genders but are more pronounced among women. The second study investigates whether paid work influences the occurrence of violence against women, examining the potential protective effect of employment, as proposed by the "marital bargaining theory", or an increased risk of victimization, in line with the "backlash theory". To mitigate the widely recognized endogeneity problem in the literature, instrumental variable models were applied. The results suggest that, when correcting for endogeneity, there is evidence of a protective effect of paid work. Finally, the third study connects the key discussions from the previous essays 3 reproductive labor and victimization 3 to women's mental health. Linear probability models and Probit models were estimated, along with robustness tests using the Propensity Score Matching methodology. The findings indicate that experiencing violence significantly increases the likelihood of exhibiting depressive symptoms, particularly moderate, severe, and very severe depression. Regarding reproductive labor, the results suggest that engaging in caregiving increases the probability of experiencing depressive symptoms, especially severe cases, while performing household chores does not consistently yield significant effects. In addition to the main analyses, all studies conduct heterogeneity analyses, highlighting the importance of considering sociodemographic differences among women and the distinct ways in which gender-related phenomena impact them.