The wage penalty for motherhoodEvidence on discrimination from panel data and a survey experiment for Switzerland

Daniel Oesch, Oliver Lipps, Patrick McDonald•View original
HighlightsBalanceddamiaan voice
Four to eight percent. That's how much less a mother earns per child compared to a childless woman with the same qualifications. But here's the question that number can't answer: is that a productivity gap, or is someone deciding she's worth less? That's exactly what Oesch, Lipps, and McDonald set out to untangle in their study of Switzerland. The puzzle has two competing explanations. On the supply side, motherhood might affect productivity in ways standard controls miss, such as career interruptions and reduced on-the-job investment. On the demand side, employers may treat motherhood as a negative signal, penalizing women not for what they do, but for what recruiters assume they'll do. Switzerland makes this especially pointed: mothers, not fathers, remain the primary caregivers there, which makes both explanations plausible at once. To separate them, the study uses two different methods. The first is fixed-effects panel analysis. By tracking the same woman over time, the estimator removes stable unobserved traits like ambition and ability, and isolates the wage change that coincides with having a child. The researchers apply this to two Swiss longitudinal datasets: the Swiss Household Panel, covering 1999 to 2015, and the Swiss Labour Force Survey rotating panel, spanning 1991 to 2009. Together, these cover tens of thousands of women across hundreds of thousands of person-years. The second method flips the perspective entirely. The team ran a factorial survey experiment, which is a vignette study, with real human resources managers. Each recruiter evaluated twelve fictional but realistic résumés, with parenthood status randomized across otherwise identical profiles. When the only thing that changes between two résumés is whether a candidate has children, any wage difference in the recruiter's recommendation is causal. Panel data captures the supply side; the experiment isolates the demand side. That two-pronged design is the study's main methodological contribution. The panel results confirm that the gap is real. After controlling for human capital and job characteristics, the net wage penalty runs roughly three to four percent per child in the Labour Force Survey and six to nine percent in the Household Panel. A large chunk of the monthly wage loss is driven by reduced hours — about six to ten fewer hours per week per child — but an hourly penalty survives even after accounting for that. The experiment then shows who's creating part of that gap. Human resources managers recommended two to three percent lower starting wages for mothers on otherwise identical résumés. For younger mothers, those under forty, the penalty jumped to around six percent. Older mothers, aged forty-five to fifty-five, faced no penalty at all. Recruiters also rated mothers as less likely to be called for an interview: two children reduced that likelihood by zero point twenty-four points on a ten-point scale. Female recruiters were no less discriminatory than male ones. Together, the two methods make a strong case that the motherhood wage gap isn't just a productivity story. The authors caution that panel estimates may still capture some unobserved productivity differences that fixed effects can't fully remove, and the recruiter sample skews toward larger employers. But both methods converge on the same pattern — a persistent penalty for mothers, largest for younger women — pointing squarely at discrimination. The implication for policy is direct: equal-pay audits need to account for parenthood status, not just gender. This lecture was created by ennepō. Go to https://ennepo.ai to Discover, Create and Follow the latest research in your field. Read when you can. Listen when you want to.

Four to eight percent. That's how much less a mother earns per child compared to a childless woman with the same qualifications. But here's the question that number can't answer: is that a productivity gap, or is someone deciding she's worth less? That's exactly what Oesch, Lipps, and McDonald set out to untangle in their study of Switzerland. The puzzle has two competing explanations. On the supply side, motherhood might affect productivity in ways standard controls miss, such as career interruptions and reduced on-the-job investment. On the demand side, employers may treat motherhood as a negative signal, penalizing women not for what they do, but for what recruiters assume they'll do. Switzerland makes this especially pointed: mothers, not fathers, remain the primary caregivers there, which makes both explanations plausible at once. To separate them, the study uses two different methods. The first is fixed-effects panel analysis. By tracking the same woman over time, the estimator removes stable unobserved traits like ambition and ability, and isolates the wage change that coincides with having a child. The researchers apply this to two Swiss longitudinal datasets: the Swiss Household Panel, covering 1999 to 2015, and the Swiss Labour Force Survey rotating panel, spanning 1991 to 2009. Together, these cover tens of thousands of women across hundreds of thousands of person-years.

The second method flips the perspective entirely. The team ran a factorial survey experiment, which is a vignette study, with real human resources managers. Each recruiter evaluated twelve fictional but realistic résumés, with parenthood status randomized across otherwise identical profiles. When the only thing that changes between two résumés is whether a candidate has children, any wage difference in the recruiter's recommendation is causal. Panel data captures the supply side; the experiment isolates the demand side. That two-pronged design is the study's main methodological contribution. The panel results confirm that the gap is real. After controlling for human capital and job characteristics, the net wage penalty runs roughly three to four percent per child in the Labour Force Survey and six to nine percent in the Household Panel. A large chunk of the monthly wage loss is driven by reduced hours — about six to ten fewer hours per week per child — but an hourly penalty survives even after accounting for that. The experiment then shows who's creating part of that gap. Human resources managers recommended two to three percent lower starting wages for mothers on otherwise identical résumés. For younger mothers, those under forty, the penalty jumped to around six percent.

Older mothers, aged forty-five to fifty-five, faced no penalty at all. Recruiters also rated mothers as less likely to be called for an interview: two children reduced that likelihood by zero point twenty-four points on a ten-point scale. Female recruiters were no less discriminatory than male ones. Together, the two methods make a strong case that the motherhood wage gap isn't just a productivity story. The authors caution that panel estimates may still capture some unobserved productivity differences that fixed effects can't fully remove, and the recruiter sample skews toward larger employers. But both methods converge on the same pattern — a persistent penalty for mothers, largest for younger women — pointing squarely at discrimination. The implication for policy is direct: equal-pay audits need to account for parenthood status, not just gender. This lecture was created by ennepō. Go to https://ennepo.ai to Discover, Create and Follow the latest research in your field. Read when you can. Listen when you want to.

More in Social Sciences →

Related lectures