Retirement patterns are changing in the U.S. and in many other countries as well. During the mid-to-late 20th century, labor force participation rates dropped for older workers and rose for younger ones. These trends have recently reversed, especially among men and younger workers.

The reversal is expected to continue, according to Towers Watson’s 2013/2014 Global Benefit Attitudes Survey (GBAS). Increasing numbers of employees are postponing retirement, many driven by financial necessity. And many of the employees putting off retirement tend to fit a certain profile: less healthy, highly stressed and disengaged from their jobs.

Plan type affects retirement timing as well, with defined benefit (DB) plan participants retiring earlier than employees with only defined contribution (DC) plans, even where counter influences exist. These findings have implications for employers as they provide some indications of which employees are likely to delay retirement and why.

Shifts in career timing

Over the last two decades, American workers, particularly men, have been entering the workforce later and staying there longer than earlier generations, according to administrative labor force data (Figure 1). Between 1950 and 1990, labor force participation rates fell among older workers and rose among younger ones. Earlier retirement among older workers was driven by the large pool of younger baby boomers entering the labor market over this period. Social Security and many employer-sponsored retirement programs were designed with generous early-retirement subsidies and other features to encourage earlier retirement.

The movement to earlier retirement was unique to men, and the trajectory of women’s careers is still unfolding. Workforce participation rates have jumped for women of all ages since the 1950s, more than doubling for women ages 55 to 64 and nearly doubling for those 65 and older.

Figure 1. U.S. labor force participation rates by age and gender
Click Image to Enlarge

Figure 1. U.S. labor force participation rates by age and gender

Source: The 1950 rates are from 1950 Census of the Population, Vol. IV, Special Reports, Part 1, No. 1A. 1970+ rates are based on Towers Watson tabulations of Current Population Survey’s Annual Social and Economic Supplement surveys.

Recent trends are being driven by both workplace dynamics and adverse economic conditions. Many of today’s jobs are knowledge- or technology-based, enabling workers to continue their careers well into their 60s and 70s. Moreover, many workers who might otherwise retire believe they need to keep working to rebuild savings lost to the 2008 financial crisis. Some retirement delays are prompted by caution, with workers deciding to extend their careers to buffer any future stumbles in the financial markets.

On the other hand, today’s workplace calls for more education, so many young adults are putting off a job in favor of spending more time in school, which typically delays entry into the labor market or results in breaks in labor force participation. While some young people may be deferring a career by choice, however, many younger would-be workers are struggling to get a foot in the door. The sluggish economy has been limiting new opportunities for employment as companies focus more on cost management than expansion. And retirement delays among older workers impede opportunities for the young, as fewer new jobs open up to replace retiring employees. Younger workers are beginning to recognize this — 34% of those under 40 believe that retirement delays among older workers are restricting their career opportunities.

Changing career and retirement patterns are not just a U.S. phenomenon. Labor force participation rates among older men have jumped in many developed economies (Figure 2). This trend has not spread to developing nations, as many of them already maintain higher labor force participation rates than their developed counterparts.  

Figure 2. Global labor force participation rates of men 55+ for selected years of participation in GBAS
Figure 2. Global labor force participation rates of men 55+ for selected years of participation in GBAS

*Data were available for 2000 and 2010 only; the lowest labor force participation rates were in 2010.
**Data were available for 2010 and 2012 only and rates were the same in both years.
Source: Organization for Economic Cooperation and Development (OECD), Employment and Labor Market Statistics

Are later retirements here to stay?

The trend of longer working careers is expected to continue, possibly even intensifying, according to recent Towers Watson survey results.

The number of workers planning to delay retirement is growing. Expected retirement delays are reflected in changes to workers’ planned retirement ages and/or retirement dates. In 2009, just over one-third of full-time employees planned to postpone retirement versus 43% in 2013, a nine percentage point jump (Figure 3).

Figure 3. Has the age at which you plan to retire from full-time employment changed over the last three years?
Figure 3. Has the age at which you plan to retire from full-time employment changed over the last three years?

Notes: Based on full-time employees enrolled in a retirement plan. Responses of “Don’t know” have been removed.
Source: 2013/14 Global Benefit Attitudes Survey – U.S.

Conversely, a smaller group (12%) is planning to retire earlier — up from 5% in 2011. Perhaps the recent uptick in equity markets has improved affordability prospects for some. Nearly three-quarters of those delaying retirement are planning a delay of three or more years (Figure 4).

Figure 4. Approximately how much later do you expect to work?
Figure 4. Approximately how much later do you expect to work?

Note: Based on full-time employees enrolled in a retirement plan who plan to work later
Source: 2013/14 Global Benefit Attitudes Survey – U.S.

As a result, expected retirement ages are rising. Fifty percent of employees plan to retire after age 65 — up nine percentage points from 2009 (Figure 5). And almost one-third (31%) either plan on retiring at 70 or older or don’t plan to retire at all. Consequently, we can expect the employment patterns of older workers shown in Figures 1 and 2 to continue for the immediate future.

Figure 5. At what age do you expect to retire from full-time employment?
Figure 5. At what age do you expect to retire from full-time employment?

Note: Based on full-time employees enrolled in a retirement plan
Source: 2013/14 Global Benefit Attitudes Survey – U.S.

Throughout the remainder of this article, we examine employee characteristics, including expected retirement ages and recent changes in expectations, to describe employees’ career plans. It is important to note that workers’ expectations will not necessarily pan out. While greater numbers of workers might expect to wait until 70 (or even older) to retire, previous research suggests that factors such as poor health often force workers to retire earlier than desired.1 However, the trends reported here suggest that the need or desire among workers to extend their working careers beyond traditional ages is strengthening, which will likely perpetuate rising rates in old age labor force participation, even if some don’t quite reach their intended retirement age.    

The role of savings adequacy in retirement timing

A key question is why employees are changing their retirement plans. Do they want to keep working because they enjoy it, or do they have to work longer because they can’t afford to retire, either because they’re living longer or for other reasons? It appears that many workers are delaying retirement because they haven’t saved enough and need to make up for financial gaps (Figure 6)

Figure 6. Savings adequacy and retirement timing
Figure 6. Savings adequacy and retirement timing

Note: Based on full-time employees enrolled in a retirement plan
Source: 2013/14 Global Benefit Attitudes Survey – U.S.

As we showed in earlier Towers Watson research, three in five employees believe they need to save more for retirement, and 83% are not meeting their own savings goals.2 Under-savers are around twice as likely as other workers to plan to delay retirement, presumably to catch up. Additionally, nearly half (44%) of all under-savers plan on a delay of five or more years. Figure 7 shows that employees who believe they aren’t saving enough are three to four times more likely to expect to retire after age 70 than their more financially prepared counterparts. Those who are over-saving are the most likely to plan to retire before age 65.

Figure 7. Savings adequacy and expected retirement age
Figure 7. Savings adequacy and expected retirement age

Note: Based on full-time employees enrolled in a retirement plan
Source: 2013/14 Global Benefit Attitudes Survey – U.S.

Who plans to delay retirement?

Employees planning to delay retirement tend to fit a certain profile (Figure 8). While these employees are most likely older, age has no effect on planned retirement age. Young workers are already thinking about retirement and many expect to work longer than their parents. Employees planning to retire after age 70 include 30% of those younger than 40 and 32% of those 50 or older. Lower-income workers are more likely to both delay retirement and retire at older ages. This makes sense given that many retirement delays are because of savings inadequacy. But Social Security replaces a greater share of income for lower-income workers, so their intentions to retire later could change as they approach retirement and become more aware of the value of their Social Security benefits. 

Figure 8. Profile of those delaying retirement and those who plan on retiring at age 70+ or never
Click Image to Enlarge
Figure 8. Profile of those delaying retirement and those who plan on retiring at age 70+ or never

Note: Based on full-time employees enrolled in a retirement plan
Source: 2013/14 Global Benefit Attitudes Survey – U.S.

Employee engagement is linked to retirement decisions as well. We use Towers Watson’s sustainable engagement measure, which captures employees’ connection to their employer by three factors: (1) being engaged — the extent of employees’ discretionary effort applied to work goals; (2) being enabled — a work environment that provides the support and resources employees need to work efficiently and effectively, and removes barriers to success; and (3) feeling energized — a work experience that promotes social connectedness and feelings of enthusiasm and accomplishment.3

As shown in Figure 8, while one might expect more highly engaged workers to want to continue careers they’re enjoying, disengaged workers are 11 percentage points more likely to delay retirement and 12 percentage points more likely to expect to retire at older ages. This reinforces the supposition that many retirement delays reflect the employee’s need to keep working rather than his or her strong connection to work or the employer. Many of these workers simply can’t afford to retire.

Financial literacy does not significantly affect retirement delays, although employees with high financial literacy are less likely to plan to retire at older ages. Along those lines, employees who lack clear expectations of their retirement plan or awareness about how much they need to save are also more likely to delay retirement (Figure 9). This could be a “just in case” strategy: Being uncertain of how much they need, they decide to err on the side of caution and work longer to save more.

Figure 9. Retirement timing by understanding savings needs and plan benefits
Figure 9. Retirement timing by understanding savings needs and plan benefits

Note: Based on full-time employees enrolled in a retirement plan
Source: 2013/14 Global Benefit Attitudes Survey – U.S.

Aside from retirement savings, health and stress have the strongest links to retirement timing. Poor health adversely affects both workers’ household budgets and their ability to save for retirement. And trends over time show that workers with health issues are increasingly likely to delay retirement compared with those in good health (Figure 10). Health concerns piled on top of financial worries can take an emotional toll, draining employees’ ability to focus on their jobs and be at their best. This raises concerns for employers about productivity in the growing segment of their workforce that is on the job primarily because they can’t afford not to be. 

Figure 10. Prevalence of delaying retirement and retiring at later ages by health status and chronic conditions, 2009 – 2013
Click Image to Enlarge
Figure 8. Profile of those delaying retirement and those who plan on retiring at age 70+ or never

Note: Based on full-time employees enrolled in a retirement plan
Source: 2013/14 Global Benefit Attitudes Survey – U.S.

Plan adequacy and type strongly linked to retirement age

In addition to serving other functions, retirement plans have traditionally been tools to help employers control the retirement timing of their workforces. Defined benefit plans in particular can offer very strong incentives to encourage older workers to retire on a predictable timetable, thus enabling employers to anticipate and develop succession plans.

Regardless of plan type, employees are far less likely to delay retirement if their retirement plan meets their needs (Figure 11), with a 41 percentage point difference between plans that do and don’t meet employees’ needs. Employees who say their plans don’t meet their needs are also 31 percentage points more likely to expect to retire after age 70 or never.

Figure 11. Prevalence of delaying retirement and retiring at older ages by whether retirement plan meets needs
Figure 11. Prevalence of delaying retirement and retiring at older ages by whether retirement plan meets needs

Note: Based on full-time employees enrolled in a retirement plan
Source: 2013/14 Global Benefit Attitudes Survey – U.S.

As shown in Figure 12, plan type has little effect on employees’ decisions to change their planned retirement ages. Regardless of plan type, employees adjust their retirement plans as necessary to their circumstances. However, planned retirement age is strongly linked to plan type. Thirty-nine percent of DB plan members expect to retire before age 65 compared with only 21% of DC-plan-only members. Similarly, a third of workers with only a DC plan expect to retire after age 69 or never, while only two in 10 DB plan members anticipate working beyond age 69.

Figure 12. Prevalence of delaying retirement and retiring at later ages by plan type
Figure 12. Prevalence of delaying retirement and retiring at later ages by plan type

Note: Based on full-time employees enrolled in a retirement plan
Source: 2013/14 Global Benefit Attitudes Survey – U.S.

This pattern has been evident since 2009, but the gap in retirement timing has widened (Figure 13). In 2009, DB plan members were only 4% less likely than DC-plan-only employees to expect to work past 70, but by 2013, the gap had widened to 14 percentage points. This suggests that plan type still offers employers some control over when older employees retire, and that DC-plan-only employees are still struggling to prepare for retirement.

Figure 13. Prevalence of delaying retirement and retiring at later ages by plan type, 2009 – 2013
Figure 13. Prevalence of delaying retirement and retiring at later ages by plan type, 2009 – 2013

Note: Based on full-time employees enrolled in a retirement plan
Source: 2013/14 Global Benefit Attitudes Survey – U.S.

Plan type augments effects of other employee characteristics

Employees with DB plans tend to be more engaged than those with only DC plans, which might also influence retirement decisions. However, plan type works together with engagement levels and health status to affect when employees expect to retire. Among workers age 50 and older, disengaged workers with DB plans anticipate leaving the workforce five years earlier than their counterparts with only DC plans (Figure 14). The focus here is on workers age 50 or older to allow for more accurate measures of engagement and health in the years closer to traditional retirement age.

Figure 14. Expected retirement age by plan type, engagement level and health status, ages 50+
Figure 14. Expected retirement age by plan type, engagement level and health status, ages 50+

*Good health = Excellent/very good health status or no chronic conditions
**Poor health = Fair/poor health status, diabetes or heart problem
Note: Based on full-time employees ages 50+ enrolled in a retirement plan
Source: 2013/14 Global Benefit Attitudes Survey – U.S.

The effect is similar for health status. Employees in poor health with a DB plan expect to retire four years earlier than those in poor health with only a DC plan. This further supports the idea that DB plans can help employers control retirement timing, as well as affect who chooses to continue working, which has important implications for workforce performance.

Multivariate analysis

As shown above, many factors are linked to employees’ expected retirement age. This analysis expands on the previous findings by using multivariate regression to isolate the impact of different factors — whether the plan meets needs, plan type, health status, engagement and stress — while controlling for demographic and socioeconomic characteristics to predict the retirement ages of different groups of workers. 

Engagement, health, stress and plan type influence retirement age independently. The effects of combinations of “plan meets my needs” and plan type with engagement, health and stress on retirement age hold up as well, particularly as workers near retirement (Figure 15). For example, older DB plan members in poor health expect to retire about six years earlier than unhealthy DC plan-only participants. Very similar results are seen when comparing estimates by whether the plan meets employees’ needs or not.

Similarly, disengaged DB plan members anticipate retiring an average of four years earlier than their disengaged DC-only counterparts, and highly stressed workers with DB plans expect to retire five years earlier than those with only DC plans. In fact, the earliest retirement ages shown are for highly stressed employees whose retirement plans meet their needs. This suggests that both financial and factors related to the work experience (e.g., workload, feeling like a valued employee, effectiveness of managers, alignment with business goals) and employees’ ability to cope with this stress as they get older are important drivers of retirement decisions. But overall, this strengthens the case that a more generous and secure retirement plan enables employees to retire at earlier ages, even those who are in poor health, stressed and disengaged.

Figure 15. Predicted retirement ages of various employee profiles, ages 50+
Figure 15. Predicted retirement ages of various employee profiles, ages 50+

Notes: Based on full-time employees ages 50+ enrolled in a retirement plan. Regression results are based on an ordinary least squares (OLS) fully interactive model with plan meets my needs and plan type. Independent variables include age, gender, household income, marital status, self-reported health status (5-point), existence of chronic condition (self or family), stress levels (cluster of eight items), sustainable engagement and plan type. 
Source: 2013/14 Global Benefit Attitudes Survey – U.S.

Conclusion

Recent shifts in career timing are posing challenges for many employers in the U.S. and other developed economies. Older workers are retiring later, while many younger people are struggling to find work or positions that fit their skill sets. In part, older workers’ delayed retirements are blocking younger workers’ entries into the workforce and limiting their opportunities. But a growing number of older employees fear that if they retire “on time,” they won’t be able to afford a comfortable retirement. 

This research shows that retirement adequacy concerns are clearly a driving factor behind employees’ decisions to extend their careers. This is highlighted by the trend of delayed retirement becoming increasingly common among employees who are disengaged, unhealthy and/or stressed, all of which pose productivity challenges for employers. Furthermore, many of those postponing retirement and planning to retire at older ages fear they are not saving enough for a secure retirement. 

As our previous research has shown, employees look to their employer-provided programs as their primary way to accumulate retirement savings. It is important for employers to stay on top of retirement patterns and determine whether their retirement programs are inadvertently encouraging unproductive workers to stay on the job longer. Offering DB plans gives employers considerably more control over who continues to work and who retires “on time” or early. But generous defined contribution arrangements can achieve similar outcomes if employees make good choices throughout their working careers in terms of their contribution amounts and investments.  

Employees need to establish good habits early on, particularly when it comes to saving more, when it has the greatest impact on their long-term financial security. Giving employees access to a range of emerging decision-support tools and education, particularly when tailored to specific population segments and delivered at key life events, can go a long way to helping them make good decisions about complex financial situations. 

In the end, employers should consider rewards in terms of the appropriate mix between security for all workers and incentives that reward the behaviors needed to deliver on the business strategy. Offering employees greater retirement security can prove to be a financial advantage for employers by cultivating a less stressed, healthier and more engaged workforce. Employers who can find the right mix of security and short-term incentives are in the best position to effectively manage their human capital risks/opportunities and get the greatest value out of their programs.

About the survey

Towers Watson’s 2013/2014 Global Benefit Attitudes Survey is a nationally representative survey fielded in 12 countries.4 The U.S. survey includes 5,070 respondents employed by nongovernment organizations with 1,000 or more employees. It builds on several previous Towers Watson surveys to track evolving employee attitudes.

This article reflects responses from a subset of 4,248 retirement plan participants working full time. All respondents are provided a DB and/or DC retirement plan by their current employer. DB plan participants are those who currently participate in an active DB plan. Respondents with only a DC plan include both those who contribute to the plan and those who decline to participate. All results are weighted by age, gender and salary to the national average of similar workers.5


Endnotes

1. Casanova, Maria, “Revisiting the Hump-Shaped Wage Profile,” UCLA Working Paper, August 2013; EBRI’s 2014 Retirement Confidence Survey.

2. See “Workers Still Uneasy About Financial Security and Retirement,” Towers Watson Insider, March 2014.

3. Four employee segments are derived from the three core components of engagement, enablement and energy. Engaged employees score high on all three components. Unsupported workers are high in engagement but lack enablement or energy. Detached employees are enabled or energized but lack engagement, and disengaged employees lack all three components.

4. Countries include Australia, Brazil, Canada, Chile, China, Germany, India, Japan, Mexico, the Netherlands, the U.K. and the U.S.

5. The margin of error is +/- 1.38% for the total sample.