National Paternity Leave Policy Adoption and Public Gender Equity Beliefs: A Quasi-Experimental Study of African Countries

• In Event: Paid Leave as Health Policy: Program Design, Uptake, and Health Impacts

Abstract

Background: Paternity leave (or second-parent leave) in addition to maternity leave is increasingly becoming a priority for countries worldwide. About 63% of countries worldwide have national paid paternity leave, comparable to 98% of countries who have national paid maternity leave. While all African countries have established paid maternity leave, only about 60% have implemented paid paternity leave, creating an imbalance in caregiving expectations. Despite global commitments to gender equity through initiatives led by the United Nations and African Union, limited causal evidence exists on whether adopting paternity leave policies meaningfully shifts public attitudes toward gender equality.

Research Purpose and Data: The primary research question asks: What is the causal effect of adopting national paid paternity leave policies on public gender equity beliefs? To answer this, the study uses individual-level survey data from the Afrobarometer, spanning three rounds (2002–2013) across 12 African countries, comprising 55,958 adults. These data are linked with policy data from the WORLD Policy Analysis Center and additional country-level indicators, including GDP and unemployment rates from the World Bank. The analysis compares four countries that implemented paid paternity leave during the study period to eight countries that did not.

Methods: A quasi-experimental difference-in-differences (DiD) design with two-way fixed effects is employed to estimate the causal impact of policy adoption. This approach compares changes in gender equity beliefs before and after policy implementation between treated and control countries, while accounting for time-invariant country characteristics and broader temporal trends. The study further explores heterogeneity by examining concurrent workplace gender discrimination policies, duration of leave, respondent gender, and age groups. Given potential violations of the parallel trends assumption, a generalized synthetic control method is also considered to strengthen causal inference.

Results: Preliminary descriptive findings suggest that agreement with gender-equitable statements increased substantially in countries that adopted paternity leave, from 71.7% pre-policy to 82.5% post-policy, while remaining relatively stable in control countries. These trends indicate a potential positive association between policy adoption and gender equity beliefs, though a DiD regression will be needed to determine causal magnitude. In addition, I conducted a preliminary analysis of the primary DiD regression and the result is a significant treatment effect of OR=1.5 (p<0.01), however this is prior to the synthetic control method. These regressions will be further built out in the analyses, in addition to the secondary analyses of effect modification. 

Implications: This research is important because it evaluates whether policy interventions can influence not only behaviors but also societal norms. Understanding this relationship is critical in regions with large informal economies, where formal policy uptake may be uneven but symbolic policy changes could still shape public attitudes. The findings have significant implications for policymakers in worldwide considering the adoption or expansion of paternity leave. By demonstrating whether and how such policies shift gender norms, this study contributes to broader efforts to advance gender equity, improve family well-being, and inform evidence-based policy design.

Author

• Marie Salem, University of California - Los Angeles
  Presenting Author