Economic and Social Reports
The likelihood and timing of mothers returning to work after parental leave

Abstract

This study aims to examine the likelihood and timing of mothers returning to work after parental leave (or a break from work), using data from the Employment Insurance Coverage Survey. Focusing on mothers with an infant younger than 12 months who worked as a paid employee before childbirth or adoption and took a break following childbirth or adoption, this study analyzed two binary indicators of returning to work after parental leave: whether a mother returned to work or planned to return to work (1) within any known length of leave or (2) within 12 months of leave. Mothers’ individual and job characteristics and return-to-work rates were compared between the 2009 and 2019 cohorts of mothers. This study found differential changes between the 2009 and 2019 cohorts in return-to-work likelihood and timing by certain individual and job characteristics. Further analysis delineated the characteristics that were associated with the likelihood and timing of mothers’ return to work among two recent cohorts of mothers (2018 and 2019) when all characteristics were considered simultaneously in nested logistic regression models.

Authors

Youjin Choi is with Social Analysis and Modelling Division, Statistics Canada

Acknowledgements

The author would like to thank anonymous reviewers from Employment and Social Development Canada and Kate Rybczynski for their advice and comments on an earlier version of this paper.

Introduction

Before the COVID-19 pandemic, women’s employment rate had steadily risen in Canada. Although its growth slowed since the 1980s, the employment rate of women aged 25 to 45 rose steadily and peaked at 79% in 2019, 9 percentage points higher than three decades earlier (1989) and 3 percentage points higher than 10 years earlier (2009) (Statistics Canada, n.d. a). During the same period, the educational attainment of women in Canada showed remarkable improvement. The percentage of women aged 25 to 64 who completed a postsecondary education increased by about 10 percentage points every 10 years in the last two decades (from 52% in 2000 to 62% in 2009 to 72% in 2019) (Statistics Canada, n.d. b). These statistics suggest that continuing employment after becoming a parent became widely common, and women now invest in their formal education and career more than ever.

With these social and economic changes, policy efforts have been made to encourage and help new mothers continue their careers after childbirth or adoption. One such policy is paid maternity and parental leave, followed by child care subsidies and tax credits. Maternity and parental leave helps parents take time off work to spend time with their child and guarantees their return to the same or equivalent job once the leave ends. This job protection is supplemented by financial support for new parents through a provincial parental insurance program in Quebec and the Employment Insurance (EI) program’s parental benefit in the rest of Canada.

Whether and when a mother returns to work after childbirth or adoption are closely related to available paid parental leave. In many cases mothers return to work once they exhaust their parental leave. The duration of paid parental leave in Canada has increased a few times over the past decades. The EI program expanded the maximum length of leave a mother can use from 6 to 12 months in 2001 and provided an option for 18-month parental leave as of December 2017. As parental leave was extended, more mothers could take longer parental leave after childbirth or adoption before returning to work. With the total amount of parental benefits remaining the same, mothers who choose extended parental benefits receive lower weekly benefits (33% of their average weekly earnings for up to 61 weeks) than parents who choose standard parental benefits (55% of their average weekly earnings for up to 35 weeks). Extended parental leave benefits were not available in Quebec under the Quebec Parental Insurance Plan (QPIP). During this time, compared with the EI parental benefits, the QPIP provided a lower threshold for eligibility (minimum earnings of $2,000), a greater payout (up to 75% of annual earnings, depending on the selected plan) and designated benefits for fathers.

Historically, an extension of parental leave led Canadian mothers to spend a longer time at home and increased job continuity with their pre-birth employer (Baker and Milligan, 2008a). Canadian studies showed that there were changes in the timing of mothers returning to work, but the percentage of mothers who withdrew from the labour market within two years after childbirth did not change between the early 1990s and the early 2000s (Marshall, 1999; Baker and Milligan, 2008a; Zhang, 2007). The current study revisits the employment of mothers after parental leave in Canada using more recent data from before the COVID-19 pandemic.

Previous research

Using panel data from the Survey of Labour and Income Dynamics, Marshall (1999) analyzed the post-childbirth employment of mothers who gave birth in 1993 and 1994, when six-month parental leave was in place in Canada, and found that the majority of mothers returned to work within six months. An extension to one-year leave in 2001 increased the percentage of mothers with infants younger than 1 year who were employed and on leave (Baker and Milligan, 2008a). When mothers who gave birth between 1998 and 2000 and between 2001 and 2003 were compared, the extension increased the duration of mothers being away from work in the first year after childbirth by 2.3 months over 8.2 months (Baker and Milligan, 2008b). Also, the percentage of mothers staying away from work 9 months or longer increased by 37 percentage points from 47%, and the percentage of those taking 12 months of leave or longer increased by 29 percentage points from 41% (Baker and Milligan, 2008b). However, the employment rate of mothers two years after childbirth remained around 90% in the early 1990s and early 2000s (Marshall, 1999; Zhang, 2007). The current study examines whether mothers’ use of leave and return-to-work behaviours have changed in recent years.

Canadian and international studies have found that the availability of paid parental leave with job protection increases job continuity for mothers who want to continue their careers after childbirth. International studies found evidence that access to and use of parental leave increased the likelihood of mothers returning to work (Waldfogel, Higuchi and Abe, 1999; Pylkkänen and Smith, 2003; Baxter, 2008). In addition, the duration of parental leave influenced the timing of a mother’s return to work (Burgess et al., 2008; Aisenbrey, Evertsson and Grunow, 2009; Lalive and Zweimüller, 2009; Del Rey, Kyriacou and Silva, 2021) and had a non-linear relationship with mothers’ labour market participation (Del Rey, Kyriacou and Silva, 2021). Mothers’ return to work was concentrated around the time when their leave ended, and most mothers used their full leave entitlement (Rønsen and Sundström, 2002). Complementing this literature, the current study also examines whether receiving parental benefits was a determinant of the likelihood and timing of mothers returning to work.

Characteristics of the jobs held before childbirth may also be related to mothers’ return to work. Top-up payments from employers may reflect that an employer is family-friendly or provides high-quality jobs that offer generous benefits to its employees. Unionized jobs may also offer more generous benefits and protection against potential unfair treatment toward mothers after they return to work than non-unionized jobs. Baxter (2008) examined whether the use of parental leave and pre-birth job characteristics, with an interest in contract types, were associated with the timing of the return to work after childbirth among Australian mothers. At the time of the study, one-year unpaid maternity leave was available to permanent employees with at least one year of continuous employment with an employer, and access to paid maternity leave was not universal in Australia (Baxter, 2008). Permanent employees had a higher rate of returning within 18 months after childbirth than contract or casual employees, but the difference between contract types was related to a higher likelihood of using leave for permanent employees (Baxter, 2008).

The presence of children affects the economic behaviours of families, for example, by increasing the consumption and financial needs of families in the short term and the long term (e.g., for children’s postsecondary education) (Browning 1992). Primary caregivers, who are often mothers, adjust their labour supply after considering the increased value of their time at home and the additional costs of working out of the home (e.g., child care service fees). A well-known factor in mothers’ return-to-work decision is their human capital, because it largely determines the opportunity cost of not working or, equivalently, the marginal benefits of returning to work. The human capital theory suggests that mothers with a higher level of human capital are more likely to return to work because their earnings are higher on average, and the benefits of working outweigh the short-term costs of child care in the long run.

Depending on available information, human capital is proxied by several variables in labour economics. Educational attainment is commonly used to measure general human capital that can transfer across jobs in the labour market. In the literature, a higher level of education was associated with returning to work sooner (Pylkkänen and Smith, 2003; Baxter, 2008). Age is often used to measure general work experience when information on the number of years working in the labour market is unavailable in the data. Some job characteristics also measure human capital to some degree. For example, job tenure (the length of continuous employment with an employer) reflects the level of job-specific human capital that a worker has accumulated while continuously working at a job. Hourly wages directly measure how much money mothers can make when they return to work, reflecting the value of human capital in the labour market.

Family and demographic characteristics may also matter to mothers’ return to work. For example, family income may be an important determinant. Depending on available child care options, low-income mothers may be unable to afford to stay away from work for long or may stay away for longer if working and paying child care costs are not financially viable (Findlay, Wei and Arim, 2021). Generally, lone-parent mothers and low-wage mothers with an unemployed or low-wage spouse or partner may be in a situation similar to that of low-income mothers. The immigrant status of mothers and their cultural background may also be related to their post-childbirth employment. Mothers who grew up in a culture where mothers’ labour market participation is not a social norm may be less likely to return to work (Kingsbury et al., 2021).

Earlier studies found that the availability of low-cost child care had a positive impact on mothers’ employment (Baker, Gruber and Milligan, 2008; Lefebvre and Merrigan, 2008; Geyer, Haan and Wrohlich, 2015; Zoch and Hondralis, 2017). Quebec introduced subsidized low-cost daycare services for all 4-year-olds in 1997 and expanded the program to 3-year-olds in 1998, 2-year-olds in 1999 and infants aged 1 year and younger in 2000. This policy change was found to increase the employment of mothers with preschool-aged children (Baker, Gruber and Milligan, 2008; Lefebvre and Merrigan, 2008).

Current study

This study revisits the employment of mothers after parental leave in Canada using more recent data from before the COVID-19 pandemic. It uses cross-sectional data from the Employment Insurance Coverage Survey (EICS), which collects information about mothers’ return to work by asking questions to birth or adoptive mothers with an infant younger than 12 months. Specifically, this study aims to add to a better understanding of the likelihood and timing of mothers’ return to work after parental leave by addressing the following three research questions:

1. Did characteristics and return-to-work rates change between the 2009 and 2019 cohorts of Canadian mothers?
2. Were there differential changes in the return-to-work behaviour of mothers with different characteristics?
3. Which characteristics were associated with the likelihood and timing of returning to work among recent mothers?

Descriptive statistics of the most recent cohort of mothers (2019) were compared with the cohort of mothers from 10 years earlier (2009). Factors associated with the likelihood and timing of mothers’ return to work among the two most recent cohorts (2018 and 2019) were examined. An outcome of three categories that summarize whether a mother returned to work within any length of leave and whether a mother returned within or after 12 or more months of leave was analyzed in a multinomial logistic regression analysis. This study considered a rich set of job characteristics, such as receipt of maternity or parental benefits, receipt of employer top-up payments, job tenure and hourly wages, as well as sociodemographic characteristics. It demonstrated that the characteristics and return-to-work behaviour of the recent cohort of mothers were different than those of the cohort from a decade earlier and that not all mothers experienced the same changes during the decade. The low-cost child care program was available only in Quebec at the time of this study, and Quebec had different parental insurance programs than the rest of Canada. Provincial differences would reflect the difference in child care costs and parental benefits available in Quebec and the rest of Canada.

The remainder of the article is organized as follows. The next section describes the data and methodology, followed by the section presenting descriptive results for observed differences in the characteristics and return-to-work behaviour of mothers between the 2009 and 2019 cohorts. This is then followed by the section that presents the results of multinomial logistic regression analyses to demonstrate factors associated with the likelihood and timing of returning to work among recent new mothers. The final section concludes with a discussion of the significance and implications of the main findings.

Data sources and methodology

Data sources

This study uses data from the EICS from 2009 to 2019. The EICS was first designed to better understand the coverage of the Canadian population by EI benefits, focusing on the unemployed, the underemployed and those not in the labour force. It is administered to a subsample of the Labour Force Survey (LFS). The scope of the EICS was broadened in 2000 to cover access to maternity and parental benefits. With this change, the target population of the survey was expanded to mothers with an infant (by childbirth or adoption) younger than 12 months during the LFS reference week.^Note This cross-sectional survey is nationally representative, except for the territories (Statistics Canada, 2020).^Note It collects information on whether mothers returned to work after a break following the birth or adoption of a child and whether they intended to return if they had not yet done so. The survey data also provide information on the individual, family and job characteristics of mothers.



The current sample includes mothers with an infant younger than 12 months. Because of this limited age range, the likelihood and timing of mothers’ return to work measured in this study were based on mothers’ intention of whether and when they would return to work for most mothers, rather than on their actual return-to-work decision. From 2009 to 2019, about 900 to 1,200 mothers with an infant younger than 12 months responded to the EICS each year; in 2019, 1,028 mothers participated in the survey and represented about 388,000 mothers living in the 10 provinces.

As this study aims to understand mothers’ return to work after parental leave, further exclusion criteria were necessary, and the reductions in the sample size for the 2019 data are shown in Table 1. The first subsample includes only those who had worked in the past two years and who took a break from working following childbirth or adoption. Mothers taking a break from work include those taking parental leave and those ending employment shortly before childbirth or adoption.^Note The sample is then further limited to mothers who were paid employees in the past two years and excludes self-employed mothers and those who were unpaid family workers. This exclusion was applied because questions to collect some of the key information, such as hourly wages, were asked only to workers with a paid job and not to the self-employed or family workers.^Note These sample restrictions excluded about 21% of mothers, mostly because they did not work in the past two years at the time of the survey (about 68% of exclusions).^Note The final sample sizes for the 2009 and 2019 cohorts were 903 and 804, respectively.

The collection for the 2019 EICS, which is the last survey year considered in this study, was completed in February 2020, just before the COVID-19 pandemic began in Canada. The mothers’ responses about their intention to return to work were unlikely to be affected by the COVID-19 pandemic because changes to the labour market were not yet salient. Mothers with an infant in the 2019 EICS data gave birth to or adopted an infant between April 2018 and December 2019.

Measures and definitions

Two binary measures of return to work were created based on the time frame of returning to work after childbirth or adoption. The first measure indicates whether a mother has returned to work or planned to return to work after any length of leave.^Note Those who responded that they had no plan to return to work or those who did not know or did not state the planned duration of their leave (i.e., unknown length of leave) were coded as 0. Because mothers who did not answer the planned duration of leave question were viewed as having no specific planned return to work and coded as not returning to work, the return-to-work measure provides a conservative estimate.^Note The second measure, “return to work within 12 months,” was coded as 1 if a mother has returned or planned to return to work within 12 months of leave. This measure was coded as 0 if mothers had no plan to return to work or had a plan to return to work after an unknown length or after 12 months or more of leave. The first measure reflects mothers’ general intention of returning to the labour market after parental leave, possibly with a long career break after childbirth or adoption. The second measure reflects whether a mother intended to return to work shortly after parental leave (less than 12 months). Before 2018, a birth mother could receive maternity and parental benefits for up to 12 months.

This study considers two sets of characteristics that may be associated with mothers’ returning to work. The first set is sociodemographic characteristics, including the mother’s age and immigrant status (whether the mother was born in Canada). The mother’s region of residence was grouped into six regions (Atlantic region, Quebec, Ontario, Manitoba and Saskatchewan, Alberta, and British Columbia).^Note The mother’s education was the highest educational attainment of three groups (high school diploma or less, non-university postsecondary certificate or diploma, or university degree). For family characteristics, whether a mother lives in a couple family (yes or no) was considered. Mothers not living in a couple family were lone parents living with or without other family members (e.g., their parents or relatives) in most cases.^Note If mothers lived with a spouse or partner, the education (the same categories as the mother’s education) and the employment status (employed or not) of the spouse or partner and the number of their own children younger than 13 years (one, two, or three or more) were also considered.

The second set of characteristics are the job characteristics measured during the survey reference week or months. However, the job referred to in the EICS is not necessarily the same as the pre-childbirth job (e.g., some mothers had already returned to work for a different employer during the survey reference week). For mothers of children younger than 12 months who worked within the last two years, took a break from work and have not returned to work, job characteristics refer to the job they held before their leave. For mothers who have returned to work, job characteristics refer to the job they held at the time. If mothers returned to the same employer, their job characteristics would remain the same. Otherwise, the characteristics are of the new job. For the mothers of children younger than 12 months who have returned to work for a new employer after their leave, the characteristics of the job they held before their leave were unavailable in the data; therefore, information about their pre-birth job was unavailable.

Two binary variables were created to measure whether mothers received maternity and parental benefits and whether mothers received additional payments from their employer while taking leave after the birth or adoption of an infant. The set of job characteristics also includes permanent job status (temporary or permanent), full-time status (part-time or full-time job), occupation based on the National Occupational Classification 2016 in five categories (management, business, finance and administration occupations; health occupations; occupations in education, law and social, community and government services; sales and service occupations; or other occupations), union status (whether the mother was a union member or covered by a collective agreement or not)^Note and job tenure (the number of months for which an employee continuously worked at a job). The hourly wages of mothers were expressed in 2020 dollars and measure the market value of human capital.

Methodology

First, descriptive statistics delineate changes in mothers’ characteristics from 2009 to 2019 and observed differences in the rate of mothers returning to work across various subgroups of mothers by their characteristics. The percentage distribution of mothers with infants and the percentage of those who returned or planned to return to work within any known length (or 12 months) of leave are presented by sociodemographic, family and job characteristics and compared for the survey years of 2009 and 2019.

Second, to examine which characteristics were associated with the likelihood and timing of mothers returning to work, two logistic regression models were estimated. The first model examined the likelihood of mothers’ return to work, with the binary outcome of returning to work versus having no planned return. The second model examined the timing of the return to work among mothers who had returned or planned to return to work. This analysis will show whether the association was different for the two return-to-work time frames (less than 12 months versus after 12 months). The sample for this regression analysis pooled the 2018 and 2019 cohorts of mothers to increase the sample size. An indicator of the survey year was added to capture any differences between the two survey years that are not explained by the differences in other characteristics. The main regression model includes the mother’s age and age squared, education, immigrant status, region of residence, couple family status, permanent job status, full-time job status, occupation, union status, job tenure, hourly wages, whether a mother received maternity and parental benefits, and whether a mother received employer top-up payments. For mothers living in a couple family, an additional model was estimated that controlled additionally for the educational attainment and the employment status of the spouse or partner and the number of own children.^Note

For results from logistic regression models, average marginal effects (the differences of predicted probabilities between the reference and comparison groups) were reported. The interpretation of the marginal effects is similar to the interpretation of coefficients from an ordinary least squares regression model. For example, for categorical explanatory variables, positive estimates for the outcome category of returning to work (versus no planned return) are interpreted as a higher likelihood of a mother in a specific group returning to work, compared with the reference group.

As a sensitivity analysis, an outcome for the timing of returning to work with a cut-off of 13 months of leave (returning within 13 months versus returning after 13 months of leave or longer) was estimated to test whether the findings were robust to an alternative return-to-work time frame. The results are presented in the appendix.

For results to be nationally representative (except for the territories), survey weights were applied for all analyses. Bootstrap weighting was used to estimate standard errors, coefficients of variation and 95% confidence intervals with 1,000 bootstrap replicates.

Descriptive statistics

Mothers’ return to work, 2009 to 2019

Chart 1 demonstrates the changes in the percentage of mothers who returned or intended to return to work after childbirth or adoption during the decade from 2009 to 2019. During this period, the total number of mothers with an infant younger than 12 months who were paid employees in the past two years and took a break from work following childbirth or adoption remained relatively stable between 283,600 and 306,700—2.5% of women aged 15 to 64 years.^Note Among these mothers, the percentage who received maternity or parental benefits was similar for the 2009 cohort (84.9%) and the 2019 cohort (86.6%). However, the percentage who returned or intended to return to work sometime after taking a break from work gradually increased over time from 82.3% in 2009 to 88.4% in 2019. According to Table 2, which presents a breakdown by the duration of leave from work following childbirth or adoption for the 2009 and 2019 cohorts, this increase in mothers’ return-to-work rates was because of a decrease in the percentage of mothers who did not intend to return to work from 10.1% in 2009 to 4.7% in 2019. The percentage who responded “Do not know” or did not specify a duration remained stable (from 7.6% in 2009 to 7.0% in 2019).

When a return-to-work time frame of 12 months was considered, the percentage of mothers returning to work within 12 months increased from 64.8% in 2009 to 74.1% in 2015 and decreased to 68.7% in 2018, with a large decrease to 59.6% in 2019. These findings suggest that while an increasing share of mothers intended to return to work after parental leave, there was an increase in the percentage who were away for longer than 12 months. Mothers may be able to use more than 12 months of leave by possibly combining paid parental leave with other types of leave, such as personal vacation and sick leave (paid or unpaid), or by opting for the longer 18-month parental benefits, especially for mothers who gave birth outside Quebec on or after December 3, 2017.^Note

Data table for Chart 1



Data table for chart 1
Table summary
This table displays the results of Data table for chart 1 . The information is grouped by Survey year (appearing as row headers), Return to work within any length of leave, Return to work within 12 months, percent and 95% confidence interval, calculated using from and to units of measure (appearing as column headers).

Survey year	Return to work within any length of leave			Return to work within 12 months
	percent	95% confidence interval		percent	95% confidence interval
		from	to		from	to
2009	82.3	78.2	86.4	64.8	59.9	69.7
2010	83.5	79.7	87.3	64.6	59.9	69.4
2011	83.8	80.0	87.7	68.7	64.3	73.0
2012	83.3	79.8	86.8	66.9	62.4	71.3
2013	86.8	83.8	89.7	72.3	68.2	76.4
2014	85.0	81.6	88.5	72.1	67.7	76.5
2015	87.8	84.8	90.8	74.1	69.9	78.3
2016	85.0	81.7	88.3	67.0	62.7	71.3
2017	86.9	83.1	90.8	67.7	62.9	72.4
2018	89.3	86.1	92.5	68.7	64.2	73.2
2019	88.4	85.3	91.4	59.6	55.2	63.9
Note: The sample is mothers with an infant younger than 1 year who worked as a paid employee in the past two years and took a break from work following childbirth or adoption. Source: Statistics Canada, Employment Insurance Coverage Survey, 2009 to 2019.



Characteristics of mothers and changes in their characteristics from 2009 to 2019

There were some changes in individual and family characteristics between the 2009 and 2019 cohorts of mothers with an infant (Table 3). On average, the 2019 cohort was older than the 2009 cohort and had a smaller share of mothers younger than 30. The 2019 cohort was more educated. The percentage of mothers who did not have a postsecondary education decreased from 23% for the 2009 cohort to 16% for the 2019 cohort, and the percentage who had a university degree increased from 38% to 51%. Also, the share of foreign-born mothers increased from 17% for the 2009 cohort to 27% for the 2019 cohort. For both cohorts, 9 in 10 mothers lived in a couple family and 1 in 10 mothers were lone parents.

There were also some differences in job characteristics between the 2009 and 2019 cohorts of mothers with an infant. The percentage of mothers who received maternity and parental benefits was similar for both cohorts, but the percentage of mothers who received top-up payments from employers increased from 26% in 2009 to 30% in 2019. Both cohorts had similar shares of mothers working full time (8 in 10) and being a union member or covered by a collective agreement (1 in 3). However, there were some differences between the cohorts in terms of job characteristics, such as occupation, job tenure and hourly wages. A noticeable change in the occupational distribution of mothers was an increase in the percentage of mothers working in health occupations (from 13% for the 2009 cohort to 19% for the 2019 cohort), while the percentage who worked in management, business, finance and administration occupations decreased by the same amount. The 2019 cohort was more likely to have a longer job tenure, with an increasing share of mothers working at their job for five or more consecutive years (42%), compared with 35% for the 2009 cohort. The 2019 cohort was more likely to earn higher hourly wages. The percentage of mothers earning less than $15 per hour (in 2020 constant dollars) decreased from 22% to 14%. This decrease was replaced with an increase in the share of mothers earning $20 or more per hour, with evenly increasing shares of mothers earning an hourly wage of $20 to $29, $30 to $39, and $40 or more.

To summarize, the 2019 cohort of mothers was, on average, more educated, had longer job tenure and earned higher hourly wages compared with the 2009 cohort. These changes suggest that the mothers in the 2019 cohort were more invested in their human capital—both general and job-specific human capital—before the birth or adoption of their child. These changes in mothers’ characteristics would lead to an expected higher return-to-work rate for the 2019 cohort of mothers.



Changes in mothers’ return to work from 2009 to 2019 by characteristics

This section examines whether there were differential changes in the return-to-work behaviour of mothers with different characteristics.

Changes in the percentage of mothers returning to work within any length of leave

Tables 4 and 5 present the return-to-work rates for mothers of the 2009 and 2019 cohorts by individual, family and job characteristics. Overall, the return-to-work rate increased from 82% for the 2009 cohort of mothers to 88% for the 2019 cohort. The last column of tables 4 and 5 shows that mothers with certain job characteristics experienced a larger increase in the rates than other mothers.

Across both cohorts, mothers working in a temporary job showed particularly large increases in their return-to-work rates, compared with mothers with a permanent job. For the 2009 cohort, mothers with a temporary job were much less likely to return to work than mothers with a permanent job (68% versus 84%). During the 10-year period, the return-to-work rate increased by 20 percentage points among mothers with a temporary job, while it grew by 4 percentage points among mothers with a permanent job. These differential changes led to a similar percentage of mothers in the 2019 cohort who returned to work regardless of permanent or temporary job status (88%).

Although it was not a statistically significant difference, mothers with the lowest level of hourly wages (less than $15) showed a larger increase in their return-to-work rate (11 percentage points from 64%) than mothers with higher wages over the 2009 to 2019 period. During the same period, the return-to-work rate remained unchanged for mothers earning a middle range of hourly wages (between $15 and $29) and increased by 5 percentage points for high-wage mothers earning $40 or more per hour. For most characteristics, differential changes in the rates across subgroups were not statistically significant.

 

Changes in the percentage of mothers returning to work within 12 months of leave

On average, compared with a decade earlier, more mothers intended to return to work after childbirth and more of them expected to use a leave of 12 months or longer (Table 6). Overall, 60% of mothers in the 2019 cohort returned or planned to return to work within 12 months of leave, 5 percentage points lower than the rate for mothers in the 2009 cohort (65%). However, mothers with certain sociodemographic characteristics experienced a larger decrease in the 12-month return-to-work rate over the 10 years than other mothers. Although it was not statistically significant, the rate decreased by 10 percentage points for university-educated mothers, whereas it increased by 3 percentage points for mothers with no postsecondary education. For the 2009 cohort, mothers with postsecondary education were more likely to return within 12 months than mothers with no postsecondary education, but the 2019 cohort no longer showed statistically significant differences by educational attainment.

Statistically significant differences were found across regions. From 2009 to 2019, the share of mothers who intended to return to work within 12 months decreased among mothers living in British Columbia (from 59% to 37%), Alberta (from 63% to 49%) and Ontario (from 68% to 55%) but increased substantially among mothers in Quebec (from 64% to 79%). These differential changes led to a higher rate of returning to work within 12 months for Quebec mothers than mothers in other regions in the 2019 survey. The difference between Quebec and other regions may be related to the availability of low-cost child care for infants and parental benefits that were limited to 52 weeks for mothers in Quebec.

For family characteristics, the rate of returning to work within 12 months decreased only among couple-family mothers over the decade (from 66% to 60%), and the rate for lone-parent mothers remained around 58%.

Mothers with certain job characteristics experienced a larger decrease in the 12-month rate of returning to work over the 10 years than other mothers (Table 7). Mothers receiving employer top-up payments and mothers with a unionized job showed a larger decrease in the 12-month rate of returning to work than other mothers. The share of mothers who received employer top-up payments and intended to return to work within 12 months decreased from 71% for the 2009 cohort to 56% for the 2019 cohort, whereas the return-to-work rates for the other mothers remained similar. Similarly, the rate decreased for mothers with a unionized job, from 74% to 61%, but remained similar for mothers with a non-unionized job over the period.

While most mothers experienced a decrease in the rate of returning to work within 12 months, some showed an increase depending on their occupation and hourly wages. While the rate decreased for most occupational groups over the 10-year period (e.g., from 76% to 65% for health occupations), it increased for sales and service occupations, from 54% to 63%. In terms of hourly wages, mothers earning less than $15 returned or expected to return to work sooner than their counterparts a decade earlier. The rate of returning to work within 12 months for mothers earning an hourly wage of $30 to $39 decreased from 74% for the 2009 cohort to 54% for the 2019 cohort. However, over the decade, the rate increased for mothers earning less than$15, from 47% to 53%.

 

Regression analysis of the likelihood and timing of mothers returning to work

Many characteristics considered in this study are correlated with each other. For example, age, education, tenure and hourly wages reflect human capital, while unionization, permanent job status and employer top-up payments may reflect job quality. Observed differences by one characteristic may therefore be explained by group differences in another characteristic. This section presents regression analyses that examine the factors associated with the likelihood and timing of mothers with an infant younger than 12 months returning to work after other group differences were considered. A binary outcome of having a planned return (versus no planned return) was analyzed in a logistic regression model with all mothers in the sample, and another binary outcome of returning within 12 months (versus returning after 12 months) was analyzed separately in a logistic regression model with mothers returning to work. Results from the regression analysis are reported in Table 8.

Several characteristics had a statistically significant relationship with the likelihood of mothers returning to work, including mothers’ immigrant status, receipt of maternity or parental benefits, receipt of additional payments from their employer and job tenure. Canadian-born mothers were 7.5 percentage points more likely to return to the labour market than foreign-born mothers. Mothers who received paid maternity or parental benefits were 13.0 percentage points more likely to have a planned return to work than mothers who did not receive parental benefits. Mothers who received additional payments from their employer were 8.5 percentage points more likely to return to work than mothers who did not receive additional payments from their employer. Lastly, job tenure had a statistically significant relationship with whether a mother returned to work or not. Compared with mothers who worked less than a year before giving birth or adopting a child, mothers who worked continuously at their job for 2 to 10 years were 8 to 10 percentage points more likely to return to work after taking time off.

The second regression results showed that among mothers returning to work, the timing of their return had a statistically significant relationship with some characteristics, such as province of residence, hourly wages and survey year. Region of residence was a strong predictor of the timing of mothers’ return to work after childbirth or adoption. Mothers living in Ontario, the Prairie provinces and British Columbia were less likely than mothers living in Quebec to return to work within 12 months and were more likely to return to work after 12 months. British Columbia showed the largest differences when compared with Quebec. Compared with mothers living in Quebec, mothers living in British Columbia were 32.3 percentage points less likely to return to work within 12 months. The likelihood and timing of returning to work were similar in Quebec and in the Atlantic provinces. The findings may be related to the availability of low-cost child care services in Quebec and an earlier finding that mothers in Quebec and the Atlantic provinces were more likely to use child care (Zhang et al., 2021). The likelihood of not returning to work was similar across provinces.

Hourly wages had a statistically significant association with the timing of mothers’ return to work (within 12 months versus after 12 months). The first analysis showed that there were no statistically significant differences in the predicted probabilities of having planned to return to work across all five hourly wage groups. However, the second analysis showed that mothers who earned $15 or more per hour were more likely than mothers who earned less than $15 to return to work within 12 months of leave.^Note For example, the likelihood of returning to work within 12 months was 21.5 percentage points and 16.8 percentage points higher for mothers with hourly wages of $15 to $19 and of $40 or more, respectively, than low-wage mothers.

Regression results showed a difference between the 2018 and 2019 cohorts in the timing of their return to work. The 2018 and 2019 cohorts were equally likely to return to work at some point after childbirth or adoption. However, among mothers returning to work, the 2019 cohort was 8.3 percentage points more likely to return to work after 12 months of leave or longer than the 2018 cohort, after controlling for all other differences in their characteristics. The observed difference between the two cohorts (seen in Table 2) remained after various characteristics were considered. The availability of extended parental benefits, in every province except Quebec, may partially explain the difference between the two cohorts. All mothers in the 2019 cohort gave birth on or after December 3, 2017, when extended parental benefits became effective in the EI parental benefit program, while some mothers in the 2018 cohort gave birth before that date.

Differences in the likelihood of mothers having planned to return to work by mother’s age, education and other job characteristics, such as permanent job status, occupation, union status and full-time job status, were statistically insignificant after all other differences were considered. Also, whether a mother was a lone parent was not a statistically significant factor in understanding mothers’ return to work.^Note ,^Note



Discussion and conclusions

This study examined the likelihood and timing of mothers with an infant younger than 12 months returning to work after a break following childbirth or adoption using data from the EICS.

This study first compared the characteristics and patterns of new mothers returning to work after parental leave in the 2009 and 2019 cohorts. The results showed that, in Canada, mothers with an infant had a higher level of human capital and made more investment in their career before childbirth than mothers a decade earlier. Compared with the 2009 cohort, mothers in the 2019 cohort were more likely to have a university education, long job tenure and high hourly wages. They were also more likely to receive parental benefits or employer top-up payments. On average, recent mothers were less likely to withdraw from the labour market. Compared with mothers in the 2009 cohort, mothers in the 2019 cohort were more likely to return to work at some point after parental leave (88% versus 82%).

More importantly, the use of leave and the return-to-work behaviours have differentially changed by mothers’ characteristics in recent years. A larger increase in the rate of post-childbirth employment was observed among less advantaged mothers, especially mothers with a temporary or low-paying job with an hourly wage of less than $15. The return-to-work rate also increased among mothers who did not receive parental benefits and mothers who worked in sales and service occupations. On top of the overall increase in the return-to-work rate, differential changes in the timing of the return to work were found based on certain characteristics. The percentage of mothers returning within 12 months of leave decreased among high-earning mothers in the 2019 cohort (highly educated or high-wage mothers) but increased among less advantaged mothers (with no postsecondary education and with an hourly wage less than $15). These findings suggest that it may be difficult to take a long break from a low-paying job possibly because of financial difficulties or the type of job. According to a recent qualitative study based on interviews with 46 Canadian employers, employers reported a low uptake of extended parental leave among employees because of the lower EI payment, and the employers had concerns related to the adaptation of this program with their human resource practices, such as backfilling positions and potential extra costs associated with top-up payments (Pettigrew, 2020).

The study further used regression analysis to examine the characteristics associated with the likelihood and timing of mothers returning to work after parental leave. Canadian-born mothers were more likely to return to work than foreign-born mothers. Mothers who received parental benefits showed higher return-to-work rates than other mothers. Mothers who received employer top-up payments and mothers with a longer tenure at their pre-childbirth job were more likely to return to work at some point after childbirth or adoption than other mothers. The last finding suggests that mothers who work for a family-friendly employer or an employer that offers generous employee benefits and mothers who accumulated a larger amount of job-specific human capital were more likely to return to work than other mothers.

Province of residence was an important predictor of when mothers return to work. Among recent mothers, mothers living in Quebec were 17 to 31 percentage points more likely to return to work within 12 months of leave than mothers living in Ontario and Western Canada, whereas the likelihood of not returning to work was similar across all provinces. Descriptive results demonstrated that, from 2009 to 2019, the share of mothers who intended to return to work within 12 months decreased among mothers living in British Columbia, Alberta and Ontario (e.g., from 68% to 55% in Ontario) but increased substantially among mothers in Quebec (from 64% to 79%). The provincial differences in the timing of the return to work may be related to differences in child care affordability and accessibility and the difference in the availability of extended parental benefits in Quebec and other regions. Child care for an infant younger than 18 months is costly in many provinces; however, low-cost child care may be available in Quebec, even for infants younger than 1 year old, since 2000. In the other provinces, where an option for 18-month parental leave was introduced and child care costs for infants were higher compared with Quebec, the percentage of mothers who returned to work within 12 months decreased. These findings suggest that the availability of low-cost child care is an important factor in the timing of mothers’ return to work. Another relevant difference between Quebec and the other regions at the time of interest in this study is the availability of paternity leave, which became available in Quebec in 2006. Fathers’ use of parental leave (including paternity leave) in Quebec may complement the availability of low-cost child care and help mothers return to work sooner.

Regression results found a difference between the 2018 and 2019 survey cohorts in the timing of the return to work, with no difference in the likelihood of not having a plan to return to work. This finding suggests that mothers in the 2019 cohort were more likely to use parental leave longer than 12 months than those in the 2018 cohort. One plausible explanation for the differences in the 12-month return rate between the two cohorts is the availability of extended parental benefits as of December 2017. These extended benefits provided mothers with up to 61 weeks of maternity and parental benefits. While all mothers in the sample for the 2019 cohort had a child who was born after the effective date of the extended parental benefit program, some of the mothers in the 2018 cohort gave birth to or adopted a child before the date and did not have this option. This finding aligns with the trends that were documented using the EI administrative data. According to the 2019/2020 Employment Insurance Monitoring and Assessment Report (Canada Employment Insurance Commission, 2021), new claims of extended parental benefits increased from 8,700 in the 2017/2018 fiscal year to 31,910 in the 2018/2019 fiscal year and to 37,770 in the 2019/2020 fiscal year.^Note

The findings and limitations of this study put forward some avenues for future research. The results suggest that provincial differences between Quebec and the rest of Canada, such as the availability of low-cost child care and an option for extended parental benefits, are an important factor in the timing of mothers’ return to work. Previous research has shown that subsidized low-cost daycare services in Quebec increased the employment of Quebec mothers with preschool-aged children (Baker, Gruber and Milligan, 2008; Lefebvre and Merrigan, 2008). The implementation of the Canada-wide early learning and child care (ELCC) program, through lower cost and increased accessibility of ELCC programs, may bring changes similar to those experienced in Quebec, in terms of the intentions and the timing of mothers’ return to work, to the rest of Canada. Replicating the current study with subsequent years of data may be insightful in the examination of how return-to-work rates will be impacted by the establishment of affordable and accessible ELCC across Canada and by the COVID-19 pandemic.

Although this study provided important information about mothers’ return (or intention to return) to work, it did not examine post-childbirth or -adoption careers in the long term. This study suggested that mothers were away from work longer as extended parental benefits became available. Some studies have found that a very long parental leave can have a negative impact on mothers’ employment and careers after they return to work (Lalive and Zweimüller, 2009; Evertsson and Duvander, 2011). For example, Evertsson and Duvander (2011) found that Swedish mothers who used leave for 16 months or longer were less likely to experience an upward occupational move after their return. Examining whether the post-childbirth or -adoption careers of Canadian mothers were affected by the availability of extended parental benefits is an interesting area for future research.

The present study provides an in-depth analysis of Canadian mothers’ return to work after parental leave. Results suggest that over this recent decade, as mothers in Canada had a higher level of human capital and made more investment in their career before childbirth, they were more likely to return to work after parental leave and use longer leave. However, not all mothers experienced the same changes over time. While more mothers, on average, expected to use longer leave over time, less advantaged mothers were less likely to do so. Results also suggest that access to low-cost child care and paternity leave in Quebec may be associated with a higher likelihood of returning to work within 12 months (versus returning to work later). Canada-wide access to low-cost child care for families may facilitate post-childbirth or -adoption employment for Canadian families and reduce the differences in the return-to-work behaviour between advantaged and less advantaged mothers.

Appendix

 

References

Aisenbrey, S., M. Evertsson, and D. Grunow. 2009. “Is there a career penalty for mothers’ time out? A comparison of Germany, Sweden and the United States.” Social Forces 88(2): 573–605.

Baker, M., J. Gruber, and K. Milligan. 2008. “Universal child care, maternal labor supply, and family well‐being.” Journal of Political Economy 116(4): 709–745.

Baker, M. and K. Milligan. 2008a. “How does job‐protected maternity leave affect mothers’ employment?” Journal of Labor Economics 26(4): 655–691.

Baker, M. and K. Milligan. 2008b. “Maternal employment, breastfeeding, and health: Evidence from maternity leave mandates.” Journal of Health Economics 27: 871–887.

Baxter, J. 2008. “Timing of mothers’ return to work after childbearing: Variations by job characteristics and leave use.” Research Paper No. 42. The Australian Institute of Family Studies: Melbourne, Australia.

Browning, M. 1992. “Children and household economic behavior.” Journal of Economic Literature 30(3): 1434–1475.

Burgess, S., P. Gregg, C. Propper, and E. Washbrook. 2008. “Maternity rights and mothers’ return to work.” Labour Economics 15(2): 168–201.

Canada Employment Insurance Commission. 2021. 2019/2020 Employment Insurance Monitoring and Assessment Report. Employment and Social Development Canada.

Connelly, R. 1992. “Self-employment and providing child care.” Demography 29(1): 17–29.

Del Rey, E., A. Kyriacou, and J. Silva. 2021. “Maternity leave and female labor force participation: evidence from 159 countries.”  Journal of Population Economics 34: 803–824.

Evertsson, M. and A. Duvander. 2011. “Parental leave—possibility or trap? Does family leave length effect Swedish women’s labour market opportunities?” European Sociological Review 27(4): 435–450.

Findlay, L., L. Wei, and R. Arim. 2021. “Patterns of participation in early learning and child care among families with potential socioeconomic disadvantages in Canada.” Economic and Social Reports 1 (8): 1–19. Statistics Canada Catalogue no. 36-28-0001.

Geyer, J., P. Haan, and K. Wrohlich. 2015. “The effects of family policy on maternal labor supply: Combining evidence from a structural model and a quasi-experimental approach.” Labour Economics 36: 84–98.

Jeon, S.-H., and Y. Ostrovsky. 2019. “Balancing family and work: Transitions to self-employment among new mothers.” Oxford Economic Papers 71(1): 47–72.

Kingsbury, M., L. Findlay, R. Arim, and L. Wei. 2021. “Differences in child care participation between immigrant and nonimmigrant families.” Journal of Childhood Studies 46(4): 46–58.

Lalive, R. and J. Zweimüller. 2009. “Does parental leave affect fertility and return to work? Evidence from two natural experiments.” The Quarterly Journal of Economics 124(3): 1363–1402.

Lefebvre, P. and P. Merrigan. 2008. “Child‐care policy and the labor supply of mothers with young children: A natural experiment from Canada.” Journal of Labor Economics 26(3): 519–548.

Marshall, K. 1999. “Employment after childbirth.” Perspectives on Labour and Income. Statistics Canada Catalogue no. 75-001-XPE.

Pettigrew, R. 2020. “Canadian employers’ reaction and policy adaptation to the extended, 61-week parental leave.” Canadian Studies in Population 47: 97–109.

Pylkkänen, E. and N. Smith. 2003. “Career interruptions due to parental leave: A comparative study of Denmark and Sweden.” OECD Social, Employment and Migration Working Papers, No. 1, OECD Publishing, Paris.

Rønsen, M. and M. Sundström. 2002. “Family policy and after-birth employment among new mothers – A comparison of Finland, Norway and Sweden.” European Journal of Population 18: 121–152.

Rybczynski, K. 2015. “What drives self-employment survival for women and men? Evidence from Canada.” Journal of Labor Research 36(1): 27–43.

Statistics Canada. No date, a. Table 14-10-0327-01  Labour force characteristics by sex and detailed age group, annual. https://www150.statcan.gc.ca/t1/tbl1/en/tv.action?pid=1410032701.

Statistics Canada. No date, b. Table 37-10-0130-01 Educational attainment of the population aged 25 to 64, by age group and sex, Organisation for Economic Co-operation and Development (OECD), Canada, provinces and territories. https://www150.statcan.gc.ca/t1/tbl1/en/tv.action?pid=3710013001.

Statistics Canada. No date, c. Table 17-10-0005-01 Population estimates on July 1st, by age and sex.

Statistics Canada. 2020. Microdata User Guide, Employment Insurance Coverage Survey, 2019.

Waldfogel, J., Y. Higuchi, and M. Abe. 1999. “Family leave policies and women’s retention after childbirth: Evidence from the United States, Britain, and Japan.” Journal of Population Economics 12(4): 523–545.

Wellington, A. 2006. “Self-employment: The new solution for balancing family and career?” Labour Economics 13(3): 357–386.

Zhang, X. 2007. “Returning to the job after childbirth.” Perspectives on Labour and Income. Statistics Canada Catalogue no. 75-001-XPE.

Zhang, S., R. Garner, L. Heidinger, and L. Findlay. 2021. “Parents’ use of child care services and differences in use by mothers’ employment status.” Insights on Canadian Society. Statistics Canada Catalogue no. 75-006-X.

Zoch, G. and I. Hondralis. 2017. “The expansion of low-cost, state-subsidized childcare availability and mothers’ return-to-work behaviour in East and West Germany.” European Sociological Review 33(5): 693–707.