Population Change and Lifecourse

2010 Statistics Canada Socio-Economic Conference

Monetizing Caregivers' Lost Wages

Donna Dosman, University of Alberta, Edmonton, Alberta and StatCan; Janet Fast, University of Alberta, Edmonton, Alberta; Geoff Rowe, Modelling Division (StatCan)

Current demographic, economic and policy imperatives mean that more people face competing employment and care demands and experience workfamily conflict. Many respond by reducing their paid work, but little is known about long term monetary consequences of such decisions.

Using Statistics Canada Labour Force Survey data for 1997 to 2006, we estimated that: • the number of employees who lost wages because of eldercare almost doubled over the last 10 years; • women are three times more likely than men to report lost wages; • self-employed caregivers are more likely to work part time or miss days of work while private sector employees are more likely to leave their job to meet care demands; • annual aggregate lost wages due to care-related employment disruptions increased from $207 million to $359 million over the last 10 years; and • economic costs generally increase in later life when earning power is highest.

Some households can’t easily absorb even the modest losses associated with absenteeism and reduced work hours. Those who leave their jobs entirely to accommodate eldercare face much more severe financial consequences that far exceed the maximum Compassionate Care Benefit, a program intended to mitigate employment consequences of care.

Child Care: Preferences and Opportunity Costs

Roderic Beaujot, Zenaida Ravanera, and Ching Du, University of Western Ontario, London, Ontario

Preferences and constraints are useful in the analysis of childbearing and child care. The preferences and constraints lead to alternative models of earning and caring, or the division of paid and unpaid work in couples. The paper uses various data sources including the 2001 Census, the 2006 General Social Survey of Families, and the National Longitudinal Survey of Children and Youth. Studied first are work patterns, in particular, the extent of paid work (full-time, part-time, not employed) by gender, parental status (including age of youngest child in the household), and marital status (married/cohabiting, separated/divorced/widowed, never married). Work preferences of employed persons are then studied in terms of the desire to work “more hours for more pay, less hours for less pay, or the same hours for the same pay” tabulated by gender, presence of children, presence of spouse and current work status (full-time, part-time). Child care is further studied in terms of the proportions of two-parent and oneparent families using various care facilities for children aged 0-5, for Quebec and the rest of Canada. The paper ends with a discussion of policy alternatives that would support parental preferences and reduce opportunity costs, given the diversity of family structures and of models of earning and caring.

Earnings and Registered Pension Plan Participation

Alex Grey, Human Resources and Skills Development Canada, Gatineau, Québec

Research has tended to focus on the difficult empirical issue of the extent to which there is a trade-off between wages and pension benefits. With the decline in registered pension plan (RPP) coverage, the issue of earning differences between workers with and without an RPP has become more relevant. A key issue is whether the shift reflects a change in the composition of compensation or whether it reflects a decline in labour productivity. The framework proposed by Smith and Ehrenberg (1983) to explain the mix of compensation between wages and pensions is used to develop hypotheses. The analysis is conducted using cross-sectional comparisons of the earnings of RPP members and non-members from the Survey of Labour and Income Dynamics (SLID) for 1996, 2001 and 2006. Results for 2006 indicate that non-RPP members earned on average 15% less than RPP members, controlling for a set of personal and job characteristics, consistent with idea that jobs without RPPs have lower labour productivity on average. This gap is significantly less for younger workers. Over 1996-2006, a period model does not indicate any change in the relative earnings disadvantage of all those without an RPP. However, cohort models (both age and tenure-based) indicate an increase in the relative earnings of some younger cohorts/more recent hires without an RPP. This is confirmed by separate regressions for those with tenure of five years or less, where both a period model and an age-based cohort model indicate an improvement over time in the relative earnings of those without an RPP. These developments are consistent with a shift among younger workers/recent hires in the compensation package away from RPPs reflecting either a change in either the preferences of workers or in the costs of pensions to firms. One implication is that a continued rise in the share of jobs without RPPs over time would be associated with a decline in overall employment compensation, holding other characteristics constant. The relative earnings of individuals without an RPP in younger cohorts have been improving, a development that may persist as these cohorts age.

Creating a Typology of Recent Pension Policies Across OECD Countries

Mehmet Fatih Aysan, University of Western Ontario, London, Ontario

The substantial declines in mortality and fertility have resulted in a gradual increase in the ratio of older to younger people in the population. This demographic transformation has brought pressure for reforms in pension policies, particularly as the large baby boom cohorts begin to exit the labour market. Many welfare states have been trying to achieve the sustainability of the pension schemes by pursuing pension reforms, such as changing the benefit rates, raising the minimum retirement ages, increasing the contribution period, and promoting private schemes.

In this study, our main purpose is to evaluate the recent pension policies in the context of family and work policies relating to younger generations. Even though population ageing and retirement policies have similar characteristics across OECD countries, different welfare states may respond differently. By using recent OECD, EUROSTAT, we aim to analyze differential policies in 21 OECD countries and to place them into an empirical typology that helps to understand the similarities and differences across OECD countries, including the impact of pension policies on intergenerational social justice across OECD countries.

Early Career Difficulties for Visible Minority Postsecondary Graduates in Canada: A Longitudinal Approach

Pierre Canisius Kamanzi and Jake Murdoch, Université de Montréal, Montréal, Quebec; Jacques Ledent, Urbanisation, Culture et Société- Institut national de la recherche scientifique (INRS), Montréal, Quebec

Visible minorities (immigrant and non-immigrant) hold more frequently jobs that are socially less valued and receive lower income than other ethnic groups (Palameta, 2004, 2007). Few studies have looked at possible difficulties of visible minorities in finding employment or being considered for employment even if they have Canadian qualifications. Using a longitudinal approach, we examine whether visible minority postsecondary graduates have equal chances of finding employment over time compared to other groups. We use data from the Statistics Canada’s 1995 and 2000 National Graduates Surveys which enable us to follow the graduates up to five years after graduation. Our results show visible minority graduates (particularly Black, Latin American and Arab) experience more unemployment during these first five years, even with equivalent levels and fields of education. 14 For the former, these employment difficulties remain even five years after graduation, even if they are reduced. There appears to be little difference in general between the 1995 and the 2000 cohorts. These results lead us to question the role of human capital in accessing employment and to underline the importance social and network capital. We discuss the possible policies required in order to improve the equity of access to employment and to reduce also possible ethnic discrimination.

U-shaped Assimilation or Economic Entrenchment? Factors Affecting Occupational Status Recovery for Highly Skilled Immigrants in Canada

Laura J. Templeton, University of Alberta, Edmonton, Alberta

Does a U-shaped assimilation curve best describe the occupational status recovery of highly skilled immigrants in Canada? Using data from the Longitudinal Survey of Immigrants to Canada as well as the Boyd-NP Occupational Status Scale, this question is investigated by comparing the occupational status of a migrant’s pre-migration job with the status of job(s) held in Canada. Controlling for a host of human capital variables, survival analysis is used to calculate the likelihood of occupational status recovery for white men, non-white men, white women and non-white women for each month since arrival in Canada. Overall, study results fail to demonstrate immigrant recovery according to a classic U-shaped trajectory. Rather, each group appears to plateau following initial gains in post migration occupational status, a trajectory more fitting with an argument that immigrants are economically entrenched. Although the status scores of white and non-white men and women cluster closely prior to migration, women and non-whites emerge as particularly disadvantaged post-migration as witnessed by their greater post-migration declines in status as well as their more modest rates of improvement during their first four years of settlement in Canada.

Future Older People in the Most Vulnerable Situation in Terms of Health, Family Situation and Living Arrangement: A Comparison of Canada with Some European Countries.

Jacques Légaré, Yann Décarie, Patrick Charbonneau, Janice Keefe, Joëlle Gaymu and Équipe FELICIE, Université de Montréal, Montréal, Québec

With advancing age, the probabilities of having health problems and living without a spouse increase very rapidly and unequally based on gender. This situation places children in the position of primary helpers, and the failure of this support system automatically entails a greater need for official assistance or recourse to institutionalization. The purpose of this paper is to analyze future changes in this minimal family network for individuals with a poor health who live alone – one of society’s most vulnerable groups. A comparison between Canada and some European countries will enable us to determine how their specific demographic history will impact differently on both changes in the numbers of this population, dealing with numerous disadvantages, and on the risks of experiencing these disadvantages. The study is based on projections that have been developed to 2030, and will deal specifically with the population aged 75 and older.

Policy and Fertility: An Empirical Study of Chilbearing in Canada

Benoît Laplante and Jean-Dominique Morency, Urbanisation, Culture et Société- Institut national de la recherche scientifique (INRS), Montréal, Québec

Using frameworks developed by G. Esping-Andersen and U. Beck, we study the influence of the resources individuals get from the labour market and of policies, and especially the role of uncertainty and costs, on couples’ fertility decisions. We focus our analysis on couples in which both partners are employed before the birth of their first child. We use data from SLID two most recent panels. We use the subsample of couples whose two partners live in the same household and where the woman is as a longitudinal individual. Effects on the hazard of the first, second and third births are estimated using the Cox model. Results show that the effects of the factors vary according to birth order. The decision relative to the first child is related mainly to the level of insecurity faced by the woman: having a permanent job and expected maternity benefits increase the hazard of having the first child. The decision relative to the second child is related mainly to the level of security brought by the man through his income. The decision relative to the third child is not related to the factors we are interested in. Discussion bears mainly on policy issues.

Reconfiguring Intergenerational Relations and Exchanges: Policy Challenges to the Production/Protection Nexus

Susan A. McDaniel, University of Lethbridge, Lethbridge, Alberta

Intergenerational relations and exchanges are the essence of societal reproduction, continuity and interaction. Globalization and population aging raise questions about how policy regimes coordinate with markets and families in both production and protection. Policy discussions about the production/protection nexus could benefit from clearer understanding of the nature, direction and determinants of shifting intergenerational relations and exchanges. In this paper, contrasting trends from across the world on co-residence, on type and direction of transfers in both family and public contexts are examined. Relying on the concept of intergenerational interlinkages (IGILS), to capture the dynamic and interactive set of social and economic processes occurring among generations, the paper focuses particularly on a comparison of patterns in Canada in international context. Because data on intergenerational transfers and exchanges are scarse, we rely on multiple sources including OASIS (Old Age and Autonomy: The Role of Service Systems and Intergenerational Solidarity), Oxford Global Ageing Survey, the Generations and Genders Survey, as well as various Statistics Canada data. Our conclusions are that intergenerational interlinkages are much more complex in nature, direction and determinants than previously acknowledged in research or policy.

What is the Contribution of Early Life Circumstances, Adult Circumstances and Adult Health Behaviour to Educational Differences in Major Depressive Episode During Young Adulthood in Men and Women?

Alison Park, Amélie Quesnel-Vallée and Rebecca Fuhrer, McGill University, Montréal, Québec

Using a life course perspective, this paper examines the contributions of early life circumstances, and adult circumstances and health behaviours, to educational inequalities in Major Depressive Episode (MDE) in men and women. Data come from a subsample of the longitudinal National Population Health Survey (NPHS), a nationally representative sample (n=769) of Canadians that were between age 12 and 19 during the first wave of the NPHS (1994). Respondents were followed over 14 years (7 waves) so that individuals were observed in adolescence and adulthood. Data analysis was stratified by gender and life-stage, defined by 1) ages 19 to 23, and 2) ages 24 and older. Between age 19 and 23, females with less than highschool experienced more MDE compared to those with higher education; this effect was amplified when controlling for early life factors. There was a protective effect of lower education in males between ages 19 and 23, and a negative effect of lower education after age 24. Early life circumstances explained most of the educational effect in males 19 to 23, and adult circumstances explained most of the education effect in older males. Results suggest that educational inequalities in MDE differ both by gender and life-stage.

Cost-related Prescription Drug Non-adherence Across Canada

David Haardt, Dalhousie University, Halifax, Nova Scotia

It is well-known that prescription drug coverage varies substantially across Canada. We also know that lack of coverage, or partial coverage, can leadto adverse health effects. However, while we know who has and who does not have coverage, surprisingly little is known about who purchases fewer drugs than prescribed. This is because most surveys do not contain information on actual drug purchases compared to what has been prescribed. Previous research which did have such information lacked in terms of socio-demographic and regional information. I use the 2001 Participation and Activity Limitation Survey (PALS) which includes information on cost-related non-adherence as well as detailed socio-demographic and provincial data. In particular, I aim to answer the following questions: 1. Who is at particular risk of cost-related non-adherence? 2. Are there differences between provinces? Are they large enough to matter? 3. Are the same people at risk in each province, or are the risk groups different across the country? 4. How important are coverage differences in explaining differences in cost-related non-adherence? Since drug prescriptions rise and prescription drug coverage changes as people grow older, I pay special attention to those aged between 50 and 64, and to life-cycle changes in non-adherence.

Defining Disability: Recognizing the Heterogeneity of Care Receivers and Its Consequences for Projecting Future Care needs of Canadians

Janice Keefe, Mount Saint Vincent University, Halifax, Nova Scotia; Jacques Légaré and Patrick Charbonneau, Université de Montréal, Montréal, Québec; Yann Décarie, Université de Sherbrooke, Sherbrooke, Québec

Attempts to improve projections of future home care needs revealed the sensitivity of measures and the diversity within groups of care receivers. Specifically, increasing the number of activities taken into account has created a problem when calculating the future total number of hours of assistance per week. We have created three distinct Groups (A, B, C) within 7 activities that take into account the heterogeneity of the population aged 65 and older receiving assistance in the community. While hours of help received by Groups A and C are important to our projections in terms of obtaining an overall amount of hours, we believe they are less critical to the projection of human resources needs for chronic home care providers. We will emphasize Group B membership in our projections. Understanding the characteristics of older persons in Groups A, B and C further demonstrates the continuum of services provided and the need to have multiple policies to meet the needs of the aging population. Source of assistance varies significantly depending on the type of activity. Subsidized home care may fit the needs of Group B but increased civic engagement both in terms of private enterprise (Group A) and informal support (Group C) will also be necessary.

How Can We Measure and Reduce Social Health Inequalities?

Paul Bernard and Marie-France Raynault, Université de Montréal, Montréal, Québec

How should social health inequalities (SHI) be measured in order to justify and evaluate interventions aimed at reducing them? We propose that this strategy begin with Mackenbach’s approach (Kunst and Mackenbach, 1994; Mackenbach and Stronks, 2002), which sets out four “entry points” for intervention to reduce SHIs: 1) reduce socioeconomic disadvantages, 2) reduce the amplifying effects of poor health on socioeconomic disadvantages, 3) reduce the effect of mediating factors (living conditions and lifestyle), and finally, 4) improve disadvantaged individuals’ access to healthcare. We are faced here with a paradox with respect to efforts to reduce SHIs: these interventions are listed in descendind order of effectiveness, but in ascending order of feasibility. We would like to expand Mackenbach’s perspective by joining it to two complementary approaches: life course and the rules of access to resources. With these three sets of concepts, a strategy for measuring SHIs that accomplishes three goals can be proposed. First, it would identifiy the “fundamental causes” (Willson, 2009) of SHIs that translate socioeconomic inequalities (income, wealth, education, occupation) into health disparities, regardless of the proximal causes of the disease. Second, it would focus on health phenomena 37 that have long-term effects on individuals’ life courses, impeding their chances to avoid SHI (for example, low birth weight, chronic stress in early childhood, workplace accidents early in the working life). Lastly, it would identify relevant health resources in local environments and on a societal level, and measure the level of access to these resources which characterizes individuals and families from different socioeconomic levels.

Participants’ Perspectives on the Use of Elder Mediation (EM) As a Support to Enhance Social Participation and Inclusion among Families Coping with Alzheimer’s and Dementia

Judy McCann Beranger, Elder Mediation Canada, Newfoundland; Judy-Lynn Richards, University of Prince Edward Island, Charlottetown, Prince Edward Island

Background: Typically, researchers on aging study social participation/inclusion in terms of seniors who volunteer and are active in their communities. Needed is research that focuses on other social networks, such as families dealing with Alzheimer’s or dementia, where levels of social participation and inclusion of the senior, caregiver, and family members can be low, resulting in poorer quality of care.

Methods: This qualitative study uses iterative, thematic analyses to examine the experiences of 41 Atlantic Canadians who used EM because they have a loved one with Alzheimer’s or dementia and were feeling isolated and unable to cope. Results: The narratives reveal several themes related to social participation and inclusion. For example, EM provides direction for increased family cooperation and collaboration, the maximization of family resources, feelings of honesty, support, "overcoming the odds," and family bonding, all of which enhanced social participation/inclusion and quality of care for the senior. Finally, almost all experienced a prolonged length of time when they desperately needed, but were unaware of, the benefits of EM.

Policy Implications: Policymakers/researchers need to collaborate on identifying supports, such as EM, for long-term care initiatives (awareness campaigns and training) to provide families with a diversity of continuum-of-care services that can enhance quality of care for seniors.

Growth and Decline: Differing Demographic Destinies for Canadian Communities

Kevin McQuillan and Erin Gibbs Van Brunschot, University of Calgary, Calgary, Alberta

The mainlines of Canadian demographic development in recent years is well known: below-replacement fertility, low and still declining mortality, and relatively high levels of immigration have produced a population that is growing slowly and aging. Behind the national picture, however, are a variety of paths being followed by Canadian towns and cities. Using 1996, 2001, and 2006 census data for Canada’s CMAs and CAs, this paper explores the different demographic circumstances of Canadian communities and their likely destinies in the near future. The paper pays particular attention to issues of growth versus decline, the pace of population aging, and the association between age structure and other significant community characteristics. Following William Frey’s analysis of American communities, the paper explores the differences between populations that are “aging in place” versus those that are aging, to a significant degree through migration. The latter group consists of communities that are far more affluent and better prepared to respond to the challenges posed by aging. The paper concludes with a brief discussion of key policy issues that will face national, provincial, and local governments.

Cohort Progress Toward Home Ownership and Household Formation: Immigrant Experiences in the U.S. and Canada Compared

Michael Haan, University of Alberta, Edmonton, Alberta; Zhou Yu, University of Utah, Salt Lake City, Utah

This paper compares residential assimilation among Mexican, Asian Indian, Chinese, and black immigrants in the U.S. (Los Angeles) and Canada (Toronto), using native-born, non-Hispanic whites as a common reference group. We focus on the arrival cohort that came in 1995-99 and analyze the pace of assimilation from 2000 to 2006. The two indicators to be examined are homeownership attainment measured at the household level and household formation at the individual level. Preliminary results show that, although homeownership levels are fairly similar across countries, there are considerable differences in household formation patterns. We interpret this as evidence of groups having a similar desire to own their dwelling in both countries, and that the differential use of household formation represents a household strategy to achieve the desired goal of ownership in the face of different employment prospects, dwelling price constraints, and purchase incentive structures (such as mortgage interest deduction).

Changing Demand for Basic Education and Training in the Workplace

Paul Bélanger, Université de Québec à Montréal, Montréal, Québec

The demand for work-related education and training usually concentrated in higher skilled jobs and among younger workers is undergoing a dual change, first among less-skilled employees, and increasingly, among older workers. Empirical research conducted in two industrial sectors had identified socioeconomic and demographic factors behind these trends, and has found certain obstacles responsible for the delayed response to these new demands.

The Impacts of Changing Levels of Educational Attainment on the Labour Market Outcomes and Employment Income Levels of Youth with Disabilities

Aron Spector, Human Resources and Skills Development Canada, Gatineau, Québec

The last decade has seen a considerable increase in the proportion of younger adults with disabilities who are completing both secondary and postsecondary schooling. Cross-sectional data from both the Survey of Labour and Income Dynamics (SLID) and the Participation and Activity Limitations Surveys indicate that:

  • The proportion of young men with disabilities completing high school and subsequently completing post-secondary certificate and diploma programs is now in line with that of other young men
  • The proportion of young women with disabilities successfully completing university degree programs is fast converging with levels of other young women.

Furthermore, there is strong evidence of a surge in labour force participation rates among youth with disabilities between the ages of 20 and 29. At the same time, we know less about how well the jobs opening up to youth with disabilities mesh with their educational background and compensation levels of employment at hiring and through early career progression. Using SLID longitudinal data from the 1999, 2002 and 2005 waves, this study will focus on these two issues by tracing education/employment start/career progression trajectories for youth with and without disabilities.

Inequality in Workplace Disability Accommodation

Kim M. Shuey and Emily Jovic, University of Western Ontario, London, Ontario

Workplace supports for workers with disabilities are often crucial to a worker’s ability to enter and remain in the labour force. This analysis uses the 2006 Participation and Activity Limitation Survey PUMF to examine workplace accommodation for Canadian workers with disabilities. We examine whether workers with disabilities who possess additional ‘vulnerable’ characteristics associated with position in the labour market (for example, lower levels of education and earnings, lack of union protection) and social structural location (age and gender) are at greater risk of having unmet needs in the workplace. Identifying whether subgroups of workers have a different likelihood of receiving necessary workplace supports allows a better understanding of the extent to which employers may direct their resources toward their more valued workers and away from less privileged ones. Results from logistic regression models predicting unmet need provide evidence of inequality in the receipt of needed workplace supports. Models indicate an increased likelihood of having unmet needs among low wage, less educated, non-union, non-permanent workers, as well as workers in manufacturing occupations. Workers in management and service positions face the lowest level of risk. The majority of these relationships persist in the presence of controls for level of disability.

Life Course and Structural Factors in Childlessness: The Waiting Game and Constrained Choices in the Second Demographic Transition

Zenaida Ravanera and Roderic Beaujot, University of Western Ontario, London, Ontario

Childlessness has increased with the delays in life course transitions and the changing normative context of the second demographic transition. It could be seen as a result of decision making processes that take place within structural and normative contexts. The structural context of one’s life course includes schooling and work domains that have drastically changed, especially for women, over the past decades. This paper presents the Canadian trends in childlessness and uses the 2006 General Social Survey on Family Transitions to study the determinants of childlessness and intention to remain childfree for women aged 30-49. The level of childlessness that was lowest (at around 12% to 13%) for the cohorts of women who gave birth to the “baby boomers” has since increased in subsequent cohorts (to about 20%). Age, work orientation, and marital and work constraints are factors that influence childlessness and the intention to remain childfree. A high proportion of childless women (61% by age 40-49) cite being “too old” as the reason for not wanting to have a child. The need of women to first establish themselves in the economic realm through education and work presents constraints in the achievement of their fertility goals.

Comparative Analysis of the Effects of Caregiving on Hours Worked for Sandwiched Versus Non-sandwiched Caregivers

Arianna Waye and Janet Fast, University of Alberta, Edmonton, Alberta

Caregivers tend to be susceptible to employment consequences as a result of care responsibilities. These include: changes in hours worked; absenteeism; labour force exits; and restricted availability in work schedules. Attention has been drawn to the sandwich generation, those caring for both an older and a younger generation, because of their unique circumstance of caring for multiple generations with differing, and even competing, needs. However, there is little evidence as to whether sandwiched caregivers experience more significant employment consequences than do non-sandwiched caregivers. Using individual time use data, GSS Cycle 15, this paper examines how sandwiched, child, and elder caregivers, are differentially impacted in terms of time spent in paid work. The implications of the findings from this research are of importance because the sandwich generation is expected to grow in the future as the population is aging, and individuals are increasingly delaying marriage and childbearing. In addition, women are increasingly committed to the workforce. Evidence of the impact of multigenerational care responsibilities on employment can help inform future public policy and community services to support the growing number of employed sandwiched caregivers.

Adequacy of Canadian Women's Financial Resources of Retirement

Mika Kawaguchi and Karen A. Duncan, University of Manitoba, Winnipeg, Manitoba

A sample of 2,435 pre-retirement women from Statistics Canada’s 2007 General Social Survey was used to explore who among pre-retirement Canadian women aged 45 to 64 were more likely to believe their financial resources for retirement are adequate, and whether selection or availability of financial advice and information affects perceived adequacy of financial resources for retirement. The effect of sources of financial advice and information was examined controlling for education, income, marital status, presence of children, subjective health, and immigrant status. Descriptive statistics and logistic regression analysis were conducted. The results of logistic regression indicated that the level of income, being in a married or common-law relationship, having a better state of subjective health, and being born in Canada, were positively associated with women’s perceived adequacy of financial resources for retirement. Retirement planning experts, financial institution employees, accountants, partners, and employers were the key sources of financial advice and information that increased perceived adequacy of financial resources for retirement. The results of this research can be used to better understand who among pre-retirement women are more or less likely to perceive their financial resources for retirement as adequate.

Retirements by Industry, in Canada, 1993 to 2004

Patrick Charbonneau, Université de Montréal, Montréal, Quebec

The aging of the Canadian population, combined with the scope of the numerical imbalance between the baby boom and the subsequent generations, will inevitably continue to impact the labour market dynamics. In addition, baby boomers are only beginning to reach retirement age, and will continue to do so for another fifteen years (Légaré, 2003). As a result, the median age of the labour force is constantly rising, and retirements should become more prevalent over time (HRSDC 2005; Sunter, 2001). It is therefore essential to address the issue of retirements. This poster will shed new light on this matter by studying recent retirement trends by industry in which workers leaving the job market in Canada were employed. Using the Survey on Labour and Income Dynamics, we will first attempt to provide an overall portrait of the workforce in terms of the socio-demographic composition of the workers in each selected industry. The variation in retirement rates in recent years by industry will then be presented. Three panels of workers will be considered in order to track changes in worker behaviour over time.

Retirements Based on Occupation in Canada, 1993 to 2004

Marc-André Fortin, Université de Montréal, Montréal, Québec

The aging of the Canadian population, combined with the scope of the numerical imbalance between workers from the baby boom generation and those from subsequent generations, should lead to a levelling off, followed by a decrease in the relative share of the working age population over the next two decades. Essentially, the decline will be the result of massive and imminent retirement of baby boomers. These departures could affect certain types of occupations more than others. This poster presents the preliminary findings of an analysis of the recent evolution in retirements based on occupation in Canada using data from Statistics Canada’s Survey of Labour and Income Dynamics (SLID). It will thus be possible to track changes in retirement rates of three panels of workers between 1993 and 2004. Using these data, a longitudinal analysis of retirement can be undertaken based on selected occupational groups, and the principal explanatory factors can be determined by highlighting in particular the relative importance of age structure for each of these groups.