Working in Retirement

Abstract

Despite the relatively large number of working retirees, very little research has focused specifically on their job experiences. This brief report aims to address this gap in the literature by examining what facets of workplace environment affect job satisfaction and engagement for people who are working in retirement. Data from the 2008 National Study of the Changing Workforce, a sample representative of United States workers, are used to compare workers aged 50 and above who consider themselves retired (N = 203) to those in the same age group who do not consider themselves retired (N = 936). Results suggest that although the economic security offered by the job is less important to job satisfaction and engagement among those who are working in retirement than it is for other older workers, their relationship with their supervisor may be more important. Implications of these findings are considered along with potential directions for future research.

Keywords

employment retirement job satisfaction job engagement

Today, many people in industrialized nations think of retirement as a normal and expected stage of life. However, it was not until the 19th and 20th centuries that retirement was financially possible and culturally accepted for most workers. Instead of viewing early labor force withdrawal as evidence of an inability or unwillingness to work, people began to see retirement as a positive stage in the life course (Atchley, 1976; Quadagno, 1982), a time of rest that older workers had earned. Once retirement became well established, not everyone would have agreed exactly what it meant to be retired. Retirement could be understood as withdrawal from the labor force, receipt of pension or social security benefits, or some combination (Purcell, 2009). The definition of retirement is in further flux today as increasing numbers of older workers consider themselves retired. For instance, a recent study based on a sample of retirees in the United States indicated that almost half of retirees followed a nontraditional retirement path, either remaining in the labor force part-time or returning to the labor force after retirement (Maestas, 2007).

Existing research offers insight into why people define themselves as retired yet remain in the workforce. First, people may believe that they are “retired” because of their age and outlook, yet remain employed for other reasons. For instance, increasing numbers of people are remaining employed past the traditional ages of retirement (Organization for Economic Co-Operation and Development, 2010) because they need additional income or health insurance (Smyer, Besen, & Pitt-Catsouphes, 2009) or to reap psychological and physical benefits (Morrow-Howell, Gao, Zou, & Xie, 2009). Second, many workers transition from their career jobs to part-time or short-tenure bridges job prior to full retirement (Giandrea, Cahill, & Quinn, 2009). As these workers have left their career jobs, they may think of themselves as retired despite remaining employed. Third, substantial numbers of workers leave and then subsequently return to the labor force (Armstrong-Stassen, 2008). For example, Maestas (2007) traced workers for at least 6 years after retirement, finding that 26.4% of retirees returned to work.

Little research has looked specifically at the job experiences of older adults who work in retirement. This report begins to explore this gap in the literature by examining two separate but related research questions. First, what facets of workplace environment affect the job satisfaction (i.e., an expression of a person’s feelings about their job; Bellou, 2010) of workers who consider themselves retired? Second, what facets of workplace environment affect job engagement—how fulfilling employees perceive their job to be (Wefald & Downey, 2009)—among workers who consider themselves retired?

Method

This analysis drew on data from the 2008 National Study of the Changing Workforce (NSCW), a nationally representative study of workers conducted by the Families and Work Institute with the assistance of Harris Interactive. Using a random digit dial procedure, Harris Interactive conducted 3,502 interviews, including 2,769 wage and salaried employees, 478 independent self-employed workers, and 249 small business owners. The response rate was 54.6%. Although the NSCW data were collected approximately every 5 years from 1992 to 2008, the 2008 survey included a new question series related to working in retirement, asked of wage and salaried employees aged 50 and older (N = 1,139).

The dependent variables were job satisfaction and job engagement. Both of these measures were computed so that higher scores represented more positive job experiences and (although based on ordinal level items) were treated as interval level, consistent with previous literature (e.g., Giallonardo, Wong, & Iwasiw, 2010; Rich, Lepine, & Crawford, 2010). The independent variables (described in Table 1) drew on scales identified as distinct components of effective workplaces (Bond, Galinsky, & Hill, 2004; Bond & Galinsky, 2006; Galinsky, Carter, & Bond, 2008; Jacob, Bond, Galinsky, & Hill, 2008), including job challenge and learning, supervisor task support, autonomy, climate of respect, and economic security. Individual variables were standardized and then averaged to compute each scale. The mean variance inflation faction was 1.40 (whereas the maximum was 1.84), below the threshold for concern.

Table 1.

Description of Measures

Autonomy	This scale (α = .76) is the mean of three standardized items:
	Extent to which respondent agrees they have the freedom to decide what to do on their job (from 1 = strongly disagree to 4 = strongly agree)
	Extent to which respondent agrees that it is their own responsibility to decide how their job gets done (from 1 = strongly disagree to 4 = strongly agree)
	Extent to which respondent agrees that they have a lot of say about what happens on their job (from 1 = strongly disagree to 4 = strongly agree)
Climate of respect	This scale (α = .82) is the mean of three standardized items:
	Extent to which respondent agrees that their managers seek information and new ideas from employees (from 1 = strongly disagree to 4 = strongly agree)
	Extent to which respondent agrees that their managers trust what they say (from 1 = strongly disagree to 4 = strongly agree)
	Extent to which respondent agrees that their managers deal honestly and ethically with employees and clients (from 1 = strongly disagree to 4 = strongly agree)
Economic security	This measure is a standardized score, based on the respondent’s likelihood of losing their job and having to look for a new job within the next year (from 1 = not at all likely to 4 = very likely)
Job challenge	This scale (α = .74) is the mean of five standardized items:
	Extent to which respondent agrees that they need to keep learning new things (from 1 = strongly disagree to 4 = strongly agree)
	Extent to which respondent agrees that their job requires them to be creative (from 1 = strongly disagree to 4 = strongly agree)
	Extent to which respondent agrees that they get to do different things on their job (from 1 = strongly disagree to 4 = strongly agree)
	Extent to which respondent agrees that the work they do is meaningful to them (from 1 = not at all true to 4 = very true)
	Extent to which respondent agrees that their job lets them use their skills and abilities (from 1 = strongly disagree to 4 = strongly agree)
Job engagement	This scale (α = .58) is the sum of four standardized items:
	Extent to which the respondent agrees that they look forward to going to work (from 1 = strongly disagree to 4 = strongly agree)
	How much effort respondent puts into job beyond what is required (from 1 = none to 4 = a lot)
	Extent to which the respondent agrees that time on their job passes quickly (from 1 = strongly disagree to 4 = strongly agree)
	How often respondent thinks positively about job when not working (from 1 = never to 5 = very often)
	An abbreviated form of this scale was also computed, excluding the extent to which the respondent agrees that they look forward to going to work (α = .61)
Job satisfaction	This scale (α = .81) is the mean of four standardized items:
	How satisfied the respondent was with their job (from 1 = not satisfied at all to 4 = very satisfied)
	Extent to which the respondent would take the job again (from 1 = definitely not to 3 = yes, without hesitation)
	Extent to which the respondent would recommend the job to a friend (from 1 = would advise against it to 3 = would strongly recommend it)
Supervisor task support	This scale (α = .84) is the mean of three standardized items:
	Extent to which respondent agrees their supervisor keeps them informed of things needed to do job (from 1 = strongly disagree to 4 = strongly agree)
	Extent to which respondent agrees their supervisor recognizes when they do a good job (from 1 = strongly disagree to 4 = strongly agree)
	Extent to which respondent agrees their supervisor is supportive when they have a work problem (from 1 = strongly disagree to 4 = strongly agree)

Respondents aged 50 and older were asked whether they were “working for pay in retirement,” and those who considered themselves working in retirement were compared with other older workers. Control variables considered were numbers of hours worked in paid employment per week (fewer than 35, 35 to 45, or more than 45), the logarithm of wage rate per hour (to reduce skewness), race and ethnicity (non-Hispanic White, Hispanic, Black, and Other), whether female, age in years, whether in fair or poor health, whether had a college degree, and the ratio of income (minus the wage income of the respondent) to the federal poverty line.

Item nonresponse ranged from 0% to 14%. Multiple imputation using chained equations (Royston, 2004) was used to deal with missing data, and coefficients from the imputed data sets were combined using Rubin’s (1987) method. In addition, weights based on the 2007 Current Population Survey were used to calculate all descriptive statistics. Multivariate models were replicated with weights, but as coefficient direction and size were similar in weighted and unweighted models, the unweighted models are discussed in this article.

Results

Among workers aged 50 and older in our sample, slightly fewer than 1 in 5 older workers considered themselves retired. As shown in Table 2, these workers were similar to other older workers in their race and ethnicity, gender, and health. However, respondents who were working in retirement tended to be somewhat older (with a median age of 59 years) and to have fewer financial resources outside of work. In this sample, 36.9% of working retirees would have fallen below the federal poverty line without their wage and salary income. Despite the evidence that many retirees worked for financial reasons, they tended to work fewer hours than their nonretired counterparts. For example, 41.6% of working retirees worked fewer than 35 hr a week, compared with only 12.7% of older workers who did not consider themselves retired. Working fewer hours might be an antecedent or result of considering oneself retired; combined with their lower wage rates (i.e., working retirees averaged only US$14.4 per hour), shorter hours meant that part-time workers reaped less substantial financial rewards for their employment than other older workers did. The financial differences between working retirees and working nonretirees were only part of the story. Those who were working in retirement had noticeably higher median job satisfaction (.65 compared with .15, p < .01), but the differences in job engagement were nonsignificant (p > .05).

Table 2.

Sample Characteristics

	Working and not retired (N = 936)	Working in retirement (N = 203)	Significance^a
Outcome variables
Median job satisfaction	0.15	0.65	***
Median job engagement	0.15	0.19
Job characteristics
Median job challenge	0.22	0.03
Median supervisor task support	0.12	0.42	****
Median autonomy	0.02	−0.04
Median climate of respect	−0.01	0.34	***
Median economic security	−0.11	−0.11
Median wage rate	20.7	14.4	***
Number of hours worked per week			****
Works fewer than 35 hr a week	12.7	41.6
Works 35-45 hr per week	61.7	42.4
Works more than 45 hr a week	25.6	16.0
Demographic and financial characteristics
Race and ethnicity			**
Non-Hispanic White	79.2	81.6
Hispanic	2.5	1.1
Black	10.1	14.2
Other	8.2	3.2
Whether female	52.1	50.0
Median age in years	55	59	****
Whether in fair or poor health	20.5	19.2
Nonwage income as a percentage of poverty line
Below the poverty line	27.9	36.9
Between 100% and 299% of the poverty line	32.4	26.4
More than 300% of the poverty line	39.7	36.7

Based on t values from OLS or logit models, as appropriate.

p < .05. ***p < .01. ****p < .001.

OLS regression models, shown in Table 3, lend further insight into the ways in which the dynamics of job satisfaction and job engagement differ for those who consider themselves retired. A series of separate preliminary models tested interactions between retirement status and each job characteristic and included measures such as nonwage income relative to the poverty threshold, whether in fair or poor health, ethnicity, pension status, health insurance status, whether the respondent had a college degree, age in years, and whether spouse was working. The models shown in Table 3 included coefficients that were significant in preliminary models. Each coefficient represents the increase in the dependent variable for a one-unit increase in the predictor. For instance, a one-unit increase in job challenge is associated with a .24-unit increase in job satisfaction. Among older workers in general, job challenge (b = .24, p < .001), supervisor task support (b = .20, p < .001), climate of respect (b = .30, p < .001), and economic security (b = .12, p < .001) were positively associated with job satisfaction. Similarly, job challenge (b = .36, p < .001), autonomy (b = .08, p < .01), and climate of respect (b = .12, p < .001) were positively associated with job engagement. How engaged and satisfied people are with their jobs depends on how challenging those jobs are, the extent to which their supervisors support them in their work, how much autonomy they have, how respectfully their supervisors and peers behave within the workplace, and how much economic security the job offers. Despite the commonalities, the sizes of the coefficients differ for the engagement and satisfaction models. For instance, a one-unit increase in climate of respect is associated with a .30-unit increase in job satisfaction (Model 1) but only a .12-unit increase in job engagement (Model 2). Although the same factors play a role, they may not have equally strong effects on each outcome.

Table 3.

OLS Regression Models Predicting Job Satisfaction and Job Engagement (N = 1,139)

	Model 1—Job satisfaction				Model 2—Job engagement
	B	SE(b)	t		b	SE(b)	t
Job characteristics
Job challenge	0.24	0.04	6.73	****	0.36	0.03	12.44	****
Autonomy	0.04	0.03	1.35		0.08	0.02	3.23	***
Supervisor task support	0.20	0.03	5.88	****	0.04	0.03	1.65
Climate of respect	0.30	0.03	9.55	****	0.12	0.03	4.76	****
Economic security	0.12	0.02	5.49	****	0.02	0.02	1.02
Working in retirement
Whether working in retirement	0.03	0.06	0.51		0.06	0.05	1.28	*
Whether working in retirement × Supervisor task support	0.16	0.08	1.96	*	0.13	0.07	1.95	*
Whether working in retirement × Economic security	−0.10	0.05	−1.84	*	−0.14	0.05	−3.01	***
Control variables
Whether female	0.00	0.04	−0.1		0.08	0.03	2.26	**
Number of hours worked per week
Works fewer than 35 hr a week (reference = 35-45 hr per week)	0.11	0.06	1.87	*	−0.01	0.05	−0.16
Works more than 45 hr a week	−0.10	0.05	−2.04	**	0.07	0.04	1.85	*
Age in years	0.00	0.00	0.64		0.00	0.00	0.48
Constant	−0.14	0.21	−0.66		−0.16	0.17	−0.92
R ²	.43				.34

p < .10. **p < .05. ***p < .01. ****p < .001.

The strength of these connections differed depending on whether workers saw themselves as “retired.” The effect of economic security on engagement was significantly smaller among working retirees (b = –.14, p < .01). Similar effects (p < .10) were observable for job satisfaction. Although many older adults who work in retirement did so for needed financial income, the stability of any particular job did not appear to affect their job satisfaction or engagement highly. There is also some indication (p < .10) that supervisor task support was more important for the satisfaction and engagement of those who are working in retirement than for other older workers. Working retirees place particular value on whether their supervisor keeps them informed, has realistic expectations, and recognizes both their successes and challenges.

Discussion

The aim of the present study was to examine the antecedents of job satisfaction and job engagement among workers who considered themselves retired, as compared with other workers of similar age. We found that retirees differed from other older workers in how strongly economic security and supervisor task support influenced their job satisfaction and engagement. As they may view themselves as having moved onto another life stage, their priorities and needs may change accordingly. In addition, older workers are in general more satisfied with their jobs (which takes into account both intrinsic rewards of the job and extrinsic advantages such as pay and benefits) but are not more engaged (which focuses primarily on intrinsic rewards). Careful attention should be paid to both intrinsic and extrinsic rewards of work.

These findings need to be understood in light of the limitations of the study. First, many of the scales, although well validated in previous literature, have substantially lower alpha values when applied to the 50-and-above population. The alpha for the engagement scale was much lower than ideal (α = .58), and although the major conclusions of the article remained the same when the item with the lowest interitem correlation was omitted from the scale, the results should be interpreted cautiously. Second, the available sample of working retirees is relatively small. The latter issue caused predictors that were nonsignificant in preliminary models (e.g., work–life fit, wage rate, poverty status, interactions between job characteristics and poverty status or age) to be omitted from the final models to ensure enough statistical power to test for interactions between working in retirement and job characteristics. If working in retirement does become the “new normal” (Mermin, Johnson, & Murphy, 2007), future waves of the NSCW should allow for expanded analysis of working retirees due to the aging of the target population. For instance, analysis of the 2008 NSCW data indicated that 3 out of 4 workers believed that they will work in retirement, suggesting that future waves of this survey will provide much greater sample sizes. Another major limitation of this analysis is its cross-sectional nature. Because of the lack of longitudinal data, it is difficult to identify whether the retired status influences the strength of the effects of job characteristics on engagement and satisfaction or whether people who are satisfied or engaged in their jobs are more or less likely to consider themselves retired. Although the NSCW does not lend itself to longitudinal analysis, if future waves include the question series related to working in retirement, analyses on change between cohorts should be possible. Repeated cross-sectional or cohort analyses should shed light on the extent to which the experience of those working in retirement may change if postretirement work becomes more normatively expected.

Despite its limitations, this study highlights the importance of understanding retirement as a new career stage rather than simply as complete labor-force withdrawal. Recent decades have witnessed the birth of a “third age of adulthood” (Bateson, 2010; James & Wink, 2007), a time of life when older adults actively redefine their involvement in a range of roles, including paid work. Working in retirement is one facet of this life stage. Existing theoretical perspectives may need to be modified and expanded to better account for this phenomenon. For example, research guided by career development theory has shown that career stage is linked to job satisfaction (Flaherty & Pappas, 2002; Lynn, Cao, & Horn, 1996). These studies have typically not incorporated those who are working in retirement because standard career development theory assumes that employees progress through stages (Super, 1990) ending with retirement (e.g., Cron, 1984). Hence, working in retirement is an opportunity for scholars to refine and reframe conceptual frameworks addressing the world of work.

As growing numbers of older adults blur the boundaries of work and retirement, it is crucial for policy makers, employers, and communities to understand their needs and preferences. In particular, research should investigate the connection between self-reported retirement status (as used in this study) and actual work behavior. This study found that a substantial minority of those who were working in retirement worked more than 45 hr per week, raising questions about the meaning of retirement for these workers. Combining objective and subjective measures of retirement status may prove useful. In addition, to the extent that the number of older adults working in retirement grows over time, selecting and training supervisors to promote effective relationships with employees will be crucial to ensuring quality employment. Supervisor relationships play a critical role in determining what opportunities, policies, and programs are available within the workplace (Allen, 2001; Breaugh & Frye, 2007; Eaton & Bailyn, 2000; Thompson, Beauvais, & Lyness, 1999). Particularly for those who are working in retirement, job satisfaction and engagement may be less strongly influenced by organization-level policies than by work-group-level relationships.

Footnotes

Declaration of Conflicting Interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article:

The authors and the Sloan Center on Aging and Work at Boston College would like to thank the Alfred P. Sloan Foundation for their continued support.

Bios

Tay K. McNamara, PhD, is a senior research associate at the Sloan Center on Aging & Work at Boston College. Her research interests include workplace practices and labor force participation among older adults.

Melissa Brown is a research associate at the Sloan Center on Aging & Work and a PhD candidate in the Graduate School of Social Work at Boston College. Her research interests include elder care and the workplace experiences of older adults.

Kerstin Aumann, PhD, is a senior research associate at Families and Work Institute, New York. Her research interests include work–life fit and the impact of gender and generations.

Marcie Pitt-Catsouphes, PhD, is an associate professor in the Graduate School of Social Work and the director of the Sloan Center on Aging & Work at Boston College. Her research interests include workplace flexibility and older workers.

Ellen Galinsky is president and cofounder of the Families and Work Institute. Her research interests include the evolving intersections of work and family.

James T. Bond is the vice president for research and director of work–life research at the Families and Work Institute. His research interests span numerous work–family issues, including youth and employment and parental leave.

References

Allen

T. D.

(2001). Family-supportive environments: The role of organizational perceptions. Journal of Vocational Behavior, 58, 414-435.

Armstrong-Stassen

(2008). Organisational practices and the post-retirement employment experience of older workers. Human Resource Management Journal, 18(2), 36-53.

Atchley

R. C.

(1976). The sociology of retirement. New York, NY: John Wiley.

Bateson

M. C.

(2010). Composing a further life. New York, NY: Alfred K. Knopf.

Bellou

(2010). Organizational culture as a predictor of job satisfaction: The role of gender and age. Career Development International, 15, 4-19.

Bond

J. T.

Galinsky

(2006). How can employers increase the productivity and retention of entry-level, hourly employees? New York, NY: Families and Work Institute.

Bond

J. T.

Galinsky

Hill

E. J.

(2004). Flexibility: A critical ingredient of an effective workplace. New York, NY: Families and Work Institute.

Breaugh

J. A.

Frye

N. K.

(2007). An examination of the antecedents and consequences of the use of family-friendly benefits. Journal of Managerial Issues, 19, 35-52.

Cron

(1984). Industrial salesperson development: A career stages perspective. Journal of Marketing, 48, 41-52.

10.

Eaton

S. C.

Bailyn

(2000). Career as life path: Tracing work and life strategies of biotech professionals. In Peiperl

Arthur

Goffee

Morris

(Eds.), Career frontiers: New conceptions of working lives (pp. 177-198). Oxford, UK: Oxford University Press.

11.

Flaherty

K. E.

Pappas

J. M.

(2002). The influence of career stage on job attitudes: Toward a contingency perspective. Journal of Personal Selling & Sales Management, 3, 135-143.

12.

Galinsky

Carter

Bond

J. T.

(2008). Leaders in a global economy: Finding the fit for top talent. New York, NY: Families and Work Institute and Catalyst.

13.

Giandrea

M. D.

Cahill

K. E.

Quinn

J. F.

(2009). Bridge jobs: A comparison across cohorts. Research on Aging, 31, 549-576.

14.

Giallonardo

L. M.

Wong

C. A.

Iwasiw

C. L.

(2010). Authentic leadership of preceptors: Predictor of new graduate nurses’ work engagement and job satisfaction. Journal of Nursing Management, 18, 993-1003.

15.

Jacob

J. I.

Bond

J. T.

Galinsky

Hill

E. J.

(2008). Six critical ingredients in creating an effective workplace. Psychologist-Manager, 11, 141-161.

16.

James

J. B.

Wink

(Eds.). (2007). The crown of life: Dynamics of the early postretirement period. New York, NY: Springer.

17.

Lynn

S. A.

Cao

L. T.

Horn

B. C.

(1996). The influence of career stage on the attitudes of male and female accountant professionals. Journal of Organizational Behavior, 17, 135-149.

18.

Maestas

(2007, April). Back to work: Expectations and realizations of work after retirement (Rand Working Paper No. WR-196-2). Retrieved from http://www.rand.org/pubs/working_papers/2007/RAND_WR196-2.pdf

19.

Mermin

G. B. T.

Johnson

R. W.

Murphy

D. P.

(2007). Why do boomers plans to work longer? Journals of Gerontology: Series B: Psychological Sciences and Social Sciences, 62B, S286-S294.

20.

Morrow-Howell

Gao

Zou

Xie

(2009). Productive Aging Conference report (CSD Report 09-51). St. Louis, MO: Center for Social Development, George Warren Brown School of Social Work, Washington University.

21.

Organization for Economic Co-Operation and Development. (2010). OECD: Factbook 2010: Economic, environmental and social statistics. Retrieved from http://www.oecd-ilibrary.org/content/chapter/factbook-2010-46-en

22.

Purcell

(2009). Older workers: Employment and retirement trends. Washington, DC: Congressional Research Service.

23.

Quadagno

(1982). Aging in early industrial society: Work, family, and social policy in nineteenth-century England. New York, NY: Academic Press.

24.

Royston

(2004). Multiple imputation of missing values. Stata Journal, 4, 227-241.

25.

Rich

B. L.

Lepine

J. A.

Crawford

E. R.

(2010). Job engagement: Antecedents and effects on job performance. Academy of Management Journal, 53, 617-635.

26.

Rubin

D. B.

(1987). Multiple imputation for nonresponse in surveys. New York, NY: John Wiley.

27.

Smyer

M. A.

Besen

Pitt-Catsouphes

(2009). Boomers and the many meanings of work. In Hudson

(Ed.), Boomer bust? The new political economy of aging (pp. 3-16). New York, NY: Praeger.

28.

Super

D. E.

(1990). A life-span, life-space approach to career development. In Brown

Brooks

(Eds.), Career choice and development: Applying contemporary theories to practice (2nd ed., pp. 197-261). San Francisco, CA: Jossey-Bass.

29.

Thompson

C. A.

Beauvais

L. L.

Lyness

K. S.

(1999). When work–family benefits are not enough: The influence of work–family culture on benefit utilization, organizational attachment, and work–family conflict. Journal of Vocational Behavior, 54, 392-415.

30.

Wefald

A. J.

Downey

R. G.

(2009). Job engagement in organizations: Fad, fashion, or folderol? Journal of Organizational Behavior, 30, 141-145.