Effects of Pre-Retirement Personality, Health and Job Lock on Post-Retirement Subjective Well-being

Abstract

Retirement can be difficult, and experiences vary greatly. Although health, financial status, and family responsibilities have been associated with retirement adjustment, individual psychosocial characteristics may also play a role. Moreover, relatively little is known about the impact of perceived ‘job lock’—the belief that retirement is impossible due to financial or health constraints—and its relationship with later retirement adjustment. The current study addresses these limitations in the literature by examining the retirement transition over four years in a large sample of U.S. adults, with a particular focus on the ways in which personality may affect this transition. Data collected at baseline (2008/2010) and again four years later (2012/2014) included the Big Five personality traits, pre-retirement job lock, self-rated health, and multiple indicators of post-retirement well-being, such as global and experienced well-being (anchored within activities in a single day). Participants were drawn from the Health and Retirement Study (N = 716; M_age = 61.9 at baseline). Results indicated that experienced positive affect was the only post-retirement well-being outcome with a significant association with job lock, although only for those with low conscientiousness. Findings also suggest that pre-retirement personality and subjective health play an important role for post-retirement well-being. Thus, the current study highlights the importance for researchers and practitioners to consider both pre-retirement personality and health when evaluating individuals’ management of the retirement transition. Word Count: 224

The Baby Boomer cohort is retiring in record numbers in the United States, with an estimated 10,000 adults turning 65 every day (PEW Research Report, 2010). With this unprecedented number of older adults transitioning into retirement, it is critical that pre-retirement factors linked to post-retirement adjustment are well understood. Extant research has identified a variety of factors associated with adjustment to retirement, including physical health, financial status, and family responsibilities (e.g. caregiving) (Wang & Hesketh, 2012). After reviewing theoretical perspectives on retirement adjustment (e.g., role theory, life course perspective), Wang, Henken, and Solinge (2011) suggest that adjustment to retirement is a resource-based dynamic process. Resources include physical, cognitive, motivational, social, emotional, and economic, and the authors call for the inclusion of personality as a resource in future research about retirement adjustment.

Personality theory may provide a useful lens for the study of retirement adjustment, as it makes predictions about behavior and subjective experiences. As such, the current study utilizes personality theory to advance the study of retirement adjustment with a longitudinal investigation of the impact of pre-retirement individual characteristics (personality, job lock, health) on post-retirement subjective and experienced well-being. In addition, we consider multiple levels of subjective well-being; global perceptions of positive and negative affect and reports of positive and negative experiences in a day are examined. In this way, we aim to identify whether pre-retirement characteristics are differentially associated with subjective affective evaluations compared with daily affective experiences, which are tied to proximal behavioral and environmental factors.

The retirement transition is not a uniform experience. Previous research has focused on understanding the ways in which specific differences affect adjustment to retirement. For example, a major area of inquiry investigates forced versus chosen retirement and their implications for later well-being and health. Those who feel forced or “pushed” into retirement report negative factors about the retirement process (such as poor physical health), and typically view the decision as involuntary (Shultz, Morton, & Weckerle, 1998). On the other side, there are “pull” factors that tempt individuals to retire; for example, the desire to have more leisure time or volunteer. Individuals who describe being pulled to retire are typically those who view retirement as voluntary and have better health and well-being profiles post-retirement. Calvo, Haverstick, and Sass (2009) found that happiness in retirement was related to whether the decision to retire was forced or chosen rather than the length of time the transition took.

Although there is a large body of research concerning retirement, less is known about the impact of perceived job lock on retirement adjustment (e.g. Wilkie, Cifuentes, & Pransky, 2011). In a way, perceived job lock is similar to forced retirement, in that individuals perceive limited control over their decision to retire. However, an individual who identifies as job locked may have different post-retirement well-being profiles: whereas those forced to retire tend to have lower life satisfaction following the retirement transition (Shultz et al., 1998), those who identify as job locked may experience pronounced increases in well-being when they are free to retire. Thus, we hypothesize that those who are job locked pre-retirement may have a higher retirement bump in well-being compared to those who were not locked. This expectation is consistent with a recent study showing higher levels of life satisfaction post-retirement among those who were job locked compared to those who were not (Fisher, Ryan, Sonnega, & Naudé, 2016). Although Fisher and colleagues (2016) provide important initial support that psychosocial factors, such as job lock, are important for retirement adjustment, other important individual characteristics – such as personality - were not included in their study.

Few studies have examined how individual differences in personality characteristics predict post-retirement adjustment, but personality theory provides a novel perspective. The Five Factor Model of personality, from which the Big Five traits (extraversion, conscientiousness, openness, agreeableness and neuroticism) derive, is one way to measure and describe consistent and enduring individual differences in characteristic ways of thinking, feeling, and behaving (McCrae & John, 1992). Along with earlier theories, such as Activity Theory (Havighurst, 1968), personality traits have been used to predict participation in, and enjoyment of, various activities. For example, extraversion (including facets of gregariousness, excitement-seeking, and positive emotion) has been linked to frequency of socializing and participation in social leisure activities (Jopp & Hertzog, 2010; Stephan, Boiché, Canada, & Terraciano 2014). Relative to retirement, extraversion could potentially be associated with increased social engagement, which in turn may be linked to higher perceptions of positive affect, lower negative affect, and more frequent positive experiences in daily life. Similarly, conscientiousness (including facets such as self-discipline, dutifulness, and achievement striving) is associated with volunteering post-retirement (Mike, Jackson, & Oltmanns, 2014) and cognitively-based recreational activities, such as playing card games, doing puzzles, reading, computer use, and attending lectures (e.g., Jopp & Hertzog, 2010; Stephan et al., 2014). As such, being high in conscientiousness could potentially benefit well-being in retirement through healthier and more active lifestyles. Extraversion is associated with higher frequency of experiences of positive affect in social interactions, and conscientiousness is associated with positive and negative affect (DeNeve & Cooper, 1998). Although little research has examined personality traits and adjustment to retirement or non-retirement, Robinson, Demetre, and Corney (2010) found that extraversion and neuroticism were associated with life satisfaction among pre-retirees, whereas agreeableness and neuroticism were associated with life satisfaction among retirees.

Researchers have also examined how personality may affect well-being during major life transitions (e.g., marriage, childbirth, widowhood, unemployment; Anusic, Yap, & Lucas, 2014; Yap, Anusic, & Lucas, 2012). Yap and colleagues (2012) found that agreeableness, openness, and neuroticism only occasionally predicted changes in life satisfaction. Anusic et al. (2014) found that trajectories of global positive and negative affect differed from trajectories of life satisfaction: affect was sensitive to life events and showed more persistent changes over time than changes in life satisfaction, which were often short-lived. Given the mixed evidence with respect to the moderating role of personality on adaptation to life events (Yap et al., 2012) and the divergent trajectories of different indicators of well-being (Anusic et al., 2014), the degree to which personality moderates adaptation to retirement is still an open question.

Although most research on well-being and personality has focused on global perceptions of affect and satisfaction, Newton and colleagues (2016) examined the association of extraversion and conscientiousness with positive and negative affect experienced in a day’s activities (i.e., experienced well-being). Individuals high in extraversion or conscientiousness reported higher positive affect experienced while socializing, thus suggesting that personality plays a role in daily experiences of affect. Overall, the evidence reviewed suggests that individuals’ personality profiles, perceptions of job lock, and reason for retirement (e.g., forced) may impact behavior and perceptions during retirement that are critical for affective perceptions and daily experiences. However, little is known about how these factors relate to the retirement transition and whether associations of pre-retirement factors are consistent with global and experienced affect post-retirement.

In the current study, we address these limitations by including personality (Big Five traits), job lock, and self-rated health, and measure changes in multiple indicators of well-being over time as individuals retire. We introduce newly-developed measures of experienced positive and negative affect to assess the extent to which pre-retirement characteristics are associated with positive and negative feelings experienced throughout a single day post-retirement (Newton et al., 2016; Smith, Ryan, Queen, Becker, & Gonzalez, 2014). Without a measure of reason for retirement, we include self-rated health as an indicator likely related to an individual’s belief that they were forced out of work (due to health reasons). The current paper makes important strides in theory building across the areas of affect, retirement and personality.

We test a series of hypotheses. H1: Pre-retirement personality characteristics will be associated with global and experienced positive and negative affect post-retirement. Specifically, we expect that extraversion, conscientiousness, openness to experience, and agreeableness will be associated with higher positive and lower negative global and experienced affect; neuroticism will be associated with lower positive and higher negative global and experienced affect. H2: Individuals who report pre-retirement job lock will have higher positive global and experienced affect and lower negative global and experienced affect post-retirement because they will no longer find themselves stuck in unwanted work. H3: Individuals who report worse health pre-retirement will have lower positive (global and experienced) affect post-retirement and higher negative affect post retirement. Although we predict similar associations for global and experienced affect, there is significantly less research on affective experiences in a day, making its inclusion more exploratory. There is little evidence to support specific hypotheses about additive versus multiplicative effects among personality characteristics with job lock and health on well-being; therefore, the current paper also tests for personality interactions to fill this gap. For example, although it is predicted that those in poor health pre-retirement will have lower positive and higher negative global and experienced well-being post-retirement (H3), this may be attenuated if individuals are high in conscientiousness.

Method

Participants

This study uses data from the Health and Retirement Study (HRS), a longitudinal study (collected every two years since 1992) of nationally-representative cohorts of individuals born before 1953 in the United States. Beginning in 2006, a random split-half of the HRS received an enhanced face-to-face interview including the distribution of a self-administered questionnaire with detailed psychosocial content (for details see Smith et al., 2014). The half of HRS participants who did not complete the enhanced interviews in 2006 participated in 2008 with alternating reassessment every 4 years.

For the current study, baseline data are pooled from HRS respondents who completed the psychosocial questionnaire in either 2008 or 2010. The 4-year longitudinal follow-up was collected in either 2012 (2008 baseline) or 2014 (2010 baseline). Selection criteria for this study include participants who self-identified as not retired at baseline and who were then self-defined as partially or fully retired 4 years later at follow-up (N = 716). The analytic sample included participants aged from 44 to 85 (M = 61.9, SD = 6.03); of these, 54.8% were women, 79.2% were White, 67.9% were married, and 37.7% had more than a high school education.

Measures

Job Lock

At baseline for the current study, participants were asked ‘Right now, would you like to leave work, but plan to keep working because: (a) you need the money (Yes or No) and/or (b) you need the health insurance (Yes or No)?’ Those who said ‘Yes’ to either or both of the above were identified as being ‘job locked’. Participants identified as job locked were coded as ‘1’ and those not job locked were scored as ‘0’.

Self-rated Health

Participants’ self-rated health was assessed with one question at baseline that was rated on a 5-point Likert scale (5 = excellent to 1 = poor): ‘Would you say your health is excellent, very good, good, fair, or poor?’

Global Positive and Negative Affect

At both waves, the Positive and Negative Affect Schedule (PANAS-X; Watson & Clark, 1994) was used to assess the degree to which participants experienced 13 positive emotions (e.g. excited, inspired, interested) and 12 negative emotions (e.g. afraid, irritable, hostile) during the last 30 days, rated on a 5-point Likert scale (1 = not at all, 5 = very much). The internal reliability for positive emotions was 0.93/0.94 (baseline/follow-up) and for negative emotions was 0.90 (baseline and follow-up). Global positive and negative affect were assessed at baseline and follow-up.

Experienced Positive and Negative Affect

Participants’ Experienced Well-being (ExWB), a short measure modeled after Kahneman and colleagues’ (2004) Day Reconstruction Method (DRM), was first included in the HRS psychosocial questionnaire in 2012 (see Smith et al., 2014). The original DRM assesses how people spend their time in a day in addition to their positive and negative affective experiences associated with time across the day. The HRS version of the DRM begins with participants identifying how many of seven activities they did yesterday (watching TV, work or volunteering, traveling or commuting, taking care of health needs (e.g. doctor visits), walking or exercising, socializing, running errands). Of those activities, participants are then asked to rate the extent to which they felt a series of positive and negative emotions (positive: happy, interested, content; negative: frustrated, sad, bored) on a scale of 0 = did not experience the feeling at all to 6 = feeling was extremely strong. Alphas ranged from 0.67 to 0.72 for activity-linked positive affect, and 0.74 to 0.76 for activity-linked negative affect. Activity-linked positive and negative affect were calculated as an average of positive/negative over those activities in which the participant engaged.

Personality

At baseline, personality traits were measured with a scale adapted from the MIDI personality scales, an adjective-based measure (Lachman & Weaver, 1997). Participants rated how well each adjective described them on a scale of 1 to 4 (1 = a lot, 2 = some, 3 = a little, 4 = not at all). Neuroticism included four descriptors: moody, worrying, nervous, and calm (α = .72 in 2008, .70 in 2010); extraversion: outgoing, friendly, lively, active, and talkative (α = .74 in 2008, .75 in 2010); agreeableness: helpful, warm, caring, soft-hearted, and sympathetic (α = .78 in 2008 and 2010); conscientiousness: organized, responsible, hardworking, careless, and thorough (α = .66 in 2008, .68 in 2010); openness to experience: creative, imaginative, intelligent, curious, broadminded, sophisticated, and adventurous (α = .79 in 2008 and 2010).

Baseline Covariates

To control for physical functional limitations, problems with five Instrumental Activities of Daily Living (IADLs) were assessed and included in each analysis. Because the current study is focused on adjustment during the transition to retirement, we also controlled for household wealth at baseline (a measure of total assets minus debt), and the extent to which the participant feels control over her/his financial situation. Finally, we included demographic controls for age, gender (1 = Female, 0 = Male), race/ethnicity (1 = White, non-Hispanic, 0 = Other), education (1 = More than high school, 0 = High school or less), and marital status (1 = Married/partnered, 0 = Not married/partnered). Year of survey completion (e.g., 2008 or 2010) was not associated with any outcomes and is not considered further.

Results

Preliminary Results

Table 1 presents the means, standard deviations (or percentage), and intercorrelations for all variables. Regarding subjective well-being, global and experienced positive affect at follow-up were moderately correlated, indicating that they are related but distinct constructs. Global and experienced negative affect were correlated at a similar magnitude. Self-rated health, personality and job lock are all significantly associated with global affect, except agreeableness, which was not significantly associated with negative affect. Although most associations of baseline predictors were significant with our outcomes at follow-up, job lock was not significantly associated with experienced positive or negative affect.

Table 1.

Means, standard deviations, and intercorrelations between well-being, personality, job lock, and self-rated health

	Mean (SD)	1	2	3	4	5	6	7	8	9	10	11	12
1. Baseline Positive - Global	3.69 (.76)	--
2. Follow-up Positive - Global	3.70 (.79)	.63	--
3. Follow-up Positive – Exp.	3.75 (1.40)	.46	.57	--
4. Baseline Negative - Global	1.73 (.61)	−.50	−.32	−.27	--
5. Follow-up Negative - Global	1.73 (.64)	−.36	−.48	−.36	.63	--
6. Follow-up Negative – Exp.	.55 (.79)	−.19	−.30	−.34	.30	.52	--
7. Neuroticism	1.98 (.62)	−.49	−.37	−.23	.67	.52	.23	--
8. Extraversion	3.12 (.47)	.40	.31	.30	−.19	−.12c	−.07b	−.22	--
9. Conscientiousness	3.48 (.41)	.36	.33	.25	−.26	−.27	−.19	−.25	.22	--
10. Agreeableness	3.54 (.46)	.32	.22	.20	−.06a	−.03a	−.08b	−.07a	.44	.27	--
11. Openness to Experience	3.02 (.52)	.47	.39	.25	−.10c	−.09b	−.03a	−.18	.46	.34	.38	--
12. Job Locked	83%	−.15	−.09b	−.06a	.16	.12c	.07a	.14	−.05a	−.07a	.01a	−.09b	--
13. SRH	3.51 (.96)	.38	.37	.30	−.26	−.25	−.14	−.26	.22	.16	.05	.20	−.10c

Note. N = 716.

All correlations significant at p < .001 unless otherwise noted.

^aDenotes Non significant correlation.

^bDenotes correlation significant at p < .05.

^cDenotes correlation significant at p < .01.

Exp. = Experienced.

SRH = Self-Rated Health.

Linear Regression: Pre-Retirement Characteristics on Post-Retirement Well-being

Linear regression results for all well-being outcomes are reported in Table 2.

Table 2.

Linear regressions for global and experienced well-being

	Positive Affect - Global		Positive Affect - Experienced		Negative Affect - Global		Negative Affect - Experienced
	B (S.E.)	p-value	B (S.E.)	p-value	B (S.E.)	p-value	B (S.E.)	p-value
Intercept	2.85 (0.31)	<.001	3.10 (0.63)	<.001	0.91(0.24)	<.001	1.02 (0.37)	.01
Age	−0.01 (0.00)	.02	−0.03 (0.01)	<.001	0.00(0.00)	.63	−0.01 (0.01)	.13
Gender (Woman)	−0.01 (0.05)	.91	−0.02 (0.10)	.80	0.00(0.04)	.94	−0.17 (0.06)	.005
Race (White, non-Hispanic)	−0.11 (0.06)	.07	−0.08 (0.12)	.51	0.04(0.05)	.35	−0.07 (0.07)	.32
Years of Education	0.04 (0.05)	.46	0.01 (0.10)	.91	−0.02(0.04)	.63	−0.02 (0.06)	.69
Married	−0.05 (0.05)	.36	−0.04 (0.10)	.70	0.03(0.04)	.53	−0.13 (0.06)	.04
Household Wealth	0.00 (0.00)	.75	0.00 (0.00)	.21	0.00(0.00)	.23	−0.00 (0.00)	.37
IADL	−0.28 (0.09)	.002	−0.03 (0.19)	.86	0.12(0.07)	.11	0.13 (0.11)	.25
Control over Finances	0.01 (0.01)	.63	0.05 (0.02)	.03	−0.02(0.01)	.02	−0.03 (0.01)	.03
Positive/Negative Affect	0.47 (0.04)	<.001	0.59 (0.08)	<.001	0.48(0.04)	<.001	0.27 (0.06)	<.001
SRH	0.11 (0.03)	<.001	0.17 (0.05)	.002	−0.04(0.02)	.06	−0.02 (0.03)	.60
Neuroticism	−0.08 (0.04)	.05	0.01 (0.09)	.87	0.17(0.04)	<.001	0.06 (0.06)	.34
Extraversion	0.01 (0.06)	.84	0.36 (0.12)	.003	0.05(0.05)	.31	0.00 (0.07)	.97
Conscientiousness	0.16 (0.06)	.01	−0.48 (0.29)	.10	−0.16(0.05)	.001	−0.20 (0.08)	.008
Agreeableness	0.01 (0.06)	.84	0.11 (0.12)	.37	0.03(0.05)	.60	−0.04 (0.07)	.59
Openness to Experience	0.16 (0.05)	.003	−0.05 (0.11)	.66	0.03(0.04)	.51	0.10 (0.07)	.12
Job Locked - Any Reason	0.02 (0.06)	.77	−0.03 (0.13)	.82	0.00(0.05)	.99	−0.02 (0.08)	.78
SRH by Agreeableness	−0.10 (0.05)	.03
SRH by Conscientiousness	−0.13 (0.06)	.02
Job Lock by Conscientiousness			0.88 (0.31)	.005
R2	.46		.28		.44		.14

Note. N = 716.

SRH – Self-Rated Health

IADL – Instrumental Activities of Daily Living

Positive Affect – Global

Over and above sociodemographic controls, IADLs, perceived control over finances, and baseline global positive affect, self-rated health, conscientiousness and openness predicted higher global positive affect post-retirement, R² = .46, F(18, 697) = 32.72, p < .001, supporting H1. In addition, we found that the association of self-rated health on positive affect varied by level of conscientiousness (see Figure 1). While self-rated health was positively associated with global positive affect post-retirement, the association was strongest among those with low levels of conscientiousness. The association of self-rated health with global positive affect also varied by level of agreeableness in a pattern similar to that reported in Figure 1 for conscientiousness. Job lock was not associated with global positive affect, counter to H2.

Figure 1.

Association of self-rated health at baseline on global positive affect differs based on level of Conscientiousness. Specifically, the positive association of self-rated health with global positive affect was only significant among individuals with low Conscientiousness.

Positive Affect – Experienced

Regression results show that self-rated health and extraversion were positively associated with experienced positive affect post-retirement, partially supporting H1 and supporting H3 (R² = .28, F(17, 698) = 15.75, p < .001). Counter to H2, there were no significant effects of job lock on experienced positive affect. A significant conscientiousness x job lock interaction indicated that individuals low in conscientiousness and who were job locked had significantly lower experienced positive affect (see Figure 2).

Figure 2.

The association of baseline job lock status with experienced positive affect differed by level of Conscientiousness. Reporting pre-retirement job lock was associated with significantly lower experienced positive affect (compared to those who were not job locked) if the individual also reported low Conscientiousness. The association of job lock with experienced positive affect was not significant in those who report high Conscientiousness.

Negative Affect – Global

Results show a statistical trend that lower self-rated health pre-retirement was associated with higher global negative affect post-retirement, providing weak support for H3 (R² = .44, F(16, 699) = 34.62, p < .001). However, over and above health and other covariates, higher neuroticism and lower conscientiousness were associated with higher global negative affect post-retirement, in partial support of H1. Job lock was not associated with global negative affect, nor were there any significant interactions.

Negative Affect – Experienced

Counter to the other well-being outcomes and H3, self-rated health was not associated with experienced negative affect post-retirement. The most powerful predictor of experienced negative affect post-retirement was global negative affect at baseline; however, individuals who reported lower conscientiousness also had higher experienced negative affect after retirement (H1). There were no significant interactions with self-rated health or job lock, R² = .14, F(16, 699) = 7.13, p < .001.

Discussion

The current study investigated the impact of pre-retirement job lock, self-rated health, and personality on post-retirement global and experienced positive and negative affect using data from US adults. Counter to expectations, the current study found little evidence that job lock was associated with retirement adjustment after accounting for personality traits, self-rated health, and sociodemographic characteristics and controls. Experienced positive affect was the only post-retirement well-being outcome with a significant association with job lock, and only among those with low conscientiousness. Conscientiousness has been consistently linked to a variety of facets, such as high orderliness, industriousness, and low impulsivity (MacCann, Duckworth, & Roberts, 2009). Individuals low in conscientiousness are also more likely to go over their credit limits, oversleep, and engage in more TV watching, which has been linked to low experienced positive affect in older adults (Jackson et al., 2010; Smith et al., 2014). It may be that individuals low in conscientiousness who reported job lock had fewer financial and behavioral resources in retirement, and experience heightened daily stressors as a result, thus explaining how pre-retirement job lock may exacerbate the low conscientiousness-daily positive affect link. This pattern was not found for global positive affect, which suggests that daily experiences were less positive but that overall perceptions of positive affect are robust to deficits in conscientiousness. The hypothesis that job lock would result in higher positive and lower negative affect post-retirement, due to relief from no longer being locked into unwanted work, was not fully supported. The results suggest that the impact of job lock on experienced positive affect may be contingent on an individual’s level of conscientiousness, which would have been missed if only main effects had been tested. Overall, the results suggest that job lock may play an instrumental role in life satisfaction (Fisher et al., 2016) as well as experienced positive affect.

Conscientiousness was the personality characteristic most commonly linked to post-retirement well-being. Self-rated health was also a consistent predictor of well-being, including some important interactions with personality. The effects of self-rated health on global positive affect were strongest among people low in conscientiousness and agreeableness, suggesting that these traits are psychological resources for retirement adjustment among those with poor health.

Positive versus Negative Affect

A primary goal of the current study was to investigate whether personality moderated associations between baseline characteristics and retirement adjustment (as discussed above). Our results suggest that deficits in protective personality characteristics and health have negative impacts on global positive affect.

Why are personal characteristics unrelated to negative affect? Given the overall low levels of reported negative affect in this sample, it may be that there is too little variability in negative affect after accounting for both baseline negative affect and neuroticism, which have substantial overlap (McNiel & Fleeson, 2006). However, it is also possible that the criteria on which individuals draw to make evaluations about positive and negative affect differ. For example, positive evaluations of well-being may be more susceptible to dispositional characteristics like personality. The influence of personality characteristics on well-being variability is a fruitful subject of future research (McNiel & Fleeson, 2006).

Global versus Experienced Affect

The current paper also highlights the importance of how retirement adjustment is conceptualized by looking at positive and negative affect from both global (i.e., trait) and experienced (i.e., state) perspectives. Overall, our results explained a greater degree of variability in the global indicators (see Table 2); however, this may be due to the fact that we had baseline levels of the global affect measures available as control variables in our models but not baseline levels of experienced affect. It is possible that their exclusion may explain the smaller magnitudes of variability accounted for in the experienced well-being outcomes.

Limitations

The current study does not specifically measure the circumstances surrounding the retirement transition, beyond job lock and self-rated health. Future research would benefit from examining how personality interacts with other features of retirement that affect well-being (e.g., the volition of the decision to retire; Calvo et al., 2009). The current study was also limited by having only two assessments for global well-being and only one assessment for experienced well-being. Having more data points would allow the researchers to examine trajectories of adaptation and adjustment, and shed light on whether personality continues to predict individual differences in well-being over longer stretches of time (Anusic et al., 2014; Yap et al., 2012).

Implications for Public Health/Translational Impact

The current study’s findings have public relevance: positive emotional profiles have been associated with better physical and mental health (Lyubomirsky, King, & Diener, 2005). This is particularly pertinent, given research demonstrating that personality is related to health and well-being issues around the retirement transition, especially for “vulnerable” populations, such as those who retire early due to disability (Blekesaune & Skierbekk, 2012). The current study found consistent support that over and above demographic and financial considerations, individuals low in conscientiousness and self-rated health are more likely to have worse global and experienced affect post-retirement. Future research should consider programs that target well-being enhancement and maintenance in these at-risk individuals. As such, brief assessments of self-rated health and personality may be one of the more efficient ways to identify at-risk individuals. Our findings highlight the importance of considering individual differences and the impact of these differences on downstream outcomes for the aging population.

Much work in public health is focused on negative health outcomes, whereas subjective well-being is not simply the opposite or absence of depression. Rather, well-being exists in addition to mental illness, and placing a focus on positive components of well-being has been a noticeable gap in public health research. Understanding which factors are likely to help retirees flourish will help to foster enhanced quality of life for the growing and diverse aging population.

Public Significance Statement.

This study finds that individuals’ characteristics before retirement, such as health and personality, are important for affective well-being once retired. Given the acceleration of retirees due to the aging Baby Boomer cohort, this study highlights the need to consider pre-retirement characteristics when promoting the health and well-being of retirees.