Fertility rates are declining globally – South Korea’s rate fell to 0.72, the lowest in the world (Kim and Yum 2025) – while large gender inequalities persist in many countries (Berniell et al. 2024). Governments have expanded policies that support working families to address both challenges, but these policies create operational dilemmas for firms: should they temporarily replace the absent employee with a new hire, redistribute their workload among existing workers, or adjust production and reduce the size of the firm? The answer depends on recruitment and training costs firms face when searching for suitable replacements in a labour market with frictions, where temporary worker absences are costly.
The length of the leave is a crucial factor in this debate. Although longer leave can be beneficial for women (Olivetti and Petrongolo 2017), and generous leave can lead to higher fertility (Kim and Yum 2025), there is concern that these policies adversely affect firms. Indeed, there are two papers that find potential negative impacts on firms. Ginja et al. (2022) examine a Swedish reform that retroactively extended parental leave by three months and find that the longer leaves imposed higher wage costs on firms and some indications that sales and value added per worker were negatively impacted. Gallen (2019) studies a Danish reform that unexpectedly extended leave by 22 weeks and reports negative effects on firm survival and coworker retention, with some coworkers experiencing lower earnings when changing jobs and delaying their own fertility decisions.
One concern with this research is that it examines a sudden change to a system that firms could not anticipate. It is thus not clear whether the negative effects on firms manifest because they were surprised by workers suddenly taking longer leave, and if they could handle longer leave if given enough time to plan.
In our recent paper (Brenøe et al. 2025), we use data from Austria to examine how firms respond to births and variations in the length of leave when they have enough time to plan for a change in the system. This provides an ideal setting for analysis for two reasons. First, Austrian mothers take exceptionally long leave compared to other countries – typically between one and two years. If extended leaves impose significant costs on firms, the Austrian context should reveal these effects clearly. Second, Austria implemented major parental leave reforms that substantially altered leave durations, providing a quasi-experimental setting.
Our first key finding concerns how firms adjust their workforce in anticipation of a leave. Figure 1 shows that firms begin increasing new hires approximately two quarters before the birth event. The number of new hires peaks in the quarter immediately before birth, when many expectant mothers begin their maternity protection period. Cumulatively, firms hire approximately 0.89 additional workers for each woman going on leave, achieving nearly complete replacement.
Figure 1 How a birth impacts firms’ hiring
Notes: Figure 1 shows how many new workers are hired by firms if an employee has a child. We estimate this effect with a methodology that compares firms affected by a birth to comparable firms with no birth. We plot the effect separately for hiring of female and male workers.
Importantly, these replacement hires are almost entirely female – only 9% of excess hires during the anticipation period and 17% during the leave period are male. This pattern suggests significant gender segregation within firms in the allocation of workers to specific roles.
The gendered nature of firm responses extends beyond hiring. Figure 2 shows the evolution of the wage bill by gender. The female wage bill in treated firms initially increases during the anticipation period as replacement workers are hired, then drops sharply when the mother goes on leave. In contrast, the male wage bill shows a small but persistent increase. In the medium run, five years after the birth event, the total wage bill returns to levels comparable to control firms, but its composition has shifted toward male workers. This reallocation appears permanent, suggesting that births trigger lasting changes in firms’ gender composition.
Figure 2 How a birth impacts firms’ wage bills
Our second major finding addresses the central policy question: does the duration of leave matter for firms? Figure 3 provides clear evidence. Panel 3(a) confirms that actual leave-taking varied strongly across regimes in the short run – with approximately 40, 60, and 80 days of leave per quarter in the 1-year, 1.5-year, and 2-year regimes, respectively. Yet panels 3(b) and 3(c) show that this variation in leave duration had virtually no impact on firms’ hiring patterns or wage bills.
Figure 3 Are firms differently affected if women are on long leave?
Notes: Figure 3 shows how an employee having a child affects firms depending on how long mothers can remain on parental leave. Panel (a) shows that the Austrian parental leave reforms shifted large shares of mothers from taking 1 year to taking 1.5–2 years of leave. Panels (b) and (c) show how firms’ hiring and wage bills were not significantly different in those three policy situations.
Whether mothers could take one year or two years of leave, firms hired the same number of replacement workers during the anticipation period. The wage bill increased by similar amounts across all regimes during anticipation and returned to baseline levels in the medium run, regardless of how long the actual leave lasted.
We also find no effects on average daily wages or, importantly, on firm survival rates. Even in the most generous regime, where mothers routinely took two-year leaves, firm closure rates remained unaffected through five years after the birth event.
There is one caveat in the Austrian setting, and that is the lack of direct data on sales or value added. However, our wage and closure results broadly correspond to findings from Denmark by Brenøe et al. (2024), who do not find that leaves lead to big impacts on sales in the periods where the leave policy was stable. Ultimately we have to leave these more granular dimensions of impact on firms open for future research.
Implications for policy
Our findings offer several insights for policymakers considering parental leave reforms. First, the contrast between our results and those from Denmark and Sweden highlights the importance of implementation. When the Danish and Swedish reforms were applied retroactively, catching firms unprepared, they generated disruption and costs. In Austria, the changes did not come as a shock, and even though the reforms substantially changed leave durations, we find no negative effects on firms. This suggests that the predictability of leave policies may matter more than their generosity in terms of the maximum available duration.
Second, the finding that leave duration has minimal impact on firms challenges common assumptions in policy debates. We show that variations in maximum leave length between one and two years – a substantial difference by international standards – do not translate into differential costs for firms. Once leaves exceed a certain threshold, further extensions appear to impose little additional burden on employers.
Third, the gendered nature of firm responses reveals both how firms manage leave absences and why gender inequalities may persist despite generous policies. The near-exclusive hiring of female replacements suggests that many workplaces remain highly gender-segregated. Women appear concentrated in easily replaceable roles, which facilitates smooth transitions during leaves but may simultaneously limit career advancement opportunities.
Finally, our evidence that even very long leaves do not threaten firm survival removes a common objection to generous family policies. Concerns that extended parental leave will devastate businesses appear unfounded, at least when policies are predictable and well-established.
As countries seek to address the demographic challenges of low fertility and persistent gender inequality, the Austrian experience suggests that generous parental leave need not come at the expense of firms. However, achieving true gender equality in the labour market requires addressing the underlying patterns of occupational segregation. The ease with which firms replace women on leave may itself be a symptom of the limited roles available to them.
References
Berniell, I, R Fernandez, and S Krutikova (2024), “The state of gender inequality in Latin America”, VoxEU.org, 6 August.
Brenøe, A A, S Canaan, N A Harmon, and H N Royer (2024), “Is parental leave costly for firms and coworkers?”, The Journal of Labor Economics 42(4): 1135–74.
Brenøe, A A, U Krenk, A Steinhauer, and J Zweimüller (2025), “How do firms respond to parental leave absences?”, CEPR Discussion Paper 20140.
Gallen, Y (2019), “The effect of parental leave extensions on firms and coworkers”, working paper.
Ginja, R, A Karimi, and P Xiao (2022), “Employer responses to family leave programs”, American Economic Journal: Applied Economics 15(1).
Kim, D, and Y Yum (2025), “The effects of parental leave policy reforms on fertility and gender gaps”, VoxEU.org, 17 March.
Olivetti, C and B Petrongolo (2017), “The economic consequences of family policies”, VoxEU.org, 3 June.
