Does Retirement Change What Individuals Value in Life? Results From a 3-Year Panel Study.

OBJECTIVES: In previous research on retirement, what individuals value in life is often assumed to remain stable after the transition into retirement. However, retirement exposes individuals to new social settings and might thus prompt them to reevaluate their life orientations. Quantitative empirical knowledge about this process is limited, though. This study examines the impact of retirement on changes in the perceived importance of self-development, social status, societal contribution, and generativity in older adults' lives. We draw on the life-course framework to develop hypotheses about which life orientations are more likely to change after retirement and how.
METHODS: We analyzed data collected in 2015 and 2018 among 5,034 Dutch individuals aged 60-65 and employed at baseline. Around half had fully retired at follow-up (either voluntarily or involuntarily), and 10% worked after retirement.
RESULTS: Conditional change models reveal that voluntary full retirement was linked to statistically significant-but (very) small-decreases in the importance of self-development, social status, societal contribution, and generativity compared to continuous career work. Differences in changes of life orientations between retirement processes (i.e., postretirement work, involuntary full retirement) were also small. DISCUSSION: The findings point to relatively high levels of continuity during the transition from career employment to retirement, given the small effect sizes observed. The notion that after retirement, individuals will reevaluate what they value in life-as expected in light of the exposure-based mechanism from the life-course literature-seems to be less pronounced than initially expected.

Retirement is often portrayed as a life stage that allows older adults to develop new roles and lifestyles to contribute to the common good and maintain their well-being (James et al., 2016). Research confirms that retirement prompts behavioral changes such as increased engagement in volunteer work or contact with family members (see Fisher et al., 2016 for a review). Gerontological theories (e.g., role theory) generally expect these behavioral changes to reflect the need of individuals to compensate for the loss of valued aspects of work after retirement (Wang & Shultz, 2009). This “role replacement” mechanism assumes that the aspects in life that individuals value during working life remain important across the retirement transition. Life-course literature, however, suggests that major life events trigger individuals to adapt their views, attitudes, and values (Baxter et al., 2015; Vidal & Lersch, 2019). Yet, empirical quantitative insights about the impact of retirement on changes in the importance of different psychosocial aspects (e.g., self-development, societal contribution) in life are scarce. Therefore, this study will address the following research question: To what extent and how does the transition from career work into retirement change what individuals value in life?

Several longitudinal studies have examined how life-course transitions change views, attitudes, and values. Frequently studied transitions in the life-course literature are first-time parenthood (Katz-Wise et al., 2010), union transitions (Cunningham & Thornton, 2005), relocation (Vidal & Lersch, 2019), and unemployment (Naumann et al., 2016). These studies generally support the notion that life-course transitions expose individuals to a different normative context with new ideas and beliefs and prompt individuals to adapt their views accordingly (Kroska & Elman, 2009; Vidal & Lersch, 2019). For example, Katz-Wise et al. (2010) find that gender-role attitudes become less egalitarian when individuals become parents. Yet, this line of longitudinal research on attitude change pays limited attention to retirement as a driver of such changes, even though it might be a major transition in later life.

The gerontological literature suggests that the transition into retirement is a major life transition, which affects the lives of individuals in many ways. Retirement has been linked to changes in behaviors (e.g., volunteering, leisure), social relationships, health outcomes, and well-being (see Fisher et al., 2016 for a review). Moreover, retirement has been found to affect personal processes, such as psychological needs (Henning et al., 2019; Stenling et al., 2021), personality (Löckenhoff et al., 2009; Schwaba & Bleidorn, 2019), self-esteem (Bleidorn & Schwaba, 2018), and sense of purpose in life (Yemiscigil et al., 2021). For example, Henning et al. (2019) argue that a need for autonomy would be more important in retirement. Building on these studies, it is remarkable that insights into attitudinal or value changes during the retirement transition are still relatively scarce. Kulik (1999) compared life orientations between preretired and retired men in a small-scale cross-sectional study and showed that retired men perceive family relationships as more important while preretired men value instrumental and economic aspects of life more. Nuttman-Shwartz (2007) also found in a qualitative study that family relationships become more important for men during the retirement transition. Although these studies suggest that retirement may prompt changes in what individuals value in life, longitudinal insights are needed to better grasp the impact of retirement on changes in life orientations.

In this study, we examine life orientations reflecting different psychosocial aspects of life, which many individuals may consider important (Caro et al., 2009) and which reflect aspects of life that could be linked to paid work (Rosso et al., 2010), but also to engage in other social roles (Hirshorn & Settersten, 2013): self-development, social status, societal contribution, and generativity. Self-development refers to developing existing skills or learning new skills (Edlund & Grönlund, 2010). Social status refers to the own position within the social hierarchy and the recognition and respect from others (Steverink & Lindenberg, 2006). Societal contribution refers to contributing to the well-being of the community and society (Hirshorn & Settersten, 2013). Generativity refers to sharing knowledge with the younger generation (Mor-Barak, 1995). These life orientations have been found to guide individuals’ retirement-related intentions, plans, experiences, and behavior (Blekesaune & Hansen, 2021; van Solinge et al., 2021; Wöhrmann et al., 2016; Zhan et al., 2015). Moreover, these life orientations arise in discussions about active aging (Foster & Walker, 2015) and productive aging (Bass & Caro, 2001). Little is known, however, about the question of whether these life orientations themselves change when individuals move from career work into retirement.

This study aims to contribute to the literature in three ways. First, this study is the first large-scale study to examine changes across the retirement transition in the extent to which individuals perceive self-development, social status, making societal contributions, and generativity important in their lives. As such, it provides insights into whether and to what extent these central psychosocial aspects of life change when individuals enter retirement. Second, we draw on the life-course framework to develop hypotheses about which specific life orientations are relatively likely to change after retirement and in what ways. This allows us to understand better how individuals navigate the retirement transition. Third, we use large-scale quantitative two-wave data collected in 2015 and 2018 from 5,034 individuals (aged 60–65 at baseline) to study the research question. Approximately half of the respondents retired fully during the study period—that is, they ended continuous employment within a career trajectory by using a retirement arrangement (Denton & Spencer, 2009). These data enable us to observe the retirement transitions of many older workers and compare them with continuous career workers. Furthermore, the data offer the possibility to capture heterogeneity in retirement processes (i.e., postretirement work, involuntary full retirement).

This study took place in the Netherlands, where most collective labor agreements prescribe that employment contracts end automatically when employees reach the public pension age. This age was 65 years and 3 months in 2015, when the baseline data were collected, and is gradually being increased and will reach age 67 by 2024, at which point it will be linked to projected life expectancy. Employees in the studied birth years 1950–1955 generally work until the public pension age because leaving the labor market before the public pension age (e.g., early retirement) substantially reduces pension benefits and has become less popular (Montizaan et al., 2021). Employees can choose to remain on the labor market after retirement in a postretirement job while collecting a pension income. They will have to enter a new labor contract with an employer, which is often a fixed-term contract, or opt for self-employment (Tunney & Oude Mulders, 2021).

Theoretical Background

The life-course perspective provides the theoretical basis in this study to examine whether and how retirement affects changes in life orientations. It emphasizes—among other things—that the life course is “a sequence of socially defined events and roles that the individual enacts over time” (Giele & Elder, 1998, p. 22). These socially defined events and roles are embedded in a social normative context with specific ideas and beliefs about desirable behavior for individuals (Kohli, 2007). When discrete roles and status change during a life-course transition, individuals enter another normative context. Consequently, individuals are less exposed to the ideas and beliefs that have previously shaped their personal views and become exposed to their new context’s potentially different ideas and beliefs. Because of such a change in contexts, individuals may question their previous views and reevaluate what is important to them in life. Life-course scholars refer to this process as the exposure-based explanation of attitude change (Kroska & Elman, 2009; Vidal & Lersch, 2019). We build on this exposure mechanism to deduce hypotheses about the impact of the transition from work to retirement on changes in life orientations. In the next paragraph, we address the central social contexts to which individuals are exposed during the transition to retirement, namely, the work and retirement contexts. Thereafter, we formulate specific hypotheses about the four life orientations central to this study.

Work and Retirement Contexts

Work has developed into an integral part of life (Kohli, 2007). Several scholars note that current work cultures require individuals not just to work but to become workers (Weeks, 2011). As such, work structures might transmit what individuals value in life. Workplaces are traditionally structured to maximize workers’ productivity (Gallie et al., 2012). Research shows that this entails an emphasis on skill and knowledge development (Edlund & Grönlund, 2010), empowerment (Bunderson & Reagans, 2011), as well as teamwork (Gallie et al., 2012). Moreover, organizations have shifted toward more society-centered values, that is, greater corporate social responsibilities (Bourne & Jenkins, 2013). Several scholars point to cultures of work centering around achievement and passion (Farrugia, 2019), or as Weeks (2011, p. 8) summarizes “dreams of individual accomplishment and desires to contribute to the common good become firmly attached to waged work.”

During the transition from career work into full retirement, individuals leave the context of paid work. Consequently, they are no longer exposed to the dominant ideas and beliefs of the work context. In turn, retirement becomes the prevailing normative context. Retirement has been suggested to represent the “crown of life” (James et al., 2016), given the differentiation between the third age of postwork and the fourth age of disability and decline (Laslett, 1989). The retirement years are now expected to be a time for self-realization and fulfillment (James et al., 2016) and “a quest for freedom, flexibility, and lack of commitment” (Seaman, 2012, p. 252). At the same time, the perceptions and expectations of retirement are changing. Concepts like active aging (Foster & Walker, 2015) and productive aging (Bass & Caro, 2001) challenge the notion of an “earned retirement” filled with leisure by emphasizing that later life and retirement are not the end of activity and productivity. According to this perspective, retirees instead are expected to stay active and engaged for as long as possible to maintain their quality of life and the welfare of their surroundings (van Dyk, 2014). This refers to continuing participation in various life domains, especially in socially meaningful activities.

Hypotheses

To understand the impact of retirement on changes in life orientations, it is important to consider the changes in the dominant ideas and beliefs in both the work and retirement contexts. What individuals have valued during career employment can be expected to change with the transition into retirement because they are less exposed to the work context and more exposed to the retirement context. Below we will hypothesize how different life orientations might change during the retirement transition.

Self-development

Given that great value is placed on building skills and knowledge during a career to keep up with developments in the workplace (Edlund & Grönlund, 2010), individuals are constantly exposed to expectations to develop themselves to succeed in a career and keep up with their job (Kyndt et al., 2014). Research shows that human resource strategies are often directed toward training (van Dalen et al., 2015). Retirement can then be considered to detach individuals from the organizational expectations to participate in training. Despite societal expectations that retirees use retirement for self-growth and fulfillment, these expectations might be less explicit than during career employment. Therefore, we hypothesize that the transition from career work into full retirement is linked to a decreased perceived importance of self-development in life (Hypothesis 1).

Social status

There is a substantial concern with social status in the work context, given that workplaces have a social hierarchy that signals employees’ occupational standing (Bunderson & Reagans, 2011). Central to career work is maintaining or heightening one’s position within the social hierarchy (Piazza & Castellucci, 2014). Therefore, the workplace fosters the importance of social status. Expectations of occupational achievement might fade in retirement, as retirement is the exit of career employment (Denton & Spencer, 2009). Then, other aspects are more important than striving for higher status, such as family (Kulik, 1999; Nuttman-Shwartz, 2007). Therefore, we expect that the transition from career work into full retirement is linked to a decreased perceived importance of social status in life (Hypothesis 2).

Societal contribution

The work context signals the importance of contributing to the common good. Societal contribution is often seen as part of the job (Rosso et al., 2010) or an extension of the labor market (Wilson, 2012). The workplace encourages thus the importance of societal contributions. In retirement, these expectations increasingly prevail, at least compared to some decades ago when retirement was still considered a withdrawal from social obligations (Johnson & Mutchler, 2014). Now, retirees are encouraged to keep busy in retirement through paid or unpaid work and engage in activities that contribute to the common good (Hirshorn & Settersten, 2013; James et al., 2016). Therefore, we expect that the transition from career work into full retirement will not change the importance individuals attach to contributing to society (Hypothesis 3).

Generativity

Great value is placed on sharing knowledge with the younger generation at work. The work organization often revolves around the notion that older generations pass on their knowledge and expertise to the younger generation (Krahn et al., 2020). Thus, the workplace transmits the importance of generativity. In retirement, expectations of generativity remain (Moen, 1996) and might then come from family and the community (Szinovacz & Davey, 2001). Retirees may face more normative expectations to respond to family demands because retirees gained much free time with retirement and might be expected to prioritize the needs of others over their free time (Weiss, 2005). For instance, children might expect retirees to intensify their grandparental investment (Szinovacz & Davey, 2001). Therefore, we expect that the transition from work into full retirement will not change the importance individuals attach to generativity (Hypothesis 4).

The Complexity of the Retirement Process

Retirement is for a considerable share of individuals not an abrupt and complete exit from the labor force but could be a complex process (Denton & Spencer, 2009). Some retirees leave the workforce entirely, while others continue to work for pay after retiring from career employment (Beehr & Bennett, 2015). Furthermore, although many individuals transition into retirement voluntarily, also a considerable share is pushed toward retirement, for instance, because of external reasons (e.g., pressure by an employer, reaching public pension age) or personal reasons (e.g., poor health; van Solinge & Henkens, 2007). As such, it may be relevant to distinguish two additional groups of retirees: those who are engaged in paid work after retiring from career employment, and full retirees who perceive their retirement as involuntary.

From a theoretical perspective, the extent to which both these groups of retirees encounter a change of contexts across the retirement transition can be expected to fall in-between the “extreme” groups of continued career workers and individuals who fully retired in a voluntary way. Given that working retirees remain on the labor market, they are still to a certain extent exposed to a work context and its expectations. For involuntary full retirees, it can be expected that they keep taking the ideas and beliefs of the work context as their reference point because they resist the change into the retirement context. Research has shown, for instance, that retirees who retired involuntarily are more connected to their past work role in retirement than those who retired voluntarily (Damman & Henkens, 2017).

Design and Methods

Sample

This study uses data from the NIDI Pension Panel Study (NPPS), a large-scale longitudinal study in the Netherlands that examines older workers during their transition from career work into retirement (Henkens & van Solinge, 2019). The data for the first wave were collected in 2015. The sample was drawn from the three largest pension funds in the Netherlands, which cover different sectors (government, education, construction, care, and social work), and represent roughly 49% of Dutch wage-employed workers. A sample of organizations was selected from the pension funds, stratified by size and sector. A random sample of workers born between 1950 and 1955 who worked at least 12 h a week was then drawn. A total of 15,470 questionnaires were sent out in 2015, of which 6,793 were completed (response rate of 44%). A follow-up study took place in 2018 with the same participants receiving a new questionnaire. A total of 5,316 respondents participated in the follow-up survey (response rate of 79%). For the analysis, the analytical sample consisted of those respondents who participated in the NPPS baseline and follow-up survey. After excluding participants without information on the dependent variables (N = 282), the base analytical sample comprised 5,034 older adults.

Measures

Our dependent variables capture four life orientations: self-development, social status, societal contribution, and generativity. Respondents were asked about the extent to which they find these different aspects important in their lives. The items were constructed for this survey building on empirical studies (Caro et al., 2009; Kooij & van de Voorde, 2011; Zhan et al., 2015). The items included, for example, statements about the importance of learning new skills, prestige, helping those in need in society, and passing things on to the younger generation. Table 1 presents the grouping of the statements into the four life orientations. The life orientation social status was measured by two items and the other life orientations (self-development, societal contribution, and generativity) were measured by three items each. Reliability (Cronbach’s alpha) was generally high both at baseline and follow-up, indicating internal consistency of the items. The Cronbach’s alpha was slightly lower for the two-item scale measuring social status (0.62) than the three-item scales measuring self-development (0.80), societal contribution (0.78), and generativity (0.73) at baseline. The scales were standardized to obtain effect sizes in terms of Cohen’s d for the dummy variables in the analyses.

Table 1.

Descriptive Statistics of Life Orientations at Baseline, in Percentages (N = 4,513)

	Items	Very unimportant	Unimportant	Neutral	Important	Very important
Self-development
1	Always learning new things	0.6	5.5	34.7	50.5	8.8
2	Fully developing my talents	0.6	6.8	43.7	42.2	6.7
3	Fully developing myself	1.7	9.4	45.9	37.8	5.3
Social status
4	Having much societal prestige	10.8	35.6	45.0	7.8	0.8
5	Doing things that offer me a lot of appreciation	2.0	14.6	47.3	33.4	2.7
Societal contribution
6	Devoting myself to society	0.6	4.5	43.7	45.9	5.3
7	Helping to make the world a better place	1.8	9.2	53.8	31.3	3.9
8	Doing somethings for those in need in society	0.5	3.6	41.4	49.5	5.1
Generativity
9	Passing things on to younger generations	0.3	2.2	22.0	60.3	15.3
10	Staying in contact with younger people	0.2	1.6	18.4	65.8	14.0
11	Give opportunities to younger generations	0.4	2.2	25.2	58.7	13.6

Note: The descriptive statistics are based on the values prior to imputation.

The central explanatory variable was the retirement process. Because all respondents were in career employment at baseline, we identified the retirement process from the follow-up questionnaire that inquired about whether individuals used a retirement arrangement to exit career work (e.g., early retirement, reaching mandatory retirement age; cf. Denton & Spencer, 2009), whether they were engaged in paid work at follow-up, and whether the retirement transition was voluntary. The constructed variable comprised the following categories: (a) remaining in career employment, (b) retired and working in a postretirement job, (c) fully retired voluntarily, (d) fully retired involuntarily for organizational reasons (i.e., due to pressure from the organization and/or colleagues, and/or mandatory retirement age), and (e) fully retired involuntarily for other reasons (e.g., health, caregiving responsibilities).

We controlled for baseline sociodemographic characteristics (age, gender, partner status, health status, and wealth) and baseline employment characteristics (occupational status, work hours, and occupational sector). Table 2 presents the mean, coding, and wording of the survey questions for the dependent, independent, and control variables used in the analysis.

Table 2.

Mean/Share, Coding of Independent Variables, and Wording of Survey Questions for Career Workers (N = 2,073), Retirees (N = 2,783), and the Total (N = 4,856)

Variable	Mean/ share			Coding and psychometric properties	Description/wording (questions translated from Dutch)
	Career workers	Retirees	Total
Life orientations
Self-development				Three-item scale, range 1 (very unimportant) to 5 (very important); standardized for analysis	Question: Could you indicate for the following things how important they are in your life? (1) always learning new things; (2) fully developing my talents; (3) fully developing myself
Baseline (alpha=.80)	3.53	3.48	3.48
Follow-up (alpha=.80)	3.59	3.42	3.42
Social status				Two-item scale, range 1 (very unimportant) to 5 (very important); standardized for analysis	Question: Could you indicate for the following things how important they are in your life? (1) having much societal prestige; (2) doing things that offer me a lot of appreciation
Baseline (alpha=.62)	2.90	2.83	2.83
Follow-up (alpha=.60)	2.96	2.75	2.75
Societal contribution				Three-item scale, range 1 (very unimportant) to 5 (very important); standardized for analysis	Question: Could you indicate for the following things how important they are in your life? (1) devoting myself to society; (2) helping to make the world a better place; (3) doing something for those in need in society
Baseline (alpha=.78)	3.45	3.45	3.45
Follow-up (alpha=.80)	3.52	3.44	3.44
Generativity				Three-item scale, range 1 (very unimportant) to 5 (very important); standardized for analysis	Question: Could you indicate for the following things how important they are in your life? (1) passing things on to younger generations; (2) staying in contact with younger people; (3) give opportunities to younger generations
Baseline (alpha=.73)	3.88	3.86	3.86
Follow-up (alpha=.74)	3.92	3.82	3.82
Retirement process
Continuous career work			0.43	Dummy variable coded 0–1, 1=transition to respective category	Questions: (1) Have you made use of one of the following arrangements (e.g. early retirement) since 2015?; (2) Which situation applies to you? Response categories are 1=I work for pay, 2=I do not work (for pay) any longer; (3) Was your decision to stop working voluntary or not? Response categories are 1=yes, completely voluntary, 2=no, partly involuntary, 3=no, completely involuntary; (4) What made your decision involuntary (e.g. pressure from employer)?
Postretirement job			0.07
Full retirement: voluntary			0.35
Full retirement: involuntary (organizational reasons)			0.11
Full retirement: involuntary (other reasons)			0.05
Employment characteristics
Occupational status baseline	0.01	0.03	0.01	Coding is based on 2008 International Socio-Economic Index of Occupational Status and was standardized using the full sample	Question: What is your job or profession?; In which category could your job or profession be grouped?
Full-time work baseline	0.51	0.45	0.50	Dummy variable coded 0-1, 1=36 or more hours per week	Question: How many hours do you work on average (per week)? Excluding overtime.
Sector baseline				Categorical variable	Information about sector is obtained via the three participating pension funds: government, education, construction, care, welfare.
Government	0.28	0.27	0.28
Education	0.23	0.26	0.25
Construction	0.18	0.19	0.19
Care	0.13	0.12	0.13
Welfare	0.19	0.15	0.16
Sociodemographic characteristics
Age baseline	61.11	62.75	62.04	Continuous variable, range 60–65 years at baseline	Question: In what year were you born?
Gender	0.48	0.44	0.45	Dummy variable coded 0–1, 1=woman	Question: Are you a man or a woman?
Partnered baseline	0.79	0.81	0.80	Dummy variable coded 0–1, 1=with partner (married, cohabitation, living apart)	Question: Do you have a partner? Response categories are 1=Yes, I am married; 2=Yes, I cohabit with a partner; 3=Yes, I do have a partner but we do not live together; 4=No, I am single
Chronic health condition (CHC) CHC baseline: 0	0.41	0.33	0.36	Categorical variable	Question: Do you have one or more of the following longstanding diseases, conditions, or handicaps (diagnosed by a doctor): (a) arthritis, (b) migraine/ severe headache, (c) cardiovascular diseases, (d) asthma/ bronchitis, (e) gastrointestinal disorders, (f) diabetes, (g) psychological complaints, (h) life-threatening diseases, (i) sleep problems, (j) other disease
CHC baseline: 1	0.34	0.36	0.35
CHC baseline: 2+	0.25	0.31	0.29
Wealth baseline	158.90	188.95	175.93	Quasi-interval measure, range 2.5–750	Question: How large do you estimate your total wealth (own house, savings, stocks, etc. minus debts/mortgage) to be? Response categories are 1=less than 5.000 euros to 7=more than 500 thousand euros. We used class averages and report values in thousands.

Note: The descriptive statistics are based on the values prior to imputation.

In general, item nonresponse was lower than 8% (found in the measure of wealth) and was dealt with using multiple imputation procedures (Stata 15: mi impute chained). We imputed the variables with missing cases 25 times and used information from dependent, independent, and control variables. The models were run for all 25 data sets and combined using the mi estimate command in Stata.

Analysis

We estimated separate ordinary least squares regression models with clustered standard errors for the perceived importance of self-development, social status, societal contribution, and generativity. To examine the change in these life orientations, each life orientation at follow-up was predicted by the value of the life orientation at baseline, retirement status, and control variables. This approach is referred to as conditional change modeling, and it estimates how an independent variable is linked with the outcome at follow-up while controlling for the outcome at baseline (Finkel, 1995). We can therefore interpret the effect of retirement as the effect on the change in life orientations between baseline and follow-up. Moreover, in conditional change models, potential selection effects that might be linked to retirement and life orientations are reduced as initial differences in the outcome are controlled for (Johnson, 2005).

Results

To gain insights into what individuals value in life, Table 2 presents the distribution of the single items at baseline. The descriptive results generally show that individuals highly valued self-development, societal contribution, and generativity while they valued social status only to some extent. For example, the majority perceived learning new skills (59%), helping those in need in society (55%), and passing things on to the younger generation (76%) as (very) important. In turn, most were neutral about the importance of societal prestige (45%), whereas 9% found it (very) important.

Table 3 presents the results of the conditional change models. The models report the effect of retirement on the importance of self-development (Model 1), social status (Model 2), societal contribution (Model 3), and generativity (Model 4) at follow-up when controlling for baseline levels of these life orientations. Overall, the results show that retirement is associated with statistically significant changes in life orientations. The statistical significance and direction of the effects are used to draw conclusions about whether the findings are in line with our hypotheses. However, next to statistical significance, it is important to pay attention to effect sizes, especially given the large sample size of this study. Generally, it should be noted that the sizes of the observed effects were small (i.e., Cohen’s d ≤ −0.24; Sawilowsky, 2009).

Table 3.

Results of Conditional Change Models for Explaining Changes in Life Orientations During the Retirement Transition (N = 5,034)

	Self- development	Social status	Societal contribution	Generativity
Retirement process (ref. career employment)
Postretirement work	−0.08* (0.04)	−0.13** (0.04)	−0.02 (0.04)	−0.08 (0.04)
Full retirement: voluntary	−0.19*** (0.03)	−0.24*** (0.03)	−0.15*** (0.03)	−0.16*** (0.03)
Full retirement: involuntary (orga.)	−0.12*** (0.03)	−0.17*** (0.04)	−0.08* (0.03)	−0.15*** (0.04)
Full retirement: involuntary (other)	−0.20*** (0.05)	−0.18*** (0.05)	−0.15** (0.05)	−0.19*** (0.05)
Occupational statusbaseline	0.10*** (0.01)	−0.02 (0.01)	0.03** (0.01)	0.03* (0.01)
Full-time workbaseline	0.03 (0.02)	−0.00 (0.02)	0.03 (0.02)	0.04 (0.02)
Female	0.05* (0.02)	0.05 (0.03)	0.04 (0.02)	0.07** (0.02)
Agebaseline	0.00 (0.01)	−0.00 (0.01)	0.01 (0.01)	0.01* (0.01)
Partnerbaseline	0.01 (0.02)	0.05* (0.03)	0.03 (0.02)	0.05* (0.03)
Chronic health conditionbaseline: 1	−0.04 (0.02)	−0.00 (0.02)	−0.00 (0.02)	−0.00 (0.02)
Chronic health conditionbaseline: 2+	−0.02 (0.02)	0.02 (0.03)	0.02 (0.02)	−0.01 (0.02)
Wealthbaseline	0.00 (0.00)	−0.00 (0.00)	0.00 (0.00)	−0.00 (0.00)
Self-developmentbaseline	0.57*** (0.01)
Social statusbaseline		0.51*** (0.01)
Societal contributionbaseline			0.61*** (0.01)
Generativitybaseline				0.55*** (0.01)
Constant	−0.02 (0.43)	0.20 (0.48)	−0.39 (0.45)	−0.93* (0.44)
R 2	0.42	0.30	0.41	0.34

Notes: OLS coefficients are shown; standard errors between brackets; The models were further controlled for sector of employment. Reference category for the effect of chronic health conditions is “none”.

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

Our findings provided some support for Hypothesis 1 that retirement is linked to a decreased perceived importance of self-development. Compared to continuous work in career employment, voluntary full retirement was significantly linked to a decreased importance of self-development at follow-up (Cohen’s d = −0.19). Postretirement work was also significantly linked to a decreased importance of self-development at follow-up (Cohen’s d = −0.08), as is also the case for involuntary full retirement for organizational reasons (Cohen’s d = −0.12). Additional analyses using voluntary full retirement as the reference group showed that the difference in effect sizes between voluntary full retirement and both postretirement work (b = 0.11; p < .01) and involuntary retirement for organizational reasons (b = 0.07; p < .05) were statistically significant.

Our findings also provided some support for Hypothesis 2 that retirement is linked to a decreased importance of social status. Voluntary full retirement (Cohen’s d = −0.24), postretirement work (Cohen’s d = −0.13), and involuntary retirement for organizational reasons (Cohen’s d = −0.17) were linked to a decreased importance of social status at follow-up in contrast to continuous career work. Further analyses showed that the difference in effect sizes between voluntary full retirement and postretirement work was statistically significant (b = 0.11; p < .05). The difference in effect sizes between voluntary and involuntary retirement for organizational reasons was not statistically significant (b = 0.07; p > .05).

Furthermore, we find no support for Hypothesis 3 that retirement will not change the importance individuals attach to contributing to society. Voluntary full retirement was significantly linked to a decreased importance of societal contribution at follow-up compared to continuous career work. The effect size was small (Cohen’s d = −0.15). Also, involuntary retirement for organizational reasons was linked to a relatively decreased importance of societal contribution at follow-up (Cohen’s d = −0.08), whereas the effect for postretirement work was not statistically significant. Moreover, the effect sizes between voluntary full retirement and involuntary retirement for organizational reasons differed slightly from each other (b = 0.07; p < .05).

Next, our findings do not support Hypothesis 4 that retirement will not change the importance individuals attach to generativity. In contrast to continuous career work, voluntary full retirement (Cohen’s d = −0.16), as well as involuntary retirement for organizational reasons (Cohen’s d = −0.15), was significantly associated with a decreased importance of generativity at follow-up. Postretirement work had no significant effect on changes in the importance of generativity compared to continuous career work.

As a sensitivity check, we tested whether the effect of voluntary full retirement on changes in the perceived importance of life orientations differed by occupational status and gender, because the implications of the retirement transition have been found to differ depending on these factors (Fisher et al., 2016). Workers in higher-status jobs are generally considered to be more involved in their jobs than workers in low-status jobs and are more likely to internalize the work culture (Schieman et al., 2006). As a result, the transition into full retirement may have stronger implications for their life orientations as compared to workers in lower-status jobs. We tested this relationship by including an interaction term. The results provided some evidence that the effect of voluntary full retirement on changes in the perceived importance of social status was more negative for those with a higher occupational status (b [interaction] = −0.08; p < .01; not reported in Table 3). For the other life orientations, the interaction terms between voluntary full retirement and occupational status were not statistically significant. Next, we also tested the interaction effect between full retirement and gender. As women are often considered to be more connected to nonwork roles (Loretto & Vickerstaff, 2013), their life orientations might derive more from nonwork contexts, so that the implications of retirement for their life orientations might be weaker than for men. None of the interaction terms of voluntary full retirement with gender had a statistically significant effect (not reported in Table 3) on changes in life orientations.

Discussion

A prominent expectation in the life-course literature is that major life events can trigger changes in views, attitudes, and values because they expose individuals to new social contexts with different ideas and beliefs. Changes in views, attitudes, and values have been found during transitions that mostly occur in young adulthood, such as union formation or first-time parenthood (Baxter et al., 2015). Limited attention has been paid to retirement as a driver of such changes. Instead, a central assumption in the gerontological literature seems to be that views, attitudes, and values remain relatively stable across the retirement transition. In this study, we examined whether the perceived importance of self-development, social status, societal contributions, and generativity in people’s lives changes across the retirement transition. Overall, our findings demonstrate that what individuals have valued in career employment generally remains important after the transition into retirement.

We found very small effects of retirement on the perceived importance of self-development, social status, societal contributions, and generativity. Even though the effects were statistically significant, they seem to be minor given their size (Sullivan & Feinn, 2012). This points to relatively high levels of continuity during the transition from career employment to retirement in terms of life orientations. The notion that retirees may reevaluate what they value in life is thus less pronounced than initially expected based on the exposure-based mechanism from the life-course literature. It seems that life orientations persist despite substantial changes in contexts. In turn, our small effects might also point to a process of gradually adapting and adjusting life orientations during the retirement transition. This process might already begin before retirement when some older workers disengage from work in anticipation of retirement (Damman et al., 2013) and may continue during retirement life.

Given that retirement is for a considerable share of individuals not an abrupt, complete, and voluntary exit from the labor force, this study paid attention to the complexity of the retirement process. Next to voluntary full retirees, we paid attention to retirees who are engaged in paid work and to full retirees who perceive their retirement as involuntary. Earlier literature has shown that the impact of the retirement transition on different aspects of individuals’ lives differs considerably between the different groups of retirees (Grünwald et al., 2021; van Solinge & Henkens, 2007). In this study, statistically significant differences were also observed between groups of retirees, suggesting that taking the heterogeneity of retirement processes into account is relevant. However, in terms of effect sizes, our study finds only (very) small differences between retirees in changes in life orientations. This seems to suggest that life orientations remain relatively stable, irrespective of the level of agency that workers have in their pathway to retirement.

Some limitations should be considered when interpreting these findings. First, we examine changes in life orientations using data collected 3 years apart between baseline and follow-up, which restricted us from observing attitude changes that might occur later. For attitudinal changes to become more apparent, extra time after retirement might be needed. Some full retirees might not have internalized the dominant ideas and beliefs of the retirement context yet and do so later when they have spent more time in retirement. This might be linked to the notion that retirees might transition through multiple stages when they retire (Atchley, 1976). Future research that follows older adults further into retirement might provide additional information about changing views, attitudes, and values after the retirement transition. Second, we focus on life orientations that are rather closely linked to the functions of work (Rosso et al., 2010). This limits our ability to extend our conclusions to changes in other attitudes and values. For example, some aspects of life might become more important after the retirement transition, such as relationships with family members (cf. Nuttman-Shwartz, 2007). Third, this study takes place in the Netherlands, limiting the generalizability of the findings to countries with fewer mandatory retirement regulations and less generous retirement systems.

The retirement literature has provided ample evidence that individuals adapt their behavior to compensate for the loss of career employment after the retirement transition (Wang & Shultz, 2009). This study is among the first to uncover to what extent retirees also adapt their perceptions of what their value in life. We find that despite small changes, there seems to be much continuity in the importance of key values about self-development, social status, making societal contributions, and generativity. These findings underscore the social position of today’s retirees as socially engaged and well connected to society.