SHRM Foundation Research
Not a Member?  Become One Today!

SHRM Foundation Research


Retirement-eligible but not Retiring: A Longitudinal Study of the Organizational- and Job-related Factors Associated with the Retention of an Older Workforce

Funded: February 2008     Completed: August 2010

Samuel B. Bacharach, Ph.D
., Cornell University

Job, Organizational and Relational Factors Associated with Retirement Upon Eligibility                       

Executive Summary   Estimates put the share of the labor force held by those 55 years and older to be nearly 24% by 2020 (Toosi, 2006).  For over a decade, experts have been warning that in order to avoid a large and rather sudden loss of needed skills and proprietary knowledge as baby-boomers become eligible to retire, employers will have little choice but to adopt policies and practices aimed at motivating older, retirement-eligible employees to defer retirement. However, with little known about how job design and employment practices affect the retention of retirement-eligible workers, managers have little basis for making decisions about which practices to modify or adopt.

Defining retirement as the disengagement from a career job that is linked with the drawing of some retirement benefit, Samuel Bacharach and Peter Bamberger reviewed previous research that has explored the factors related to retirement.  Drawing from job embeddedness theory, they focused on key job, organizational and relational factors that they deemed likely to play a key role in determining whether an individual retires upon becoming eligible to draw retirement benefits.  Data on a wide range of job, organizational and relational variables were collected from a national probability sample of older workers who, at the time of the initial interview, were within six months of becoming eligible to draw some sort of retirement benefit for the first time.  A second wave of interviews—conducted one year later—was used to collect data on the criterion measure, namely retirement upon eligibility.


Although fewer than expected study participants retired upon eligibility, various job, organizational and relational factors were found to influence the likelihood of such behavior.  Moreover, age was found to play a key role in determining the magnitude and nature of these effects:
• Economic downturn and retirement:  Although the study was not intended to examine the impact of economic downturn on the decision to retire, the timing of the study offered interesting results in this regard.  Recognizing that nearly 75% of the 32.3 million retired employees in the US receiving reduced Social Security payments due to entitlement prior to full retirement age (i.e., taking a lower payment upon eligibility for social security at age 62) (Social Security Admin., 2009), and that the median retiree withdraws from his/her career job and starts drawing retirement benefits well before reaching the age of 65 (EBRI, 2009), we assumed that at least 25% of our sample would retire upon eligibility.  In fact, only 50 (i.e., approximately 10% of the sample) actually left their career job and started drawing on their retirement benefits.   This suggests that employers should be prepared for a pent-up demand for retirement as the economy recovers.

• How jobs are designed influences retirement deferment:  Retirement-eligible workers who perceive their job as more challenging were significantly less likely to retire upon eligibility.   However, the impact of job challenge on the likelihood of retirement upon eligibility is diminished among older workers.  Employers seeking to retain retirement-eligible workers should take steps to either redesign the jobs of younger retirement-eligible employees to ensure an adequate level of job challenge, or consider redeploying these workers to more engaging positions.

• Organizational policies and practices influence retirement deferment:  Retirement-eligible workers who reported that their employer offers seasonal employment, the opportunity to work reduced hours or other flex-time alternatives were significantly less likely to retire upon eligibility.  However, the impact of flexible work arrangements on the likelihood of retirement upon eligibility is diminished among older workers.  Although flex-work arrangements may be costly to implement, by allowing for the retention of younger retirement-eligible employees, they may ultimately be justified by lowering the significant costs associated with the search for and training of replacements.

• Relational factors matter: A more supportive work context was found to have divergent effects on the likelihood of retirement upon eligibility depending upon age.  For “younger” workers (i.e., those 61 or younger) increased levels of social support were associated with a lower probability of retirement upon eligibility.    Additionally, the probability of retirement upon eligibility among these “younger” retirement-eligible workers increases as a direct function of the proportion of those among their close work-based peers retiring.   For those 62 and older, increased levels of social support were associated with an increased probability of retirement upon eligibility.  However, their retirement was not influenced by the retirement decisions of their close, work-based peers.  For employers, these findings reinforce the importance of investing in supportive work climates as a means by which to prevent the “snowballing” of one or two employee retirements into a broader wave.

• Overall impact: Taken together, these job, organizational and relational factors explain nearly three times the amount of the variance in the likelihood of retirement above and beyond that explained by more widely studied individual factors such as gender and race, health, and income/wealth.

Bottom Line
HR professionals interested in retaining their older workforce may be able to do so by reviewing and modifying basic organizational policies and practices found to influence the retirement behavior of retirement-eligible workers.  The actions that are likely to have the strongest impact on retirement deferment include job redesign and/or the redeployment of older workers into more challenging positions, the wider adoption of flex-work arrangements, and the encouragement of peer support in the work environment

Study Methods
Applying a prospective study design, the researchers screened 4559 individuals from a national probability sample of older adults in order to identify a target population of 850 meeting the pre-retirement criteria for the study, namely eligibility to receive—for the first time—at least one of four main types of retirement benefits (i.e., defined benefit pension, IRA, 401(k), or Social Security). Of these, 468 participated in all phases of the study. Study participants were interviewed two times, first at 6-12 months prior to retirement eligibility (n=500), and then 6-12 months after reaching retirement eligibility (n=468).

Employee Benefits Research Institute (2009).  The 2009 Retirement Confidence Survey: Economy Drives Confidence to Record Lows; Many Looking to Work Longer, EBRI Issue Brief  (#328).  Washington, D.C.:  Employee Benefits Research Institute.

Social Security Administration (2009).  Annual Statistical Supplement to the Social Security Bulletin, (SSA Publication No. 13-11700).  Washington, D.C.: Social Security Administration.

Toosi, M. (Nov 2006). A new long at long-term labor force projections to 2050. Monthly Labor Review, 19-39.

Read the full report.                      

View the full list of SHRM Foundation funded research.