Population Change And Lifecourse

Home About Us Knowledge Mobilization Networks Students Research Infrastructure PWFC

Knowledge Mobilization



Print (PDF)

Press Release

Research Brief No. 2
December 2009


Models of Earning and Caring: Trends, Determinants and Implications


Families in Canada have undergone changes regarding the ways in which they earn a living and care for each other. Data taken from Statistics Canada time use surveys of 1986, 1992, 1998, and 2005 show changes in the average number of hours of paid and unpaid work completed by men and women. We have defined five models of the division of work, and these are combined into two models for some of the analyses, as indicated below:


Definitions of models of the division of earning and caring activities
Complementary-traditional: wife is doing more unpaid work and husband more paid work.
Complementary-gender-reversed: husband is doing more unpaid work and wife more paid work.
Women’s double burden: wife is doing the same amount of, or more, paid work, and more unpaid work.
Men’s double burden: husband is doing the same amount of, or more, paid work, and more unpaid work.
Shared roles: wife and husband doing the same amount of unpaid work.
Augmented complementary-traditional: combination of complementary-traditional and women’s double burden, with women doing more unpaid work.
Augmented shared: combination of three other models with men doing the same amount or more unpaid work.
The shared roles model accounts for approximately one quarter of couples, and these report higher average measures of happiness and life satisfaction. Analysis of the various models shows that life course questions, as well as structural and cultural considerations are relevant. In particular, the presence of children, the relative resources of men and women, rural vs. urban lifestyle, language and religion all play a role in determining the models. The indicators of well-being and social support show mixed results across models. However, when one takes happiness and life satisfaction as an indicator, the shared model seems to have more positive implications for both men and women. Therefore, we propose a focus on social policies that promote equal opportunities in the broader society, and structural supports for gender egalitarianism in households. Given the aspirations for relationships based on mutuality and sharing, we propose that these supports would facilitate a continued increase in the number of families in the shared model.


Key Findings: Trends
  • The complementary-traditional, with men doing more paid work and women doing more unpaid work, is declining but remains the largest category representing one third of respondents.
  • The shared roles, where the unpaid work that each does is within 40 to 60 percent of the total unpaid work, is a growing category now representing over 25 per cent of respondents.
  • More than 25 per cent of couples are in the women’s double burden type and this proportion is stable.
  • Men’s double burden is increasing and represents 10 per cent of respondents.
  • While complementary-gender-reversed couples are increasing, they represent the smallest group at 3 per cent.


  • Couples are more likely to be in a shared roles model when women have more resources and when the couple is less religious.
  • Structural supports, such as adequate child-care facilities, support the shared model, and this is demonstrated by the higher number of couples in the shared model in Quebec (where there is more child care) as well as urban areas.
  • The presence of children is a major determinant. Women with children under 18, and men with children under 5, are more likely to be in a complimentary-traditional or women’s double-burden model.
  • There are a greater number of younger families in the shared-roles model.
  • Many of the determinants of sharing unpaid work involve factors associated with paid work, including the relative resources of spouses.


  • Questions regarding stress, health, life satisfaction, belonging and social cohesion, show that the differences are typically small across models and there is no clear “winner”.

  • The shared role model receives high average indicators on happiness and life satisfaction for both men and women, but it is high on stress for men.

  • The shared roles model is the type that many people prefer, and it maximizes the labour supply. It also has the potential to reduce the deterioration of skills that results from longer-term withdrawals from the work force. In this model, women are less vulnerable in the case of separation, divorce, or death of a spouse.

  • The shared roles model, with its advantages in terms of gender equity and ability to maximize labour force participation by all adults, is beneficial to society in many ways.




Families may be defined as people who share resources and care for each other. (Beaujot, 2000; Beaujot et al., 2005). Questions of gender equity in paid and unpaid work have been central to social inquiry over the last half-century. The unequal division of unpaid work has been called a second shift or a double burden that represented a stalled revolution in the direction of gender equity. Time use in paid and unpaid work, and the division of this work in families, are important in understanding accommodations that need to be made in the workplace and in social policy.  

The Study

The purpose of this paper is to follow the change in hours of paid and unpaid work of men and women, to look at alternate models for the division of this work within families, and to present the determinants and implications of alternate models. Data have been obtained from Statistics Canada time use surveys of 1986 to 2005. The paper’s  goal is to encourage broad discussion and debate about important public policy issues pertaining to the division of labour within families.



The data used here are from time-use diaries that were collected in the Statistics Canada Canadian General Social Surveys of 1986, 1992, 1998, and 2005. These are representative samples of the Canadian population, with sizes of 9946, 9815, 10749, and 19597 respectively. The core content of these time use surveys provides data on the daily activities of Canadians.  



Trends : The time-use in total productive activities, both paid and unpaid, for the total population has increased from 7.5 hours per day in 1986 to 8.1 hours in 2005 (Table 1).  Most sub-categories of unpaid work, including elder care, civic and voluntary activity, and housework, increase over ages for both women and men, reaching a peak at ages 65-79. However, childcare hours are at their highest when paid work is also high, that is at ages 25-44. Total productive activity increases for both men and women over the categories of “unmarried no children” to “married no children” to “married parent”. For the total population, the difference between men and women in average total productive activity has been at most 0.1 hours per day in eaTable 1ch of the four surveys. At the same time, this total productive activity has increased for employed men from 9.0 hours in 1986 to 9.6 hours in 2005, and for employed women it has increased from 9.2 hours to 9.8 hours. For the entire population, in 1986 women’s paid work plus education represented 58.9% of men’s time in these activities, compared to 72.2% in 2005. For unpaid work, men’s time in 1986 represented 46.3% of women’s time, compared to 62.8% in 2005. The differences are smaller in the employed population, but they are still in the same direction, with women doing 87.3% as much paid work as men, and men doing 71.4% as much unpaid work as women, in 2005.



Determinants: The study also analyses how the division of work of couples might be affected by variables such as marital and parental status, and socio-economic situation of individuals and couples. We assume that the manner of sharing paid and unpaid work within households is a function of choices.  But choices are not only based on individual rational calculations; they are also made in relation to others and guided by normative frameworks (Crompton, 2006: 13). Personal resources and relative power between couples  have an impact on the model as well. For example, men who have higher income or whose education is higher or equal to that of his partner are likely to have a partner who is doing more of the unpaid work.


Values and norms play an important role as well. Men who are highly religious are less likely to be in a shared-roles model, whereas men who are less religious are less likely to be in an augmented complementary traditional model. Also, men from a more traditional culture are more likely to be in an augmented complementary traditional model. People in rural areas tend to hold more traditional values than do those from urban areas and are therefore more likely to be in an augmented complementary traditional model.  


Consequences: The women’s double burden is the second largest category, and like the complementary-traditional, is more common for older people and those with young children. The complementary-traditional model is high on stress for men, while it is low on stress for women. However, women in this model are more likely to want to work more paid work hours. The women’s double burden model shows high stress and low life satisfaction for women. The men’s double burden is low on stress for women but shows poor health for men. The complementary-gender-reversed model is associated with poor health and happiness indicators for men, while women have good health but high stress.


For both men and women, the highest average level of happiness and satisfaction with life on a whole occurs within the shared-roles model.


By supporting the “shared roles” model, there would be support for the type of family model that many would prefer, and it would maximize the lifetime paid-work hours of women and men, with less “burn-out” and aspirations for early retirement. Further, women will be less vulnerable in the event of separation, divorce or death of spouse.



If more occupations provided greater opportunities for work-life balance, women and men would be able to enter fields corresponding to their interests and skills, rather than their potential to accommodate families. Greater gender parity in occupations and wages would also promote the shared model, as men are more likely to participate in unpaid work when their wife or partner earns high personal income.


We believe that a key policy challenge in Canada is that of accommodating the shared-roles model within diverse families. The shared-roles model promotes gender equality and maximizes the potential for all adults to be in the labour force. Studies have shown that egalitarian marriages are happier (Amato et al., 2007) and that, particularly for women, shared decision-making, more equal patterns of housework, and egalitarian gender attitudes are associated with higher marital satisfaction (Wilcox, 2008).  To promote this model, we suggest policies that support equal opportunities for men and women to access education and work, conditions that would facilitate work-life balance, and promoting greater involvement of men in housework and childcare. It is also important to provide programs with the potential to increase the work/life balance. This will allow for a greater alignment across occupations. Structural supports such as adequate childcare facilities as well as equal opportunity to parental leaves should also be a focus of public policy in Canada.  




Amato, Paul R., Alan Booth, David R. Johnson, and Stacey J. Rogers. 2007. Alone Together: How Marriage in America Is Changing. Cambridge, MA: Harvard University Press.

Beaujot, Roderic, 2000. Earning and Caring in Canadian Families. Peterborough: Broadview. 416 pp.

Beaujot, Roderic, Jianye Liu. 2005. Models of time use in paid and unpaid work. Journal of Family issues 26: 924-946.

Crompton, Rosemary. 2006. Employment and the Family: The Reconfiguration of Work and Family Life in Contemporary Societies. Cambridge: Cambridge University Press.

Wilcox, W. Bradford. 2008. Review of Alone Together: How Marriage in America Is Changing. Journal of Marriage and Family 70(1): 257-258.   .


About the study

The research paper “Models of Earning and Caring: Trends, Determinants and Implications” was commissioned by the Policy Research Directorate of Human Resources and Skills Development Canada (HRSDC). The views expressed do not necessarily reflect the opinions of HRSDC or the federal government. The paper was written by Roderic Beaujot and Zenaida Ravanera of University of Western Ontario; and Jianye Liu of Lakehead University. For more information, contact Prof. Roderic Beaujot at rbeaujot@uwo.ca. This research brief was written by Kelly Quance.  


Print (PDF)

Press Release