Lara Bister writes about her master thesis on maternity leave lengths and maternal health in the long-run in Germany. She conducted this research at the University of Groningen (supervised by Dr. Roberta Rutigliano) and the Max Planck Institute for Demographic Research, Germany (supervised by Dr. Peter Eibich und Dr. Karen van Hedel).
Since the second half of the 20th century, female labour force participation has rapidly increased and this trend is expected to continue. This also implies that a larger proportion of women, after their transition into motherhood, have to reconcile their professional careers with their family life. Maternity leave or, more broadly, parental leave is an institutionalised strategy aimed at protecting maternal, prenatal, and neonatal health (especially during the statutory maternity leave) and later on, balancing family and work. However, there is marked variation in the length of maternity leave between and within countries, which might lead to large variations in maternal health outcomes (sickness occurrence). There is a growing body of literature stating a relieving effect of maternity leave on maternal health, yet, findings on the role of the length of maternity leave are still inconclusive. Some studies state a positive association between maternity leave length and maternal health while others suggest selection effects driving this relationship. This made me wonder: do longer maternity leaves mean better health for mothers?
Research exploring the relationship between maternity leave length and maternal health is of high relevance for society, policy makers, and academic research. First of all, the demographic change and the generational principle of many pension insurance models in several countries require a high level of labour market participation. As such, it is important to sustain a generally healthy workforce, and a focus on maternal health is an essential part of this. Second, sound knowledge of the “optimal” length of maternity leave for maternal and child health can help to ensure that policies are designed effectively. Third, not only the length of maternity leave itself but also the individual motivations of a leave take-up are relevant information when promoting effective reforms. Motives can reach from individual economic drivers (i.e., leave benefits or childcare availability), career-focused purposes (e.g., motherhood penalty or wage gap, but also opportunity costs), or individual health protection. Only if those individual motives are understood, maternity leave policies can be adjusted in accordance with mothers’ and families’ needs.
In my master thesis research, I investigated the association between the length of maternity leave and maternal health in the long run. I focused on selection mechanisms, e.g., mothers’ income, to address the different benefits associated with different lengths of maternity leave. I chose this focus to gain a better understanding of the take-up process of mothers’ parental leave and its relevance for long-run health of mothers. To implement my research, I focused on the case of Germany, which provides a relatively generous maternity leave policy allowing leave lengths from two months to three years postpartum and comparably high maternity leave benefits. I used administrative time series data from the German Statutory Pension Fund (DRV) and analysed the relationship between maternity leave length and post-leave sickness occurrence using an event history approach (discrete-time logit model). Although the data did not allow me to differentiate between different kinds of sickness, it provided a powerful measurement for objective health by indicating the occurrence of severe sickness. My sample consisted of N=4,237 first-time working mothers in Germany born between 1960 and 1979, who had given birth between the ages of 20 and 39 and who were observed from the age of 20 until either a post-maternity-leave sickness occurrence or when they reached the age of 50.
The results of my master thesis generally suggested that longer maternity leaves are associated with higher risks of sickness occurrence. However, this association seemed to be driven by selection. For example, women who experienced serious sickness before their first pregnancy were more likely to take up longer maternity leaves, women with above-average (pre-childbirth) income rather took shorter leaves, and women with below-average income likely took longer maternity leaves. In an additional analysis, I looked at different characteristics of women and how they were associated with both maternity leave take-up but also post-leave sickness risks. Indeed, I found different patterns in the relationship between different maternity leave lengths and sickness risks. On average, they were quite similar, yet, for women who experienced sickness before their pregnancy, the risk of sickness occurrence significantly increased with the length of maternity leave. For women with a high income, the risks of sickness occurrence increased with the length of maternity leave.
In line with previous research, I explained those results with negative health selection, meaning that women in generally worse health statuses take longer maternity leaves to recover from childbirth and the transition into motherhood. Yet, they are more likely to experience post-maternity leave sickness. My findings for high-income-women, on the other hand, were contrary to my expectations of shorter leaves being associated with higher sickness risks due to the pressure of returning to work in high-income positions. However, the results might be driven by unobserved factors. More research is needed to investigate this relationship, such as what role selection into different maternity leave lengths plays for maternal health in the long run.
Finally, I could not confirm that longer maternity leaves mean better health for mothers in my master thesis research. Although I found evidence that the association between maternity leave length and maternal health might be selective, the big picture of the significant circumstances was not clear. Also, more attention should be paid to maternity leave policy reforms in Germany, which significantly affected the available leave benefits and period of paid leave. For that reason, I have worked on extending the analysis performed in my master thesis during the last months together with my co-authors and former supervisors. We focused more on both the selection process into maternity leave lengths and on the determining characteristics for maternal health in the long run. We just submitted the first manuscript to a peer-reviewed scientific journal and hope we can share the results with you soon. Stay tuned!
Lara Bister’s master thesis won the NIDI-NVD Master Thesis Award in 2020. After finishing her master thesis, Lara Bister participated in the European Doctoral School of Demography (2019-20) and started as a PhD candidate at the University of Groningen (supervised by Prof. Dr. Fanny Janssen and Dr. Tobias Vogt). Since 2020 she is affiliated PhD student at the International Max Planck Research School for Population, Health and Data Science.