Abstract
Numerous empirical studies show that mothers are often paid less than women without children, who have similar profiles. This paper presents the results of a meta-analysis of existing estimates of the motherhood wage penalty (over 2,000 estimates on data for 38 countries).
A statistically significant publication bias was found towards a higher penalty estimate. However, the motherhood penalty persists even after correcting for the bias. Many sources of heterogeneity of the current estimates of the motherhood penalty were shown to exist, including technical characteristics of the model, precision of the estimate, and inclusion of information on women’s human capital, employment, and other individual characteristics in the regression model.
The analysis confirmed the significance of such sources of the motherhood penalty as losses caused by employment interruptions and underinvestment in human capital, exchange of part of the wage for more convenient working conditions, reduced work effort, including due to high involvement in unpaid domestic work. The hypothesis of mothers’ lower productivity is not supported.
Controlling for regional variables revealed a relatively higher motherhood penalty in Western Europe and the United States and a relatively lower one in Latin America. This provides some empirical evidence in favour of the study’s hypothesis of a possible relationship between the size of the motherhood penalty and fertility concentration. In countries with a significant heterogeneity in the distribution of women by the number of births (high prevalence of childlessness and/or multiple children), employers may favour smaller gender pay gaps along with a higher motherhood penalty.