Retaining older workers
René Morissette, Grant Schellenberg and Cynthia Silver
Given the growing number of people nearing retirement, concerns about the social and economic consequences of a mass exit from the workforce have spurred interest in increasing the labour force participation of older workers.1 A key issue is how amenable older workers would be to employer strategies and public policies designed to encourage them to remain on the job.
Possibly some older workers would retire later if offered flexible work arrangements, such as part-time hours or fewer annual workweeks. Others might postpone retirement if pension income were not affected or if they were offered salary increases. Naturally, some would be unable to continue working because of health problems. This article uses the 2002 General Social Survey to explore these issues for some 1.8 million individuals who retired between 1992 and 2002 (see Data source).
Incentives would have kept some retirees in the workforce
The 2002 GSS asked retired respondents what factors might have influenced them to continue working (Table 1). Over one-quarter indicated they might have changed their decision to retire if they had been able to reduce their work schedule without their pension being affected, either by working fewer days (28%) or shorter days (26%). In addition, just under one-fifth would have been influenced by more vacation leave. Altogether, 31% cited at least one of these three pension-related reasons.2 The importance of work arrangements is also shown by the 28% who would have continued working on a part-time basis.
Health problems were a consideration for many recent retirees; 27% said they would have continued working if their health had been better. Just over 21% would have continued working if their salary had been increased, although by how much was not asked. Fewer would have continued working if mandatory retirement policies had not existed (12%) or if they could have found suitable caregiving arrangements (6%). In all cases, men and women responded very similarly.
Overall, 60% of recent retirees indicated a willingness to continue working if certain incentives had existed. Undoubtedly, this overstates the extent to which older workers constitute a potential supply of labour since the capacity of some to remain in the workforce was limited by health problems. It is therefore important to examine the combination of factors that would have enabled or encouraged them to continue working.
One-third of recent retirees retired for health reasons (Table 2). The percentages are higher for self-employed individuals (40%) than for employees (31%), likely reflecting older retirement ages of self-employed workers. Since alternative work arrangements and retirement policies would likely not affect the retirement decision of these individuals, they are excluded from the discussion of older workers as a potential supply of labour.
Another third of recent retirees did not retire for health reasons and would not have continued working for any of the reasons offered. The remaining third—healthy individuals who would have been willing to remain in the workforce (at least partly)—clearly offer the best prospect for increasing the overall supply of labour.
About a quarter of these healthy individuals, representing 8% of all recent retirees, would have continued working if alternative working arrangements had been available.3 Almost half (15% of recent retirees) would have kept working in light of such arrangements combined with other factors. Within this group, the other factors most frequently cited were salary increase (80%), no mandatory retirement policy (35%), suitable caregiving arrangements (8%), and other reasons (23%). The remaining quarter (8% of all recent retirees) would have continued working for reasons other than working arrangements. Within this group, the most frequently cited were 'other reasons' (58%), salary increase (24%), and no mandatory retirement (24%).
Profile of those willing to continue working
While the above paints a broad picture of preferences regarding retirement, it leaves several questions unanswered. First, who would be most likely to keep working if working arrangements were the only consideration? For instance, flexible work arrangements might not influence the retirement decision of employees working at physically demanding jobs for an extended period of time. On the other hand, those whose financial position has deteriorated after retirement or who are having a hard time finding new activities might be willing to reconsider their decision.
Second, who would continue working only if other factors were changed in addition to work arrangements? Third, who would be unlikely to be swayed even in these circumstances? Would former employees in goods-producing industries react differently than their counterparts in services? Would education level play a role?
One-half of all former employees who did not retire for health-related reasons said they would have kept working if alternatives had been offered4 (Table 3). Twelve percent cited alternative working arrangements; almost a quarter indicated both working arrangements and other factors; and the remaining 13%, other factors only.
These simple averages mask important differences between groups. For instance, 28% of those with a high school diploma or university degree would have kept working if both working arrangements and other factors had been altered, compared with 19% of those with no high school diploma. The same combination of changes was indicated by one-third of those who had been offered early retirement incentives, compared with one-fifth of those who had had no early incentive (Table 4).
From an industry perspective, no more than 20% of individuals formerly employed in health care, social assistance and education; or accommodation and food services would have kept working even if both working arrangements and other factors had been altered (Table 4). The extent to which job quality discourages older workers from remaining in the workforce cannot be addressed with the GSS. Overall, if alternatives had been offered, about 60% of individuals formerly employed in utilities, transportation and warehousing; trade; and information, culture and recreation would have continued to work, compared with about 45% in construction; health care, social assistance and education; and accommodation and food industries. One reason for the lower rate among construction workers may be that physically demanding jobs are unattractive to older workers.
Do these qualitative patterns hold when healthy retirees with similar characteristics are compared? A multivariate analysis was used to examine how answers varied according to age at retirement, sex, education level (own and spouse's), occupation and industry of prior employment, and several other characteristics.5 The analysis was limited to those who were employees prior to retirement. 6
Compared with their counterparts aged 60 to 64, retirees aged 50 to 59 were more likely to report that they would have continued working. If only alternative working arrangements are considered, 14% would have stayed, slightly more than the 12% observed for those aged 60 to 64 (Table 5).7 Those who retired at 65 were much more likely than their younger counterparts to have been willing to continue working if, in addition to working arrangements, other factors such as mandatory retirement policies had been altered.
Retirees with a university degree were among the most likely to have continued working under different working arrangements (with or without other factors). This may be attributable to their relatively high levels of job satisfaction or less physically demanding jobs. Alternative work arrangements appear to be an important consideration for employers keen on retaining highly educated workers.
Immigrants and retirees who received early retirement incentives were much more likely to have considered continuing to work given other factors in addition to alternative working arrangements. Retirees formerly employed in health care, social assistance and education were the least likely to report preferences for continuing to work. This suggests less scope for retaining older workers in these industries—a consideration that takes on added importance given their disproportionately large shares of employees approaching retirement (Statistics Canada 2004).
Does a worsening financial situation in retirement affect one's view of continuing to work? Unambiguously, the answer is yes. Among comparable retirees, those whose financial situation had deteriorated since retirement were much more likely to wish they had been offered alternative working arrangements. This may reflect unexpected declines in living standards after retirement. Poor knowledge of one's employer-sponsored pension plan is an important consideration in this respect (Morissette and Zhang 2004).
Admittedly, retrospective questions about retiring must be treated cautiously since it is impossible to determine if a different course of action would have been taken. Responses may overstate the willingness of individuals at the time to continue working, particularly if they have found their retirement to be less satisfying than expected. In retrospect, continued employment may look appealing. Conversely, even those retirees who said they would not have been willing to continue working might have done so if offered a job with enough pay and the right conditions.
Despite such limitations, the findings offer some insight for the future. Alternative working arrangements appear to be an important consideration in encouraging older workers to remain in the workforce. Over one-quarter of retirees in the sample would have been willing to continue working if part-time employment had been available. Similarly, a significant proportion said that continued employment would have been an attractive option if they had been able to work fewer hours without their pension being affected. The importance of working arrangements is also evident in the 42% who returned to the workforce on a part-time basis.
However, the circumstances and conditions that shaped the experiences of retirees in the 1990s may be quite different from those in the years ahead. In a context of tighter labour markets, it is unlikely that organizations will as readily offer early retirement incentives. Indeed, in the public sector, spending on such incentives reached a peak in 1996, declining through the rest of the decade (Kieran 2001). The opportunities open to individuals facing retirement may change in the near future, as may the extent to which they are willing to remain in the workforce.
The 2002 General Social Survey (GSS) targeted all persons 45 and older residing in the 10 provinces except full-time residents of institutions.
The GSS used a subjective definition of retirement. First, individuals who reported their main activity during the previous 12 months as 'retired' were identified as retirees. Individuals with another main activity were asked "Have you ever retired?" Those who said yes were also identified as retirees, even if they had since returned to the workforce. Individuals who had never retired were asked a follow-up question that probed further:
Retirement does not necessarily mean stopping work permanently. Have you ever retired in any of the following circumstances?
- You became eligible for a pension or put in enough years for a pension.
- You received an early retirement package.
- You significantly reduced the amount of work you did for a business or farm you operated in order to retire.
- You were permanently laid off or lost a job and did not look for work or gave up looking for work in order to retire.
- You retired from a job or significantly reduced your work time because of your health.
- You retired from a job or significantly reduced your work time because you could afford to live on your savings/investments.
- You retired from a job or significantly reduced your work time because you could afford to live on your spouse's/partner's investment or retirement income.
- You reduced your work time because of the health of your spouse or relative.
Those who responded yes to any part of the follow-up question were also classified as retirees.
The analysis is limited to recent retirees—those who (first) retired between 1992 and 2002. This was done to focus on the characteristics and experiences of individuals who made the transition into retirement in recent years. The sample was also restricted to individuals 50 or older, resulting in the exclusion of a few respondents who retired earlier. The final sample was 4,464.
- This concern has been clearly expressed by the Organisation for Economic Co-operation and Development: "Population ageing means that, in the absence of any change in patterns of labour market participation, the labour force is likely to fall in relative, and even in a few countries perhaps in absolute, terms over the coming decades with major consequences for economic growth, public finance and living standards. This is why raising the employment rate for older workers is so critical." (OECD 2002, 10).
- A considerable number of retirees who did not receive income from employer pension plans in 2002 responded to these questions, perhaps because they understood such plans to include the Canada and Quebec Pension Plans. When the analysis is limited to individuals who received income from an employer pension plan, the proportion who would have continued working if they had been able to reduce their time at work without their pension being affected rises to 37%.
- Alternative working arrangements include fewer days without pension being affected, shorter days without pension being affected, vacation leave increased without pension being affected, and part-time work.
- The corresponding percentage for former self-employed individuals is only 33%. This no doubt reflects the greater flexibility that self-employment offers in terms of working arrangements and autonomy.
- The other control variables were owning a house; received an early retirement incentive; receiving pension income; employed full year, full-time prior to retirement; no change, deterioration, or improvement in financial position since retirement; and enjoyment of life as much, more, or less since retirement.
- The one-third of recent retirees who left the labour force for health reasons are excluded from the multivariate analysis. The one-third of recent retirees who did not leave the workforce because of health limitations and who said they would not have continued working even if circumstances had been different is used as a comparison benchmark. The multivariate analysis is based on a multinomial logit model and essentially compares the three 'would have stayed' groups in Table 2 with the 'would not have stayed' group.
- The probabilities shown in Table 4 are obtained by setting the other covariates to their mean values.
- Kieran, Patrick. 2001. "Early retirement trends." Perspectives on Labour and Income 2, no. 9. September 2001 online edition.
- Morissette, René and Xuelin Zhang. 2004. "Retirement plan awareness." Perspectives on Labour and Income 5, no. 1. January 2004 online edition.
- Organisation for Economic Co-operation and Development (OECD). 2002. Editorial. OECD Employment Outlook 2002: 7-12.
- Statistics Canada. 2004. "The near-retirement rate." Perspectives on Labour and Income 5, no. 2, February 2004 online edition.
Full article in PDF
René Morissette is with the Business and Labour Market Analysis Division. He can be reached at (613) 951-3608. Grant Schellenberg is with Demography Division. He can be reached at (613) 951-9580. Cynthia Silver is with the Housing, Family and Social Statistics Division. She can be reached at (613) 951-210. All the authors can be reached at email@example.com.
You need to use the free Adobe Reader to view PDF documents. To view (open) these files, simply click on the link. To download (save) them, right-click on the link. Note that if you are using Internet Explorer or AOL, PDF documents sometimes do not open properly. See Troubleshooting PDFs. PDF documents may not be accessible by some devices. For more information, visit the Adobe website or contact us for assistance.