Research Brief

In the 1990s, the states and then the federal government enacted maternity-leave legislation. This legislation guaranteed to mothers the right, after a leave of limited duration, to return to their pre-leave employers at the same or equivalent positions. A recent presidential proposal even encourages states to use unemployment benefits to provide paid time off for new parents. Such initiatives were made in response to trends that showed the number of working mothers rising sharply in the past two decades.

By protecting the right of new mothers to return to their previous jobs, maternity-leave statutes seek to help women benefit more from on-the-job training and to reduce the wage gap between mothers and women who have never had children. If maternity-leave initiatives are to fulfill their promise, it is important to understand the extent to which new mothers returned to their old jobs before the passage of such legislation.

Jacob Alex Klerman and Arleen Leibowitz use data from the National Longitudinal Survey of Youth (NLSY) and the Current Population Survey (CPS) to analyze job continuity among new mothers prior to the federal Family and Medical Leave Act of 1993 (FMLA). They find that most women working full-time before pregnancy returned to work for the same employer after giving birth. They argue it is unlikely that maternity-leave legislation will have a major effect on job continuity for working mothers, since the attractiveness of a given job may change after a woman gives birth (e.g., such women may want jobs with flexible hours or on-site day care). Such legislation, however, may have other effects on women's lifetime labor market choices.

Differences Between Mothers and Non-Mothers

Klerman and Leibowitz combine NLSY and CPS data to compare labor market behavior of mothers and demographically similar non-mothers over an 18-month period. They examine changes in employment by employer and hours worked for each group of women.

The FMLA applies to women working full-time before giving birth. Klerman and Leibowitz find that, prior to its passage, 61 percent of mothers who were working full-time twelve months before giving birth were working for the same employer six months afterward—45.4 percent full-time and 15.7 percent part-time. Among all mothers who went back to work full-time, 89 percent were working at their pre-pregnancy jobs. For these women, the FMLA guarantee of return to the pre-leave job codifies existing practice. Even without FMLA protection, relatively few of the new mothers working full-time both before and after birth did not return to their pre-pregnancy jobs.

Among demographically similar non-mothers working full-time, 82 percent were working for the same employer eighteen months later, 54.7 percent full-time and 26.8 percent part-time. Subtracting this 82 percent job continuity rate from the 61 percent for mothers shows a 21 percent deficit in job continuity or excess turnover for the new mothers. Overall, about 40 percent of new mothers working full-time before pregnancy did not return to their previous employer. The 21 percent excess turnover rate for new mothers indicates that about half of this turnover was caused by the birth of the child. The remaining job turnover for new mothers would have occurred regardless of childbirth.

Klerman and Leibowitz compute a similar excess turnover rate for those working full-time both before and after the birth of the child. Among demographically similar non-mothers, 97 percent working full-time at both the beginning and end of an 18-month period were working at the same job. Since the continuity rate for new mothers working full-time both before and after giving birth is 89 percent, their excess turnover rate is 8 percent.

Analyzing job switching among mothers and non-mothers, Klerman and Leibowitz find still smaller effects of childbirth on job continuity. Among women working full-time before their pregnancy, 11.1 percent were working at a different job six months after giving birth. Among non-mothers working full-time, 8.1 percent were working at a different job 18 months later. Thus, the excess rate of job switching among mothers who had been working full-time is only 3.0 percent. Even if all this excess were caused by lack of adequate leave before passage of the FMLA, Klerman and Leibowitz point out that the scope for a maternity leave statute to affect job continuity is small.

Moreover, much of the excess turnover for working mothers may be attributable to changes in job preferences. After the birth of a child, for example, jobs that offer more-flexible hours or on-site day care become more attractive. Thus, even if the pre-birth job were still available, a new mother might still change jobs after giving birth—leading to job discontinuity that family leave statutes cannot affect.

Differences Among Mothers

Klerman and Leibowitz also analyze differences in working mothers by parity (i.e., number of children previously born) and educational level. Most (52 percent) mothers were working full-time before the birth of their first child, while fewer than two in five (38 percent) were working full-time before the birth of their second child, and only one in three (33 percent) was working full-time before the birth of her third child. Among those working full-time before the birth of their first child, only about half were working at the same job after giving birth. Among those working full-time before the birth of their second or third child, about two in three were working at the same job after giving birth.

These numbers suggest that after their first birth, mothers divide themselves into workers and non-workers. If a woman is at work before later pregnancies, she is likely to be working after subsequent births.

Comparing results across educational groups, Klerman and Leibowitz find that high school dropouts are less likely than others are to work before and after giving birth. College graduates were as likely to work full-time as high school graduates were before pregnancy. Among those who worked full-time before pregnancy, college graduates were somewhat more likely than high school graduates were to return to work part-time rather than not at all after the birth of a child.

Effects of Maternity-Leave Statutes

Klerman and Leibowitz's analysis shows high levels of job continuity even before passage of the FMLA. Much of the excess turnover that exists for new mothers in comparison with demographically similar non-mothers likely stems from differences in labor market preferences held by new mothers, e.g., preferences for more-flexible jobs or jobs with on-site child care.

It is unlikely that even longer guaranteed leaves would have much effect on the job market behavior of new mothers. Klerman and Leibowitz find mothers' employment rates increase only slightly from six to eighteen months after giving birth. Such circumstances make it seem unlikely that many women who are not working 18 months after the birth of a child would be induced by a maternity-leave statute to return to their pre-birth employers.

It is possible that maternity-leave statutes will affect labor market behavior before birth, as the statutory guarantees of job continuity lead more women to work before pregnancy. Knowing that they would not have to quit their jobs in order to spend a few months with a newborn child, women may choose to work more. Nearly one in three women in the analysis was not at work, and only a minority were working full-time. If maternity-leave statutes lead women to work more and help them continue to collect from the investments they make in their employers, family-leave statutes may help mothers earn higher wages. Such potential effects of maternity-leave legislation merit further analysis.

This report is part of the RAND Corporation Research brief series. RAND research briefs present policy-oriented summaries of individual published, peer-reviewed documents or of a body of published work.

The RAND Corporation is a nonprofit institution that helps improve policy and decisionmaking through research and analysis. RAND's publications do not necessarily reflect the opinions of its research clients and sponsors.