Please use this identifier to cite or link to this item: http://hdl.handle.net/10419/213599
Authors: 
Frodermann, Corinna
Wrohlich, Katharina
Zucco, Aline
Year of Publication: 
2020
Series/Report no.: 
DIW Discussion Papers 1847
Abstract: 
Paid parental leave schemes have been shown to increase women's employment rates but decrease their wages in case of extended leave durations. In view of these potential trade-offs, many countries are discussing the optimal design of parental leave policies. We analyze the impact of a major parental leave reform on mothers' long-term earnings. The 2007 German parental leave reform replaced a means-tested benefit with a more generous earnings-related benefit that is granted for a shorter period of time. Additionally, a "daddy quota" of two months was introduced. To identify the causal effect of this policy on long-run earnings of mothers, we use a difference-in-difference approach that compares labor market outcomes of mothers who gave birth just before and right after the reform and nets out seasonal effects by including the year before. Using administrative social security data, we confirm previous findings and show that the average duration of employment interruptions increased for high-income mothers. Nevertheless, we find a positive long-run effect on earnings for mothers in this group. This effect cannot be explained by changes in working hours, observed characteristics, changes in employer stability or fertility patterns. Descriptive evidence suggests that the stronger involvement of fathers, incentivized by the "daddy months", could have facilitated mothers' re-entry into the labor market and thereby increased earnings. For mothers with low prior-to-birth earnings, however, we do not find any beneficial labor market effects of this parental leave reform.
Subjects: 
parental leave
wages
labor supply
JEL: 
H31
J13
J22
J24
J31
Document Type: 
Working Paper

Files in This Item:
File
Size
685.99 kB





Items in EconStor are protected by copyright, with all rights reserved, unless otherwise indicated.