The Motherhood Wage Penalty: A Meta-Analysis

Abstract

Mothers tend to receive lower wages than comparable childless women. This 'motherhood wage gap' has been reported in numerous studies. We summarize the existing empirical evidence on this topic using meta-analysis and test for several mechanisms which can be responsible for the persistence of the wage gap. Based on 208 wage effects of having exactly one child and 245 wage effects of the total number of children, we find an average motherhood wage gap of around 3.7 percent. While the gaps associated with the total number of children are mostly explained by the loss of mothers' human capital during child-related career breaks, the gaps associated with one child are predominantly driven by mothers' choice of jobs and occupations that pay less. The residual gap is smallest in Nordic countries, where public policies actively support gender equality and reconciliation of work and family, as well as Belgium and France, and largest in the post-socialist countries of Central and Eastern Europe and Anglo-Saxon countries.

Evaluating the wage differential between the formal and informal economy: a gender perspective

“…The following personal and household characteristics are included, given that wage levels have been previously identified as significantly associated with marital status (Bardasi and Taylor, 2008; Datta Gupta et al , 2007; Killewald and Gough, 2013; Watson and McLanahan, 2011), age and education level (Becker, 1975; Borjas, 2005) and the presence of children and elderly in household (Adda et al , 2017, Becker, 1975; Kleven et al , 2019, Cukrowska-Torzewska and Matysiak, 2018): Marital status – a dummy variable for marital status is included, taking a value of 1 for married individuals and zero otherwise Age – to assess for the non-linear relationship between age and wages, the model includes a variable measuring age of the employee and the age square. Educational level – three dummy variables are included in the model: a group 2 dummy variable with value of 1 if a person's highest completed education was secondary or post-secondary vocational education and zero otherwise; group 3 is set to 1 for persons whose highest educational attainment was a completed graduate tertiary degree, zero otherwise; and group 4 represents individuals that have completed post-graduate tertiary degree. The reference category refers to employees who have not completed any schooling or whose highest completed education was primary or lower secondary education. Children – in line with previous studies, the model includes a measure for the presence of children, with a binary variable indicating if the household has children under the age of 15. Presence of elderly people in household – a dichotomous variable is included identifying individuals that live in households with members aged 65 or older.…”

Gender and Gender-Role Attitudes in Wage Negotiations: Evidence from an Online Experiment

“…Discrimination in job assignments or in refusal to hire no doubt contributes to the gender gap in pay (Tomaskovic-Devey and Avent-Holt 2019). Gender differences in pay also arise from childbearing, or the "motherhood penalty" (Correll et al 2007;Cukrowska-Torzewska and Matysiak 2020). And a number of other factors besides GRAs influence gender differences in negotiation outcomes, most notably fairness concerns (Auspurg et al 2017), the activation of stereotypes (Kray et al 2001;Kray et al 2002), and fears of potential backlash among women (Amanatullah and Morris 2010;Amanatullah and Tinsley 2013;Wade 2001).…”

Beyond measurement of the motherhood penalty: How social locations shape mothers' work decisions and stratify outcomes

“…Using linked employer‐employee data from Canada, Fuller (2018) found support for one component of the compensating differentials theory: mothers' disproportionate representation in firms that pay less, but offer hypothesized mother‐friendly attributes like part‐time work. Controlling for job characteristics such as public sector work and occupation type reduced the penalty by two percentage points in a cross‐national metanalysis (Cukrowska‐Torzewska & Matysiak, 2020). Budig and England (2001) found small effects when incorporating job characteristics, while Leonard and Stanley's meta‐regression (2020) found little effect for part‐time, occupation, or industry variables.…”

“…Testing for selection effects among mothers yields conflicting results. Mothers do have lower levels of education than childless women overall (Jee et al., 2019), but among employed mothers, the selection effect works in the opposite direction; controlling for selection into employment lowers, rather than raises, the wage gap (Cukrowska‐Torzewska & Matysiak, 2020). Staff and Mortimer (2012), analyzing longitudinal employment patterns of women, find only very small differences between future mothers and those who never have children in the amount of time spent out of the labor force (and not in school) prior to motherhood.…”

“…While only 48% of women without a bachelor's degree are married at the time of their first birth, among college graduates the share climbs to 91% (Byker, 2016). Research suggests that education similarly shapes the financial consequences of motherhood, with broad empirical consensus that more highly‐educated mothers bear smaller penalties—or avoid them altogether (Amuedo‐Dorantes & Kimmel, 2005; Anderson et al., 2003; Budig & Hodges, 2014; Cukrowska‐Torzewska & Matysiak, 2020; Pal & Waldfogel, 2016).…”