By Katherine Marshall
Full article in PDF format
The paid work week has recently shown a steady decrease in Canada and in most other Organisation for Economic Co-operation and Development (OECD) countries (Usalcas 2008). While a drop in the average time spent at the job may suggest more personal time at the individual level, disposable family time is contingent on the combined paid work schedules of family members. In fact, overall family work hours have increased because the number of contributors has increased. In 2008, dual-earners accounted for three-quarters of all couples with dependent children—up from just over one-third in 1976. Although individual paid hours are well documented, less is known about employment hours and earnings within families.
More families with two earners means less time available for unpaid work and leisure activities. One potential concern might be that parents are spending less time with their children. However, this may not necessarily be true since people make choices about how to spend their time. Indeed, research has shown that, at the expense of other activities, both mothers and fathers in dual-earner families have increased the time they spend on child care (Bianchi 2000 and Marshall 2006).
A second concern is the pressure and stress parents experience when attempting to manage work and family responsibilities. The issue of juggling paid and unpaid work has helped spur the creation of many workplace programs and policies such as dependent care initiatives, work–life stress management, workplace flexibility, and leave and benefits (HRSDC 2007). Understanding the labour market dynamics within families helps with the ongoing development of such practices.
The third major area of interest is family role specialization. Research has shown that although couples are increasingly sharing economic and domestic responsibilities within families, a gender division of labour is still evident (Marshall 2006). Within many dual-earner couples, women continue to spend relatively more time on domestic work and men more time on economic work. The increase in dual-earners has slowed, so perhaps the evolution of breadwinning patterns within couples has as well.
Using the Labour Force Survey (LFS), this study examines trends in the total hours worked by employed couples (those with at least one spouse employed), the distribution of single- and dual-earning families, and the proportion of hours and earnings contributed by dual-earner spouses1 over the past decade. Finally, the General Social Survey is used to investigate work-hour preferences and perceptions of work–life balance (WLB) and personal stress among dual-earners (see Data sources and definitions).
Total weekly employment hours of couples increased from an average of 57.6 in 1976 to 64.8 in 2008 (Chart A)—a 13% increase and the equivalent of just under one full day of paid work per week (7.2 hours). However, this trend masks a change in the type of earning family (single versus dual), even though the average hours for each type have not changed. In other words, more families have two concurrent earners,2 but the time that two-earner families spend on employment has remained remarkably stable. In 1976, the combined hours of dual-earner husbands and wives averaged 77.6, whereas in 2008 the figure was 76.7. The slightly higher rate in 1976 may be due to the inclusion of usual overtime hours among employees prior to 1997 (see Data sources and definitions). As with total family hours, 2008 was the first time in five years that dual-earner hours changed significantly, down by 0.5 from 2007, likely reflecting the global economic downturn.
The proportion of couples with both spouses employed rose steadily from 4 in 10 in the mid-1970s to around 7 in 10 in the late 1990s, when it began to level off (68% in 2008). The slower growth rate of dual-earners in the last 10 years is reflected in the flattening of total family hours. While single-earner families have declined, the role of earner within these families has changed. Between 1976 and 2008, the proportion of families with a single-earner husband dropped from 53% to 21%, while those with a single-earner wife increased from 4% to 10%. These trends are further accentuated among families with dependent children at home. For example, the proportion of dual-earner families with preschool children at home (under age 6) rose from 31% to 67% over the past 30 years, while the rate among those with older children (youngest between 6 and 15) climbed from 45% to 77% (Chart B). The remainder of this paper focuses on the employment dynamics within dual-earner families from 1997 (when the Labour Force Survey began collection of earnings data) to 2008. Since 1997, usual hours have included only hours worked for regular pay.
Although the combined average work hours of dual-earner couples have remained around 77 since 1997, the distribution around this average has changed in recent years. In 2008, 59% of dual-earner couples worked between 65 and 80 combined hours per week—up from 54% in 1997 (Table 1). The growth in ‘standard' hours is due mainly to small decreases in the proportions of families working short and long hours. Furthermore, even though roughly one in four dual-earner families worked more than 80 hours per week in all years considered, the average for these long-work week couples dropped from 99 hours in 1997 to 96 in 2008. Studies show that couples working more than 100 combined hours per week are particularly pressed for time and are more likely to report increased levels of personal stress (Jacobs and Gerson 2001, and Larochelle-Côté and Dionne, forthcoming).
However, the real story behind the standardization of hours for dual-earner couples is the change in the overall contribution by each spouse. Although the length of the family work week has stayed the same, the average hours of wives have steadily increased while husbands' hours have been decreasing (Chart C). The distribution of individual spousal hours has moved towards a standard work week, particularly for wives. With wives working longer hours, the proportion of dual-earners who both work full time increased from 70% to 74% between 1997 and 2008.
In 2005, the majority of full-time dual-earners reported their current hours as the preferred arrangement, but 13% of husbands and 16% of wives preferred to work fewer hours for less pay. Also, women in dual-earner couples report higher levels of ‘time crunch' and dissatisfaction with work–life balance than men—particularly with preschool children present (see Perceptions and preferences of dual-earners).
The convergence of paid work hours within dual-earner couples has led to an increase in wives' contribution to total family employment hours, up from 43.8% in 1997 to 45.3% in 2008 (Table 2). In other words, the net difference in weekly work hours has dropped from an average of 9 to 7, and, viewed annually, this represents an increase of more than one week of full-time work for women and a drop of two weeks for men. Other research has shown a similar convergence in both paid and unpaid work hours within dual-earner couples.3 These trends suggest that the dual-earner model may be further evolving into what has been termed “marriages of equally dependent spouses” (Nock 2001).
Many factors can influence the convergence of paid work hours for spouses, including change in the industrial and occupational structure, educational attainment and labour market opportunities, and individual and family preference and choice. Recent research has shown that long hours have been declining because of an employment increase in the service sector, a decline in self-employment and self-employment hours, and a shift toward standard hours among those with higher educational attainment (Usalcas 2008). These trends are more likely to affect men's average hours since they have traditionally been more likely to work long hours. Indeed, spousal hours in self-employed couples are much more similar now than in the past. For example, dual-self-employed couples averaged the longest work week in 1997 (88 combined hours) and in 2008 (84 hours)—but while wives' average hours were similar in both years, husbands' hours dropped from 53 to 48.
Although higher educational attainment may have a dampening effect on long hours, it has also opened up labour market opportunities and enhanced women's labour market attachment. The proportion of women age 25 and over with a university degree in Canada rose from 14% in 1997 to 22% in 2008 (for men, from 18% to 23%). Furthermore, younger women now have higher levels of educational attainment than men—in 2008, 32% of women 25 to 44 had a university degree compared with 26% of men—and consequently dual-earner couples with a university-educated wife have increased substantially. Wives' higher educational investment increases their chances of strong labour market attachment since a university education is associated with higher labour force participation rates, higher-quality job opportunities and higher earnings. In both 1997 and 2008, wives' hours were, on average, longer, and family hours more similar when the wife had a university education.
Another noteworthy change in spousal hours is seen when dependent children are at home. Average paid work hours have converged considerably among those with children under 6 at home, with wives' hours increasing from 32 to 34 since 1997, and husbands' hours decreasing from 44 to 42. Not only are hours now higher for mothers with young children, but their labour force participation has also increased, from 37% in 1976 to 72% in 1997 and to 74% in 2008—further indication that parenthood does not alter women's employment patterns to the same extent as in the past. As well, men's increasing involvement in family responsibilities, such as more time spent on housework and child care, and taking parental leave may be part of the reason for their decreasing paid work hours. As the family earnings model has evolved over time, so too have the role expectations of spouses (Beaujot 2006).
In 2008, dual-earners with paid jobs (70% of all couples) earned an average of $1,770 per week before taxes—a real increase of about 10% since 1997 (Table 3). Almost the entire earnings increase was due to higher hourly earnings (up 9.2%) rather than an increase in hours worked (0.8%).4 The well-documented male–female earnings gap is evident in the average hourly rates of dual-earner spouses, with wives earning 81% as much as husbands in 2008 ($21.10 vs. $26.20) (Drolet, forthcoming). However, the difference has narrowed since 1997, when wives earned 77% of what their husbands earned. Since both hours and earning power have increased for wives, their overall contribution to family weekly earnings increased to $740 in 2008, representing over 41% of the total (Chart D). Although the change in wives' contribution to family hours and earnings has been relatively small since 1997, the trend showed steady and often significant annual increases over the past decade.
Primary breadwinning is often defined as one partner bringing in most of the family's income, but it can also be based on the contribution of time in the labour market. Time is an important factor in the notion of role specialization, and focusing on only the financial aspect of breadwinning “could be disguising interesting time dimensions to gendered breadwinning work roles” (Warren 2004). Using a common categorization of primary, equal and secondary breadwinning, 65% of wives were considered equal workers in terms of weekly paid hours in 2008, up from 60% in 1997 (Table 6). In other words, almost two-thirds of couples had weekly work hours within 10% of each other, which means each spouse contributed between 45% and 55% of the total weekly hours. For example, if a couple had a combined work week of 80 hours, the contribution by the wife would have been within the range of 36 to 44 hours.
Due to both their lower hourly earnings and their relatively shorter work weeks, most wives contributed less than 45% of total family earnings, making the majority of them the secondary earner in both years. However, between 1997 and 2008 the proportion of wives as equal or primary earners increased from 37% to 42%. The gradual convergence of hours and hourly earnings of husbands and wives in dual-earner couples suggests that the economic roles within families are continuing to change and that “equal breadwinning is on the rise” (Raley et al. 2006). Changing breadwinning roles—such as when a wife becomes the primary breadwinner, or when one spouse contributes equally or more in terms of hours but remains the secondary earner—can lead to social and psychological changes within families. The “implications of new earning arrangements for couples' marital happiness and well-being” are worth further investigation (Raley et al. 2006).
The combined paid work hours of couples increased from an average of 58 per week in 1976 to 65 in 2008. However, this statistic hides two underlying trends—changes in the type of earning family and the earning dynamics between spouses. The number of dual-earner couples rose from 1.9 million (43% of couples) to 4.2 million (68% of couples). However, the average combined hours of dual-earner couples remained constant at around 77 per week.
The increasing number of full-time, dual-earner families continues to make work–life balance an important issue. Fewer families have a parent at home, either full-time or part-time, to help manage the household, to provide child care, and, increasingly, to provide elder care. Fewer one-earner families suggests that “a decline in support at home rather than an increase in the working time of individuals underlies the growing sense that families are squeezed for time and that work and family life are in conflict” (Jacobs and Gerson 2001). Around one in four men in dual-earner families with young children at home, and more than one in three women, reported feeling severely time stressed—a state associated with significantly lower rates of WLB satisfaction. Not surprisingly, women also expressed more dissatisfaction with work–life balance than did their male counterparts. Interestingly, the majority of both men and women who expressed severe time stress and WLB dissatisfaction reported a preference for their current work hours or for even more hours, suggesting perhaps that in some cases family economic security is seen as more important than personal welfare. There is increasing documentation on the need for a more family-supportive workplace, including guides to help employers, managers and policy makers make such accommodations (see, for example, Barrette 2009 and Lero et al. 2009)
Earning patterns within dual-earner families have also changed. The average weekly hours of husbands and wives have converged from a difference of more than 9 in 1997 (43.3 and 33.8 respectively) to just over 7 in 2008 (42.0 and 34.7), placing two-thirds of couples in an equal work-hours category (their hours being within 10% of each other's). However, the combination of relatively shorter hours and lower hourly earnings placed more than half of wives (57%) as secondary earners in 2008 since they contributed less than 45% of total family earnings. On average, couples earned $1,770 per week before taxes—$1,040 by husbands and $740 by wives.
Women's increasing educational attainment and earning power offers them further opportunity to contribute equally or more to family revenue, but counter-influencing factors include the male–female earnings gap, social expectations about the primary breadwinner role, and personal and family choices about paid and unpaid work arrangements (Raley et al. 2006). However, younger women and men tend to have more neutral views on family roles since most have grown up in dual-earner households. As these younger cohorts continue to enter the labour force, even further changes in employment and earnings patterns within families may emerge.
The Labour Force Survey (LFS) is a monthly household survey that collects information on labour market activity over a one-week period from all persons 15 years and over, including questions about the usual and actual weekly hours at a main, and any other, job. This paper examines usual hours worked, which better reflects the regular or average week-to-week work schedules of families.
The General Social Survey (GSS) is an annual household survey that collects information on a wide range of social trends and policy issues. Data are collected monthly from one household member age 15 and over. Two of the cycles on time use, 1998 and 2005, collected information on both ‘time crunch' (see definition below) and work-hour preference among those employed full time.
The target population includes all married and common-law couples with at least one spouse employed at the time of the survey. Single-earner couples are those with one spouse employed and the other either unemployed or not in the labour force. Dual-earner couples are those with both spouses employed during the survey reference week. Dual-earners can also be defined as both husband and wife reporting some employment income during the past year. A dual-earner rate based on current labour market participation will be lower than one calculated using the incidence of annual employment income. The LFS collects labour market activity information at the individual level. For this study, total individual weighted counts have been divided in half to reflect a count for couples. For example, in 2008, the 12,188,000 husbands and wives from the same households equates to 6,094,000 couples.
Actual hours worked during the reference week includes any paid or unpaid overtime. This measure reflects temporary increases and decreases in weekly work hours due to illness, vacation, overtime and irregular work schedules.
Usual hours worked excludes overtime. For the self-employed, it refers to the number of hours usually worked at the business in a typical week, regardless of whether they were paid. The definition of usual hours has remained unchanged for the self-employed since 1976. However, prior to 1997 employees were to include overtime hours in their estimate if they were typical to their schedule. Although the change is likely to result in a slight downward shift in the estimates of usual hours, this is not deemed problematic for this study since the main focus is the changing dynamics within families. In other words, any downward shift in estimates would equally affect husband and wife hours.
Secondary, equal and primary breadwinner categories are based on the contribution made by each spouse to the couple's overall time spent in paid work per week and their hourly and weekly earnings. Partners are defined as having the same or equal hours or earnings if they contribute between 45% and 55% of the total, secondary if less than 45%, and primary if greater than 55%. Several studies have used the 10% range to represent ‘equal,' while others have used a wider range of 20% (Warren 2004).
Collection of earnings information began in 1997 from all employees for their main job. Respondents are asked to report their hourly rate of pay or their regular salary (weekly, bi-weekly, etc.) before taxes and other deductions, and including tips, commissions or bonuses. Hourly and weekly earnings are calculated in conjunction with usual paid work hours per week.
Time crunch stress is determined by the number of positive responses to 10 statements:
Seven or more yes responses is considered severely time crunched (Frederick 1993).
Time becomes a more precious commodity when there is less of it, and, arguably, families with two full-time jobs and preschool children at home have relatively high demands on their time. Not only is the level of care more intense with young children, but parents can also be “affected by the tension generated by day-care difficulties” (Barrette 2009). Indeed, 24% of men and 38% of women in such families report severe time-crunch stress (Table 4). Someone feels severe time stress if they have responded yes to at least 7 out of 10 statements including: “At the end of the day, I often feel that I have not accomplished what I had set out to do“ or “I feel that I just don't have time for fun any more“ (see Data sources and definitions for more details). Age and stage of life play a role as both men and women without dependent children at home report significantly lower rates of time stress.
As found elsewhere, regardless of the presence and/or age of children at home, employed women tend to report higher rates of time stress than employed men. Some possible explanations for the difference include time availability, social norms on what represents a successful parent, and the quality of personal leisure time (Marshall 2006, and Nomaguchi et al. 2005). The 2005 time stress rates are very similar to those found among dual-earner parents in both 1992 and 1998. Although the degree of time stress may not be rising among dual-earners, the number of people affected is likely higher because of the increase in the number of dual-earner families.
Not surprisingly, as the level of time stress increases, the degree of satisfaction with work–life balance decreases. For example, over 90% of both dual-earner men and women with low time stress express being satisfied with their work–life balance, whereas only half of those with high time stress report such satisfaction (Chart E). Excessive time stress is related to a diminished sense of well-being for mothers and fathers (Nomaguchi et al. 2005). Furthermore, long-term exposure to work–family conflict can also lead to negative physical health (e.g. hypertension, cardiovascular disorders, migraines) and psychological health (e.g. depression, anxiety, irritability) consequences (Frone 2000 and Barrette 2009).
Asked about preferred hours of work, an equal proportion (60%) of full-time, dual-earner men and women opted for the same hours and same pay (Table 5). Just over 1 in 10 would prefer more hours for more pay, while 13% of men and 16% of women would prefer fewer hours of work for less pay. While no differences were seen in reported work-hour preferences among dual-earner men and women without children, the desire for fewer hours increased to 20% for women with dependent children at home, and, conversely, the desire for more hours increased for men with preschool children (19%). Even among those reporting high levels of time stress or dissatisfaction with their work–life balance, only about 1 in 4 men and women reported a preference to reduce their work hours, and 1 in 10 would prefer to work even more hours, while half are content to keep their current hours. The reality may be that even though people may feel too busy, they may also feel that they cannot financially afford a reduction in their paid work hours.
Katherine Marshall is with the Labour and Household Surveys Analysis Division. She can be reached at 613-951-6890 or firstname.lastname@example.org.