Factors that Influence Retirement

by Derek Aldridge

This article first appeared in the Autumn 2011 issue of the Expert Witness.

We often note in our reports that a plaintiff’s loss of income can be sensitive to assumptions regarding his retirement age. (The retirement age assumption will generally be more important for a plaintiff who is close to retirement than for one who is still many years away.) In addition, it is also sometimes reasonable to incorporate early retirement scenarios, in which we assume that due to his residual deficits, a plaintiff will retire earlier than he would have if uninjured. This can have a substantial impact on the plaintiff’s future loss of income, especially if he is already close to retirement age.

The reason that an early with-accident retirement scenario is often sensible is because a person’s health is an important factor in his decision to retire. That is, poor health tends to hasten retirement, all else being equal. Of course, there are other factors that also influence a person’s decision to retire (financial considerations being most important, but also job satisfaction, spousal influences, and so forth), but health is an important factor, especially with respect to unplanned retirement.

Two articles that examine health impacts on retirement were recently published in Statistics Canada’s Perspectives on Labour and income. The first article (Heath factors and early retirement among older workers) used 12 years of longitudinal data1 to examine the various health factors that caused people to retire early. The second article (Retirement, health and employment among those 55 plus) used a cross-sectional survey2 of Canadians age 55 and over, to examine the reasons why they were either working (and not previously retired), partially retired, fully-retired, or previously retired but now working. Below, I will discuss some of the findings in these articles that are relevant to estimating a person’s loss of income due to an injury.

Health factors and early retirement among older workers (Park, 2010)

The first article relied on data from a group of workers who were age 40–52 in 1994/95. The survey interviewed this group every two years, through 2006/07, by which time they had reached age 52–64. By the end of the 12-year period, approximately 20 percent of the workers had left the labour force. However, among those who reported that they had fair or poor health, 35 percent had stopped working. Among those workers in poor health who stopped working, almost half reported that illness or disability was the main reason they had exited the labour force.

More specifically, the article found that when compared with men who reported that they had excellent health, men who reported fair or poor health were nearly five times more likely to exit the labour force. The researchers found a weaker health effect among women, and this effect was judged to be not statistically significant.3 The stronger effect among men was indeed found to be statistically significant. Among men, it was found that eye problems, back pain, ulcers, and migraine headaches were especially likely to increase the likelihood of retirement.

The researchers also examined health behaviours, and found that among men, daily smoking and heavy drinking also increased the chance of labour force exit (relative to males who did not practice these behaviours). In particular, male heavy drinkers were almost twice as likely to exit the labour force early, versus those who did not drink heavily. Among women, the one health behaviour that was judged to be statistically significant was obesity. Obese women were found to be 1.6 times more likely to exit the labour force early.

The article reported that if health status was controlled for, the effects of smoking and obesity disappeared. That is, smoking and obesity directly impact health, which in turn impacts retirement. Thus, the link between smoking/obesity and retirement is an indirect one. However, heavy drinking had a direct influence on retirement, even when overall health was controlled for. That is, heavy drinking was observed to be associated with early retirement among men, independent of the impact of drinking on the men’s health.

Although it is not especially relevant in the context of personal injury claims, the article also noted that working conditions had a significant impact on the likelihood of labour force exit. In particular, high job strain (stress), job dissatisfaction, and low supervisor support were all found to have statistically significant effects on retirement behaviour.

Retirement, health and employment among those 55 plus (Park, 2011)

The second article relied on a survey of a group of “older” Canadians, aged 55–85 in 2009. The survey group was divided into four subgroups: those who had never retired, those who were partially retired, those who were fully-retired, and those who had previously retired but had returned to work. The characteristics of the four groups were examined, in an effort to determine which characteristics were associated with a person’s retirement status.

The study reported some unsurprising results (for example, women were more likely to be retired than men; most of the workers who had never retired were under age 65, and so forth), and then addressed some issues that are much more interesting from our perspective.

Even when the age differences were accounted for, the study found that the retirees had worse health than those who were not retired. More than half of the retirees reportedly had three or more chronic conditions4, while more than a fifth had five or more chronic conditions. This effect was more pronounced among women than men.

The study also considered physical inactivity and found that retirees were much more likely to be considered physically inactive than those who were working.

Another factor considered by the study that is interesting from a personal injury perspective is financial status. The study reported that nearly 40 percent of the “never retired” workers reported that they had inadequate financial plans for retirement. More specifically, more than a third of them reported that they had less than $25,000 in savings and investments. In addition, among the groups who had retired at least once, the most common reason they reported for their decision to retire was that it was financially possible. These findings reinforce the notion that (especially for older plaintiffs), a person’s financial ability to retire should be considered when estimating when that person would have retired, but for the accident.


The two articles discussed above provide additional support for our common practice of assuming that a plaintiff with substantial residual deficits will likely retire earlier than she would have in the absence of the accident. The articles also identified some factors to be considered when making an assumption concerning a plaintiff’s without-accident retirement age. However, while we know that poor health tends to hasten retirement, we will continue to face the difficulty of knowing the specific effect on a particular plaintiff. That is, for most cases it will remain unclear whether it is more reasonable to assume that a plaintiff’s injuries will cause her to retire (say) three years early or two years. We are left with offering assumptions that are reasonable, providing multiple scenarios, and ultimately leaving it up to the Court to make a specific finding.


1. Park, Jungwee. 2010. “Health factors and early retirement among older workers.” Perspectives on Labour and Income. June 2010. Statistics Canada Catalogue no. 75-001-X. p. 5–13.

2. Park, Jungwee. 2011. “Retirement, health and employment among those 55 plus.” Perspectives on Labour and Income. January 2011. Statistics Canada Catalogue no. 75-001-X. p. 1–12.


1. Longitudinal data is collected over a period of time from the same set of people. [back to text]

2. Cross-sectional data is collected from a group of people at one point in time. [back to text]

3. By “not statistically significant”, I mean that the researchers believed that it might have been random chance that caused it to appear as though women with poor health were more likely to exit the labour force than women with excellent health. [back to text]

4. The number of chronic conditions was calculated based on respondents’ answers to questions about whether they had been diagnosed by professionals as having any of the following chronic conditions: asthma, arthritis, osteoporosis, high blood pressure, back problems, migraine headaches, chronic bronchitis, emphysema, chronic obstructive pulmonary disorder (COPD), diabetes, stroke, heart disease, cancer, stomach or intestinal ulcers, urinary incontinence, Alzheimer’s disease or other dementia, bowel disorder/Crohn’s disease or colitis, Parkinson’s disease, thyroid conditions, cataracts, glaucoma, mood disorders, and anxiety disorders. [back to text]


Derek Aldridge is a consultant with Economica and has a Master of Arts degree (in economics) from the University of Victoria.