Pre- and postpartum employment patterns: comparing leave policy reform in Canada and Switzerland

ABSTRACT

In recent decades, many countries modified their maternity and parental leave programmes, changing elements such as length, wage replacement levels, and eligibility criteria. We employ sequence analysis of women and men’s employment trajectories in the two years before and after a birth to explore changes occurring alongside reforms that advanced different policies: a short, compulsory, and well-compensated maternity leave in Switzerland in 2005, and a long, voluntary, but less well-compensated parental leave in Canada in 2001. Our results show that employment patterns changed little after the reform in Switzerland. Most Swiss women remained in or switched to part-time employment in the period preceding childbirth. After the reform in Canada, mothers in the province of British Columbia—the context of our study—spent more time out of employment after the birth of a child. However, they were also more likely to return to work full time. In both contexts, the employment trajectories of men did not change. Together the results highlight that parents are not passive recipients of policy change; rather, reform may reinforce old patterns or generate change depending on the extent of change and the context where it takes place.

KEYWORDS:

• Parental leave policy
• maternity leave policy
• Canada
• British Columbia
• Switzerland
• sequence analysis

Data availability statement

Two databases are used in this paper, the Swiss Household Panel and the Canadian Paths on Life’s Way. Swiss data are available under request via the Swiss Centre of Expertise in the Social Sciences – FORS (http://forscenter.ch/en/our-surveys/swiss-household-panel/). Canadian data can available under request via the Paths on Life’s Way Project (http://blogs.ubc.ca/paths/).

Disclosure statement

No potential conflict of interest was reported by the author(s).

Notes

1 We use the generic term “leaves” to refer to any type of statutorily protected time off from employment around the birth (or sometimes the adoption) of a child. This encompasses leave modalities directed specifically at mothers (maternity leave), fathers (paternity leave), or either parent (parental leave). Where appropriate, we differentiate these more specific types of leaves. This paper does not address other types of leaves (e.g., vacations, sick-leave, caregiving leave for ill family members).

2 The most critical components of leave reform reflect national-level policy initiatives in Canada – hence our focus on “Canadian” leave policies. However, it is preferable to assess the impact of implementation at the sub-national level because job protection provisions around leaves are also set via provincial Employment Standards legislation and vary somewhat in their details. Also, the province of Québec opted out of the federal system to create a distinct policy regime. Focusing on employment patterns in British Columbia thus allows for internal consistency vis-a-vis the policy context. Switzerland, too, is a federalist system. However, in terms of maternity leave, policies across cantons are similar. The small differences that do exist do not have a significant impact on our analyses.

3 The International Social Survey Program data we use to assess attitudes towards maternal employment can be broken down by province, revealing that attitudes are somewhat more conservative in British Columbia than in Canada as a whole. However, the sample size for BC is very small (N=75) so this should be interpreted with caution.

4 Although maternity leaves were part of some collective and individual contracts, 59% of Swiss firms did not guarantee any form of maternity leave (Aeppli, 2012).

5 Of note, this led to shorter leaves for some women employed in firms that previously had more generous leave lengths (i.e., 4.6% of Swiss firms [Aeppli, 2012]).

6 A later reform implemented in 2017 (beyond the time frame of our study) added the option of a longer leave, up to 61 weeks for any one parent, at the same overall benefit level. Beginning in 2019, Canadian parents were also able to access an additional 5 or 8 weeks at the same overall benefit level so long as couples split the time off (i.e., “use it or lose it” benefits for a second parent).

7 Time taken by one parent is deducted from the entitlement of the other.

8 In 2017, the government introduced extended parental benefits to a maximum of 61 weeks in exchange for a lower benefit rate. The period covered by our data does not capture this recent change.

9 Prior to the reform in 2018, provinces provided either 37 or 35 weeks of job protected parental leave and 16–18 weeks of job protected maternity leave. Job protections for parental leave have since been lengthened in most provinces to harmonize with the new federal provisions.

10 The Survey of Labour and Income Dynamics covers the period before and after reforms and allows for the exclusion of data from Québec, but does not provide precise enough birth dates for children. As noted earlier, provincial variation in job protection surrounding birth also make national-level data less useful in Canada.

11 The different time periods do not substantively affect our analysis given the cohort structure of the Canadian sample where only a small minority of births took place in the early 1990s.

12 Countries vary in how they interpret part- versus full-time employment. In Canada and BC, part-time employment is classified as 29 h or less per week. Swiss data classify part-time employment as 37 h or less per week. We use country-definitions of part-time status rather than harmonized definitions as local norms are particularly relevant to parental decision-making around employment.

13 Some of the clusters included in this and in the following analysis are small (i.e., less than 40 sequences). When clusters have a small number of sequences, a change in the trajectories of only a few people could alter the overall structure of the cluster.

14 In such cases, mothers might reduce their part-time hours or take vacation time after birth, both of which would provide some time for care but not lead to a classification of non-employment. Non-biological mothers (adoptive or same-sex) may also maintain continuous employment after a birth as they do not require the same recovery period and cannot breastfeed.

15 The definitions of part-time employment differ in Switzerland and British Columbia, with the threshold defining the number of hours higher in the former. We are confident that this does not affect our findings (e.g., Swiss and BC women working similar hours, but are classified as part-time in Switzerland and full-time in BC) as, in practice, Swiss part-time contracts are typically set at 80% or less and thus often would fall below the BC part-time threshold.

16 We also perform a sensitivity analysis considering the 9th month after the reform as a cutline. This test makes it possible to understand if policy change had a lagged effect on employment decisions, as those who are already pregnant or planning pregnancies before the reform may make decisions with pre-reform conditions in mind. Divergences between these two approaches are minimal and consistent with our conclusions.

17 Of note, we do not see dramatic shifts in childcare policy at the time of the leave reforms that would lead us to believe childcare provisions explain the patterns of results. Additionally, both countries have a relatively low share of children in publicly funded childcare. In 2008, the percentage of children age 0–2 in formal childcare and/or pre-school services was 24% in Canada and 16% in Switzerland (OECD, 2018).

Additional information

Notes on contributors

Matteo Antonini

Matteo Antonini after having completed his PhD at the University of Lausanne (Switzerland) working at the Swiss National Centre of Competence in Research LIVES, he is currently a research assistant at the La Source School of Nursing (Lausanne, University of Applied Sciences and Arts in Western Switzerland). He is currently participating to several research projects the most relevant being the national founded Exp-Care study on nurses-patients relations. Eclectic researcher, he is interested in life courses, labor market dynamics, the role of beliefs in human behaviors, and sequence analysis applied to social sciences.

Ashley Pullman

Ashley Pullman is currently a Postdoctoral Fellow at the University of Ottawa's Education Policy Research Initiative. Her main area of research is education and work across the life course. With an interest in life course studies, her current research utilizes the Canadian and German longitudinal components of The Program for the International Assessment of Adult Competencies (PIAAC).

Sylvia Fuller

Sylvia Fuller is Associate Professor of Sociology at the University of British Columbia. Her research centers on understanding how entrenched patterns of inequality in the labour market develop and erode, and in the implications of changing employment relations for workers' prospects for security and mobility. Recent publications explore how organizational context shapes motherhood wage penalties and fatherhood wage premiums, temporary workers' employment trajectories, divergence in the career pathways of new immigrants, and the medicalization of welfare among lone mothers.

Lesley Andres

Lesley Andres is a Professor in the Department of Educational Studies at the University of British Columbia. Her research focuses on the sociology of higher education, issues of inequality and access, transitions across the life course, and mixed method survey research design. She is the principal investigator of the Paths on Life's Way Project, a unique Canadian longitudinal study that has combined extensive qualitative and quantitative data over a 28 year time frame to examine the lives, actions, experiences, and perspectives of individuals within a life course framework. This research focuses on educational, occupational, and other life course outcomes in relation to various forms of inequality. Her most recent book is Designing and Doing Survey Research (2012).