Archived Content

Information identified as archived is provided for reference, research or recordkeeping purposes. It is not subject to the Government of Canada Web Standards and has not been altered or updated since it was archived. Please contact us to request a format other than those available.

Highlights Highlights Main menu Editor's corner More news Contact us Survey information Back issues Statistics Canada home page In depth Français
Statistics Canada logo


system menu - text links at bottom of page
mast-head for "Perspectives on Labour and Income"
sub-heading "The online edition"

March 2003     Vol. 4, no. 3

Benefiting from extended parental leave

Katherine Marshall

The Unemployment Insurance Act (EIA)1 note 1  of 1940 introduced unemployment insurance to Canada, but it was another 30 years before the Act provided provisions for maternity leave. Starting in 1971, mothers with 20 or more insurable weeks could claim up to 15 weeks of benefits. Almost two decades later, in 1990, 10 weeks of parental leave benefits were added. These could be used by either parent or split between them (HRDC 1996). Another significant change in December 2000 increased parental leave benefits from 10 to 35 weeks, effectively increasing the total maternity and parental paid leave time from six months to one year. As well, the threshold for eligibility was lowered from 700 to 600 hours of insurable employment. However, the rate of benefit remained unchanged at 55% of prior weekly insurable earnings up to a set maximum (see Parental benefit revision).

One aim of the 2000 amendment was to enable working parents to care for their infant for longer and still allow them secure re-entry into employment. After the extension of parental benefits, all provinces and territories revised their labour codes to give full job protection of 52 weeks or more to employees taking paid or unpaid maternity or parental leave. note 2  Many other industrialized countries have moved to provide employment-protected parental leave as well. In 1996, the European Union (EU) passed a directive on parental leave mandating the right of all workers to at least three months leave (not necessarily paid) for childcare purposes (as distinct from maternity). As of 1998, 13 of the EU countries had statutory parental leave provisions, 2 did not (United Kingdom and Ireland), and one (Luxembourg) had limited provisions (Hall 1998).

The expansion of parental benefits has the potential to alter the labour market behaviour of both mothers and fathers. Do women now remain at home longer with their infants, and are there factors, such as income, that influence the length of leave time taken? Do women return to the same employer after longer periods of leave? This paper examines the labour market activity of mothers before and after the last paid parental leave amendment. Some of the events, such as returning to work, are based on both actual and intended behaviour (see Data source and definitions).

Overview findings

In both 2000 and 2001, over 300,000 mothers had infants at home (Table 1). In both years, roughly three-quarters of these mothers had been employed for at least one of the 52 weeks prior to the birth of the child—74% in 2000 and 77% in 2001.

Among mothers who worked prior to the birth of their child, 84% in 2000 and 82% in 2001 returned or planned to return to work within two years. note 3  The extension of paid leave does not appear to have affected mothers' return-to-work rate. An equal proportion of these women reported their reference job as paid (93%) (see Data source and definitions).

More mothers with paid jobs received maternity or parental leave benefits in 2001 (84%) than in 2000 (79%). This may be a result of the heightened awareness of the highly publicized revised parental benefit program and the reduction in the entrance requirement from 700 to 600 insurable hours. In any case, the combination of increased access to parental benefits and increased labour force participation of expectant mothers elevated the overall proportion of all new mothers receiving maternity or parental benefits from 54% in 2000 to 61% in 2001. Still, 39% of mothers with newborns in 2001 did not receive birth-related benefits because they were not in the labour force (23%), were paid workers who were ineligible or did not apply for benefits (12%), or were self-employed (5%).

A slightly smaller proportion of women who received EI reported receiving a financial top-up from either their employer or another source in 2001 than in 2000 (20% versus 23%). Women were much more likely to receive a top-up if they worked for a large firm. In 2001, 31% of those employed in firms of 500 employees or more received a top-up, compared with 18% of those in smaller firms. Also, the vast majority in both years returned to the same workplace, with 2001 showing a slightly higher rate—89% versus 84%.

Only about 3% of husbands claimed or planned to claim paid parental benefits in 2000, whereas by 2001 the figure more than tripled to 10%. This is not only a statistically significant increase, but also a socially significant one. Although the length of time involved is not known, approximately 1 in 10 fathers take a formal leave from their job to be at home caring for a newborn. Administrative EI data also shows a fivefold increase in the number of men receiving parental benefits since the amendment (Pérusse 2003). This parental leave benefit claim rate for fathers moves Canada ahead of many other countries, but still leaves it considerably behind those that offer non-transferable leave to fathers—Norway, for example, where almost 80% of fathers take parental leave (see International take-up rates among fathers).

One year off work more common now

For mothers who returned or planned to return to work within two years of childbirth, the most common return time changed from 5 to 6 months in 2000 to between 9 and 12 months in 2001 (Chart A). Clearly a result of the longer paid-benefit period, the proportion of women returning to work after about a year off (9 to 12 months) jumped from 8% to 47% between the two years.

Roughly 1 in 10 women in both years took either no time, or only one or two months, off work after childbirth. The vast majority of these early returnees were self-employed or employees without maternity or parental leave benefits. At the other end of the spectrum, for both years, less than 2 in 10 women did not plan to return to work, or did plan to return and either did not know when or gave a date beyond two years.

Time off jumps from 6 months to 10 for benefit recipients only

Among self-employed women who returned to work within two years, the median time off work was only one month in both 2000 and 2001 (Chart B). note 4  Previous research supports this finding, and suggests that entrepreneurs on leave can face a double financial loss, not only because of their own lost earnings but because of the possible expense of hiring a replacement worker (Marshall 1999). And, since the self-employed do not pay into the Employment Insurance program, they are not entitled to maternity or parental leave benefits. The median length of time off work also changed very little for employees not receiving maternity or parental benefits—five months in 2000, and four months in 2001. The self-employed and employees without benefits accounted for a minority of the total who were previously employed and had returned (23% in 2000 and 19% in 2001).

Most women who returned or planned to return to work were employees in receipt of maternity or parental leave benefits: 77% in 2000 and 81% in 2001. And it is this group that appreciably extended their stay at home following the program amendment. The median time at home for women with benefits increased from 6 months in 2000 to 10 months in 2001. Although there is some variation around the median, most recipients were concentrated in a narrow band around this figure. Two-thirds (67%) took or planned to take 9 to 12 months, one-quarter took 8 or less, and the remainder took 13 to 24.

Key factors in length of paid time off

Father's take-up rate of benefits

Although most employees with benefits took advantage of the revised parental leave program and were, or planned to be, off work for almost a year, one-quarter of the women took less than 9 months off (median of 5 months) (Table 2). The two groups share many similarities; they had roughly the same median age (30), the same marriage rate (95%), and the same education (7 out of 10 had a post-secondary diploma or university degree). However, fathers' participation in the program differed significantly. Almost one-quarter of the husbands of women who took less time off claimed or planned to claim benefits, while only a handful of husbands of the long leavetakers did so. Logically, if fathers claim some of the 35 paid parental leave weeks, mothers would have less than a year of paid leave for themselves, and thus a shorter stay at home. Further analysis note 5  indicated that women with partners who claimed or planned to claim parental benefits were 4.6 times more likely to return to work within eight months than those with partners who did not claim benefits.


Significantly more mothers who returned within eight months reported annual earnings below $20,000 in their previous or current job (49%), compared with those who returned after almost a year (29%). note 6  In other words, lower individual earnings were associated with a quicker return to work (Chart C). For example, mothers with maternity or parental leave benefits who returned to work within four months had median annual earnings of just under $16,000. This suggests that women with lower earnings (and possibly lower savings) may not be financially able to stay at home for an entire year on 55% of their earnings.

Since personal income influences total household income, early returnees were also more likely to be part of a household whose total income was under $40,000—46%, compared with 38% for those who returned between 9 and 12 months (Table 2). However, when household income is compared with all other variables at the same time, by way of regression analysis, the mother's earnings are clearly the overriding factor.

Receiving an employer top-up or other compensation in addition to paid maternity and parental benefits does not appear to affect the timing of returning to work. Just over a quarter of all employees who returned or planned to return to work within two years enjoyed this benefit. note 7  Although the top-up was substantial for many—half received a supplement large enough to equal 90% or more of their previous earnings—the median duration was only 15 weeks. The median weekly EI benefit rate was somewhat lower for those who returned sooner than for those who returned later ($300 versus $323), but the difference was not statistically significant.

Job permanency

The majority of mothers who took or planned to take a year off had worked full time in their previous or current job (87%), as had those who took less time off (82%). And, almost equal proportions (one-third) reported the job as unionized. However, one job-related factor that did determine a relatively early return to work, despite receipt of maternity or parental leave benefits, was whether the mother's job was permanent. Almost all (98%) of mothers on leave for a year had a permanent job, compared with 87% of those who returned in eight months or less. The job-permanency rate for benefit recipients who returned in four months or less was only 75%. Roughly 90% of these non-permanent jobs were temporary, term, contract or casual, and so would in theory be less likely to offer job protection. Those with non-permanent work were almost 5 times more likely to return to work in less than nine months compared with those with a permanent job.

Some of the key factors influencing the time away from work for women with maternity and parental benefits may be interrelated. For example, non-permanent jobs generally offer lower wages than permanent ones, so an early return to work might reflect the possibility of job loss, economic necessity, or both. Further analyses in subsequent years, when the entire sample will include births after the 2001 parental leave extension amendment, and upcoming data from the Survey of Labour and Income Dynamics may help shed further light on these questions.


Bill C-32 added 25 weeks of paid parental leave to the pre-existing 10. Including the 15 weeks of maternity benefits, parents can now receive up to a year of benefits while caring for their newborn children. Those who received these benefits experienced a significant increase in the time taken off work after the birth or adoption. Over 80% of these women returned or planned to return to work within two years, and the median time off increased from 6 to 10 months between 2000 and 2001. Despite the extended time off taken by most women who received benefits, one-quarter of them returned to work within eight months. Significant factors linked with a shorter leave from work included a father's participation in the parental benefit program, a mother's job being non-permanent, and low employment earnings. Even with the increased time away from work, women were equally likely to return to the same employer in both years.

However, the program amendment had no effect on those without access to parental leave—roughly 46% of all mothers with newborns in 2000 and 39% of those in 2001. The increased claim rate in 2001 was likely due to the increased employment rate of women before childbirth, as well as the increase in the proportion of employees qualifying for birth-related benefits. The mothers in 2001 without maternity or parental benefits consisted of those who were self-employed (5%), paid workers who did not qualify or apply for benefits (12%), and those who had not previously been employed (23%).

Since the extension of parental leave benefits, fathers' participation rate in the program has increased from 3% to 10%. So, not only are most newborns receiving full-time care by their mothers for longer, but many more are experiencing a father at home for some of the time as well.


Data source and definitions

The Employment Insurance Coverage Survey (EICS), a supplement to the Labour Force Survey (LFS) since 1997, studies the extent of coverage of the Employment Insurance program. Starting in 2000, a special maternity supplement was added to help monitor the effect of the extended parental benefit program, which began December 31, 2000.

The supplement asked new mothers detailed questions on their labour market situation before and after the birth/adoption of their child. Other information collected included the timing of any breaks before and after the birth/adoption, the receipt of EI by type and benefit level, as well as individual and household income prior to or since the birth/adoption. The survey also asked about spousal use of parental benefits, as well as some employer- and childcare—related questions. In cases where an event had not occurred—for example, a mother's return to work or a husband's claim for parental benefits-subsequent questions about intentions were asked. Calculations of the time off work are based on both completed and intended leave spells.

The sample included roughly 1,350 mothers with children less than 13 months of age in both the 2000 and 2001 surveys. However, almost 500 of those interviewed in 2001 had given birth or adopted their child in 2000 and were therefore excluded from the analysis. This paper examines the labour market behaviour of a sample of mothers who gave birth before and after the implementation of the parental benefit amendment, which means births in 1999 or 2000, and 2001.

A number of non-sampling errors, such as incorrect skip patterns, have led to some data quality issues, particularly with the 2001 file. Several variables have some missing responses, and in these cases calculations are based on valid responses only. The extent of the problems is not believed to seriously affect the results. Future cycles of the survey will resolve these problems.

Employment prior to birth: Women were considered employed if they reported working one or more weeks for pay or profit in any of the 52 weeks preceding the birth of the child.

Annual earnings were derived for all previously employed women by multiplying usual weekly hours of work by total weeks worked before birth (maximum of 52) by usual hourly earnings.

All respondents were asked to report total household income from all sources within a list of income ranges provided.

Women had an employed spouse if at the time of the survey they reported living in a husband-wife family in which the husband was employed.

Reference job characteristics were collected at the time of the LFS, which was 4 to 6 weeks before the EICS. For women who were not yet back to work, the term refers to their last main job held; for women who had already returned, it refers to their current main job.

If mothers, while pregnant or on leave, received employer payments, private insurance payments or other benefits in addition to EI maternity or parental benefits, they were considered as receiving a top-up.

Parental leave refers to a period of job-protected time granted to employees for the care and nurturing of their children. Currently, all provinces and territories offer at least 52 weeks to mothers and 37 weeks to fathers.

Parental benefits are available to previously employed qualifying parents (see Parental benefit revision).

Duration of time off work was calculated for all women who reported taking a break of one week or more after the birth/adoption of their child. For those who had already returned to work, the total weeks off work was recorded. For those who were not yet back to work, but who knew when they would return, the planned return date was recorded. In all cases, total time off was calculated as the time between the birth month and year of the child and the month and year of return. As expected, a greater percentage of return-to-work spells based on 'intentions' was noted for mothers who gave birth after the parental benefits amendment. Of all time-off spells that took place within two years, 74% were based on a specified future return date in 2001, compared with 40% in 2000.

Some precision is lost in calculating total time off in months rather than in weeks, but the more important issue is the change between 2000 and 2001. Also, total time off work may be underestimated because some women begin their maternity leave before the birth, since this benefit can be claimed up to eight weeks ahead of time.


Parental benefit revision

In 2000, Bill C-32 amended the Employment Insurance Act regarding paid parental leave in Canada. Starting December 31, 2000, leave time for employed parents increased from 10 to 35 weeks. Parental leave benefits can be claimed only after the birth of the child, and the leave must be taken within 52 weeks of the birth. To qualify, parents must have worked for 600 hours in the past 52 weeks, down from 700 previously. The 35 weeks of benefits can be taken by one (qualifying) parent, or they can be split between both (qualifying) parents, with only one waiting period required between them. The benefit entitlement remains at 55% of average insured earnings up to a maximum of $413 per week.

Maternity leave benefits, which are administered in the same way as parental benefits, can be claimed for 15 weeks by women only, and up to 8 weeks before the birth.

Although a discussion of sickness benefits is outside the scope of this paper, as of March 2002, these benefits no longer cut into the total eligible period for maternity and parental benefits. More information is available on the HRDC Web site (


International take-up rates among fathers

Even though the EU parental leave directive was implemented in 1996, most research shows that participation rates are high for mothers (90% or more) but not for fathers, even though the benefit is usually paid and available to both parents. Data from a number of European countries indicate that fathers' participation in parental benefits is often under 5% (Austria, Germany and Finland). Participation rates tend to be higher only in countries where parents are offered non-transferable paid parental leave (each parent must use the leave or lose it), such as Sweden and Norway where rates are 36% and 78% respectively. Many reasons have been put forward for the low parental benefit take-up rates for fathers including social, cultural and employer attitudes, the income rate while on leave, the level of job protection, and also "whether or not the mother wishes it" (OECD 2001). One reason for the increased claim rate in Canada (from 3% in 2000 to 10% in 2001) may be that fathers no longer face a two-week payless waiting period if their spouse has already served one. Another reason may be the length of time now offered for benefits—with 35 weeks available, mothers may be more willing to share some of the leave time with their partners.


  1. In 1996, the Unemployment Insurance Act became the Employment Insurance Act (EIA).
  2. Under provincial or territorial labour codes, job-protected parental leave is granted to those with continuous employment, which can range from less than a week to one year.
  3. This finding differs from a 1993-94 study of women returning to work after childbirth using the Survey of Labour and Income Dynamics (SLID), where 93% of women reported being back to work within two years. One reason for the difference may be that at the time of the EICS, about 8% of mothers were undecided about their future return. With the undecided removed, 90% of the women in the EICS also reported returning within two years.
  4. An error in the questionnaire meant that all self-employed women in 2000, and most in 2001, who had not yet returned to work were not asked about their intention to return. Therefore, the calculations are based on completed spells only and likely underestimate the true time off. However, the majority of the self-employed had already returned, and well over half did so in less than three months. This is consistent with analysis of self-employed mothers using the Survey of Labour and Income Dynamics, which found that 80% of those previously employed were back to work by the end of the first month after chilbirth (Marshall 1999). Also, the full 2001 survey was used in order to have a large enough sample for calculations in Chart B (that is, self-employed mothers who gave birth in 2000 were included).
  5. A logistic regression model was used to examine the probability of having taken less than nine months off work. The dichotomous dependent variable was less than 9 months (= 1) and 9 to 12 months (= 0). More information about the model may be obtained from the author.
  6. An assumption is made that employment before and after the birth is largely similar. This is based on the fact that well over 80% of the women return to the same employer, and 90% to the same hours (Marshall 1999).
  7. The overall top-up rates of 20% and 26% found in Tables 1 and 2 respectively, differ because of the population examined. The 26% includes only employees with maternity or parental benefits who had returned to work within two years.


  • Hall, Mark. 1998.The EU parental leave agreement and Directive: implications for national law and practice. European Industrial Relations Observatory online (EIROnline) (January). Internet: (accessed March 20, 2003).
  • Human Resources Development Canada (HRDC). 1996. The History of Unemployment Insurance. Ottawa: HRDC.
  • Marshall, Katherine. 1999. "Employment after childbirth." Perspectives on Labour and Income (Statistics Canada, Catalogue no. 75-001-XPE) 11, no. 3 (Autumn): 18-25.
  • Organisation for Economic Co-operation and Development (OECD). 2001. "Balancing work and family life: Helping parents into paid employment." Ch. 4 in OECD Employment Outlook. Paris.
  • Pérusse, Dominique. 2003. "New maternity and parental benefits." Perspectives on Labour and Income (Statistics Canada, Catalogue no. 75-001-XIE) 4, no. 3. March 2003 online edition.


Katherine Marshall is with the Labour and Household Surveys Analysis Division. She can be reached at (613) 951-6890 or

Statistics Canada FIP identifier Government of Canada wordmark
Highlights ]
Main menu | Editor's corner | More news | Contact us | Survey information | Back issues ]
Statistics Canada home page | In depth | Français ]

© Statistics Canada - Conditions of  use Published: 2003 03 21