Introduction

Across high-income economies, there has been a major reduction in gender gaps in employment, earnings, and wage rates since the 1980s. International differences have also narrowed as gender gaps and child penalties in labor market outcomes declined. However, these gaps have fallen by less in the United States than in other countries. Concurrently, there has been a significant expansion in ‘family-friendly’ labor market policies, such as parental leave, government-subsidised childcare, and the right to request part-time work schedules. This research investigates the role of these policies in affecting gender gaps and child penalties across 27 countries, and evaluates why the reduction in gender gaps has been slower in the United States.

Motivating trends

The study reviews trends from 1986 to 2019, revealing considerable diversity in gender employment gaps. Southern European countries and Japan exhibit the largest gaps, while Nordic and Eastern European countries show the smallest. Between 1990 and 2019, the non-US average gender employment gap saw a sizeable decline of 17.5 percentage points, compared to a smaller decline of 7.5 percentage points in the US. This divergence reflects a robust increase in female employment/population rates outside the US. By 2019, the US employment gap moved from being below the non-US average to above it. Similar patterns are observed in earnings gaps, which fell by 0.425 log points on average across the sample.

Data

The research utilises microdata from the Luxembourg Income Study (LIS) and the International Social Survey Programme (ISSP) to obtain labor market outcomes for individuals aged 25–54. The final dataset includes 519 country-year observations from 27 OECD countries. To estimate child penalties, the sample was restricted to individuals aged 20–45, using 385 country-year observations from 22 countries. Policy variables, including weeks of parental leave and government childcare spending as a percentage of GDP, were merged from OECD datasets and legislative documents.

Empirical strategy

The authors employ a two-stage regression model to estimate the impact of within-country shifts in policy on gender gaps. The model controls for country and year fixed effects, overall wage inequality (measured by the standard deviation of male log wages), and composition-adjusted unemployment rates. To account for selection bias in wage data, the study implements a probability-weighted imputation approach to approximate latent wages for non-employed individuals. Child penalties are estimated using a pseudo-panel event study approach, matching parents to non-parents based on observable characteristics to determine the impact of childbirth on employment and earnings.

Results

The study finds that longer duration of parental leave generally lowers gender gaps in employment, wages, and earnings. However, this effect is non-linear; at very high levels of entitlement, further increases in leave raise gender gaps. Simulations suggest that moving from the average US level of parental leave (9 weeks) to the non-US average (81 weeks) would lower the gender employment gap by 5.0 percentage points. Childcare expenditures have a uniformly gap-reducing effect across all outcomes. An increase in childcare spending from the US average (0.232% of GDP) to the sample average (0.556% of GDP) implies a reduction in the gender employment gap of 2.37 percentage points and a decrease in the earnings gap of 0.053 log points.

Accounting for changes over time in gender gaps in employment, wages, earnings, and part-time employment

Policy shifts explained approximately 33% of the 17.0 percentage point reduction in the employment gap between 1990 and 2019. For the gender gap in earnings, policies accounted for about 23.5% of the observed decline. The slower expansion of family-friendly policies in the United States compared to other high-income countries accounts for 6–15% of the slower reduction in US gender employment and earnings gaps. While the US increased parental leave provision to 12 weeks by 2019, the non-US average rose to 88.5 weeks during the same period.

Separate results by education level

The impact of family policies varies significantly by education. For less-educated women, additional weeks of parental leave narrow gender gaps up to a threshold of 91 weeks for employment and 97 weeks for earnings. In contrast, for highly educated women, parental leave has little effect on the employment gap and may even raise the gender wage gap. These differences likely reflect the higher costs of career interruptions for more educated workers. Childcare expenditures, however, robustly reduce gender gaps for both education groups and are particularly effective at narrowing the gender wage gap for highly educated women.

Child penalty results

There is a strong positive relationship between the child penalty in earnings and the overall gender earnings gap. Between 1990 and 2019, the child employment penalty in the US fell by 8.5 percentage points, while the reduction outside the US was 19.6 percentage points. Expansion of parental leave entitlements and childcare expenditures were critical factors, accounting for 76% of the reduction in child employment penalties and 46% of the reduction in child earnings penalties across the sample.

Conclusion

Family-friendly policies have played an important role in the evolution of gender gaps since the 1980s. While parental leave and the right to request part-time work facilitate labor force attachment, they can involve trade-offs regarding wage gaps, particularly for highly educated women. Government expenditures on childcare consistently emerge as the most effective policy for reducing gender inequality across all labor market outcomes. The study concludes that the US has fallen behind other high-income countries largely due to its relatively slow adoption and expansion of these supportive policies.