Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Empowered to Stay Active: Psychological Empowerment, Retirement Timing, and Later Life Work

Empowered to Stay Active: Psychological Empowerment, Retirement Timing, and Later Life Work Motivating older employees both to prolong their working lives and to stay active even after retirement has become increas- ingly important due to rising old-age dependency ratios. Later life work—including both paid work and volunteering—has thus become an important topic for scholars and practitioners. We aim to extend research on later life work by hypothesizing that psychological empowerment at work increases not only desired and actual retirement ages but also levels of later life work. Second, we test differential effects of psychological empowerment on later life work, expecting it to be more strongly related to paid work after retirement (i.e., bridge employment) than to volunteering. Third, we suggest that the relationship between psychological empowerment and bridge employment depends on the employees’ level of physical limitations. We used data from a longitudinal panel study in Germany in which structured telephone interviews were conducted. A sample of older individuals who had retired between two waves of measurement was drawn (time lag: three years; n = 210). The results of a path analysis support the postulated mediation. Furthermore, as expected, psychological empowerment more accurately predicted bridge employment than volunteering, and physical limitations moderated the relationship between psychological empowerment and bridge employment. Lastly, additional analyses on the individual empowerment facets revealed that only the competence facet played a significant role in the proposed hypotheses. Overall, our findings suggest that psychological empowerment may help to increase older employees’ motivation to delay retirement and to stay active even after retirement. Keywords Aging workforce · Psychological empowerment · Retirement age · Activeness after retirement · Bridge employment The working-age population in many developed countries 2010). It depends not only on official retirement ages but is expected to decrease by more than one-third by 2060 also on individual contexts, needs, and preferences as well (OECD, 2019b). Many aging societies will thus have to as socio-economic factors (Wang & Shultz, 2010). Invol- face an immense financial burden to their public pension untary early retirement—that is, “a retirement that results and healthcare systems (Rouzet et al., 2019). Therefore, in from a situation with (often unexpected) employment con- countries such as Germany, for example, several economic straints” (Dorn & Sousa-Poza, 2010, p. 427)—has increased institutes have recently called for the statutory retirement during the COVID-19 pandemic and is mostly experienced age to be raised to 69 years (Holtemöller et al., 2021). How- by individuals working in low-wage sectors (Hofäcker & ever, even if such increases would provide a legal basis for Naumann, 2015; Truesdale, 2020). The main reasons people longer working lives, the question remains as to whether retire involuntarily are that they work in poor-quality jobs older employees would actually be able and motivated to and experience low employment stability during their 50 s delay retirement. Over the years, individual retirement has (i.e., they are not continuously employed). Thus, it is impor- become a complex and dynamic process (Giandrea et al., tant to note that, although statutory retirement ages are being raised, many older workers are not able to continue working, at least, because delaying retirement requires having a job * Ivana Drazic to retire from (Truesdale, 2020). At the same time, although ivana.drazic@srh.de early retirement (voluntary or involuntary) is widespread Department of Business Psychology, SRH Berlin University in many developed countries (OECD, 2019a), increasing of Applied Sciences, Ernst-Reuter-Platz 10, 10587 Berlin, numbers of older individuals are deciding to prolong their Germany Vol.:(0123456789) 1 3 I. Drazic et al. working lives (Grigoli et al., 2021). This heterogeneity raises that the effect of job satisfaction on early retirement deci- the following question: Which factors influence the retire- sions is negligible. ment timing decision of older employees who could theoreti- A construct at the person-job level that might more con- cally continue working? sistently predict retirement timing by reaching the root of Common variables that have been linked to retirement work-related experiences is psychological empowerment. intentions and, in consequence, retirement decisions are Initially, research on employee empowerment concentrated individual attributes (e.g., health), family factors (e.g., on socio-structural empowerment (see, e.g., Kanter, 1977), marital status), and socio-economic factors (e.g., current which can be defined as the distribution of power across economic conditions; for a review, see Wang & Shultz, organizational participants through empowering policies and 2010). In addition to these factors, job and organizational practices (Liden & Arad, 1996). Although this democratic factors—that is, characteristics of jobs and work organi- approach received great attention, it was found to be lim- zations and interactions between the person and job vari- ited. For example, it could not explain why some employ- ables—also play an important role (see Wang & Shultz, ees did not feel empowered even though they were working 2010). For example, a study by Vignoli et al. (2021) indi- in empowering structures or vice versa (Spreitzer, 2008). cates that higher levels of work ability and lower percep- Because of these limitations, researchers started studying tions of age stereotypes at the workplace are related to the employees’ subjective experiences of empowerment. Conger desire to work longer. Another recent study that inves- and Kanungo (1988) substantially influenced this change in tigated actual retirement decisions found that individual the view of empowerment (Spreitzer, 2008). Thomas and growth need and organizational climates that encourage Velthouse (1990) built on their work by defining empower - older workers’ learning and development were positively ment as an intrinsic motivation, further specifying its cog- related to older workers’ decisions to stay (vs. retire) nitive preconditions. Finally, based on Thomas’ and Velt- regardless of their retirement eligibility (Li et al., 2022). house (1990) work, Spreitzer (1995) proposed a definition Findings from a study by Sousa-Ribeiro et al. (2021) cor- of psychological empowerment that is now used widely in roborate the importance of an age-friendly work environ- empirical research. Spreitzer (1995) defined psychological ment and further showed that feeling positive regarding the empowerment as intrinsic task motivation manifested in future at work was positively related to the participants’ four work-related cognitions: meaning, self-determination, preferred, expected and actually retirement ages. In line competence, and impact. Meaning reflects the degree of with this finding, Fasbender et al. (2019) found that career alignment between employees’ values, beliefs, and behav- adaptability (i.e., the ability and willingness to manage iors and the corresponding requirements of their work roles one’s own career) and levels of personal growth were posi- (Spreitzer, 1995). Competence represents the extent to which tively related to late career planning among a sample of employees believe that they have the skills to perform their older workers, and that both relationships were mediated work well (Spreitzer, 2008). Self-determination comprises by occupational future time perspective. Findings from a “autonomy in the initiation and continuation of work behav- study by Watermann et al. (2023) underscore the impor- iors and processes” experienced by employees (Spreitzer, tance of both older individuals’ future time perspectives 1995, p. 1443). Impact reflects how strongly employees can and perceptions of age discrimination. The authors found influence important outcomes at work (Spreitzer, 1995). that age discrimination was negatively related to retire- Taken together, these four work-related cognitions constitute ment intentions via remaining time (a facet of occupational the “Gestalt” of psychological empowerment, which implies future time perspective). In general, job and organizational that all four cognitions are essential for experiencing psycho- factors seem to be particularly relevant because the aver- logical empowerment (Spreitzer, 1995). Thus, for instance, age employee spends many hours at work before retiring; individuals who perceive a great deal of autonomy but no in Germany, the time spent is approximately 76,500  h meaning at work will not feel psychologically empowered (Schermuly, 2016). Thus, organizations may influence (Spreitzer, 2008). In contrast, individuals who do feel psy- older workers’ retirement decisions by changing certain chologically empowered are more likely to take an active job and organizational factors. Meta-analytical findings orientation toward work (Spreitzer, 1995). suggest that job satisfaction and work involvement are the Meta-analytical findings indicate positive consequences strongest predictors of retirement planning at the person- of psychological empowerment for important work out- job level (see Topa et al., 2009). A more recent system- comes such as job satisfaction and performance (Seibert atic review indicates that high job satisfaction and high et al., 2011). With increasingly complex and dynamic busi- job control are associated with not only later retirement ness environments, research on psychological empower- intentions, but also actual retirement (Browne et al., 2018). ment—as one answer to this challenge—has blossomed over However, a meta-analysis by Topa et al. (2018) suggests the past decades (see Seibert et al., 2011), and the topic still generates considerable research interest (e.g., Malik et al., 1 3 Empowered to Stay Active: Psychological Empowerment, Retirement Timing, and Later Life Work 2021). By enabling employees to use their initiative in their experiences by reflecting how people interpret their job situ- daily work, psychological empowerment can be seen as an ation. It precedes common person-job factors such as job sat- important measure to counteract the challenges of increas- isfaction and represents a more complex antecedent of later ingly complex business environments (Lee & Edmondson, life work, comprising four distinct facets. Finally, from a 2017). However, another major challenge for many com- practical point of view, it is worthwhile to study psychologi- panies is the aging of the workforce, and little research on cal empowerment, as research suggests that it is malleable. empowerment has focused on the specific empowerment Meta-analytical findings indicate that it can be influenced of older employees and its consequences (Naghavi et al., by psychosocial organizational factors such as structural 2019). As one exception, Schermuly et al. (2017) found a empowerment, leadership, and trust in the organization positive relationship between psychological empowerment (Llorente-Alonso et al., 2023; Seibert et al., 2011). To test and desired retirement age (DRA). However, their study our hypotheses, we extended previous research on psycho- was based on cross-sectional data, and it did not investigate logical empowerment and DRA using panel data from the actual retirement age (ARA). Consistent with this limita- project “Transition and Old Age Potential” (TOP). Individu- tion, Browne et al.’s (2018) systematic review calls for more als aged 55 years or over were interviewed via telephone. We studies to examine ARA when investigating the influence of applied a two-wave design with a time lag of three years, psychosocial work conditions on retirement decisions. The investigating individuals who had retired between two waves first part of the present study addresses this research gap by of measurement (n = 210). In the next paragraph, we first examining whether empowered employees not only desire provide information on the German retirement system to to retire later but indeed retire later. contextualize our study. Second, we conceptualize retire- The second part of the study investigates the conse- ment from a psychological perspective and introduce the quences of psychological empowerment on activeness after temporal model of retirement as an overarching framework. retirement. We hypothesize that psychological empower- Lastly, we derive our hypotheses. The conceptual research ment during working life is positively related to levels of model for all hypotheses is depicted in Fig. 1. later life work. Later life work can be defined as “work activities of older adults just before and beyond normal retirement age” (Wöhrmann, 2021, p. 12), including both Background paid work and volunteering. Based on continuity theory (Atchley, 1989), we expect psychological empowerment to The German Retirement System be related to later life work and to better predict engage- ment in paid work after retirement than engagement in vol- Germany has a long history of institutionalized early retire- unteering. Furthermore, as suggested by Beehr and Bennett ment (i.e., the act of retiring before the statutory retirement (2015), we investigate boundary conditions of the relation- age), which was intended, inter-alia, to promote organiza- ship between psychological empowerment and the decision tional restructuring (Trampusch, 2005). Not until the 1990s to engage in bridge employment. We suggest that the posi- did German policymakers gradually begin to close pathways tive relationship between psychological empowerment and to early retirement (Hofäcker & Naumann, 2015). In the bridge employment is moderated by employees’ physical 2000s, Germany further promoted a paradigm shift from limitations. Finally, to advance research on psychological early exit to working longer by introducing labor market empowerment, we respond to calls by Spreitzer (2008) and programs for older employees and a stepwise increase of investigate the roles of the individual facets of psychological the statutory retirement age from 65 to 67 years between empowerment within exploratory analyses. 2011 and 2029 (BMAS, 2020; Hofäcker & Naumann, 2015). In summary, we aim to contribute to the literature on later In general, the current German pension program allows life work by testing whether employees who feel more psy- for flexible retirement timing: Individuals can retire before chologically empowered (a) not only desire to retire later but reaching statutory retirement age by accepting monthly pen- indeed retire later, (b) are more likely to engage in paid work sion claims, which are to be reduced by 0.3% until statu- after retirement than employees who feel less psychologi- tory retirement age is reached. On the other hand, surpass- cally empowered, and (c) are more likely to engage in paid ing statutory retirement age is also possible and even gets work after retirement than in volunteering. Furthermore, we rewarded with a surcharge of 0.5% per month (Czepek & want to shed light on boundary conditions of the proposed Weber, 2015). Furthermore, retirees who do not retire early relationships by proposing physical limitations as a moder- can continue working without income restrictions (DRV, ating variable. We also examine the role of the individual 2020), which further establishes financial incentives to facets of psychological empowerment. From a conceptual prolong working life. Concerning the sample of the present point of view, these investigations are important, because study, the statutory retirement ages for claiming full state psychological empowerment reaches the root of work-related pension ranged from 65 to 66 years. 1 3 I. Drazic et al. Fig. 1 Conceptual research model 2010). This model describes the retirement process heuris- The Temporal Process Model of Retirement tically, capturing its dynamic and complex nature. It outlines three sequential and gradually unfolding retirement stages: Retirement has been conceptualized in many different ways (1) retirement planning, (2) retirement decision making, and in the literature, with the lowest common denominator being (3) retirement transition and adjustment. In the retirement that it is not a single event, but rather a process that takes planning phase, older people begin to envision their still place over a period of time (Shultz & Wang, 2011). For distant retirement lives and make initial plans and prepara- a long time, the focus in retirement research has been on tions for retirement. As retirement nears, people enter the physical and financial factors of retirement (Shultz & Wang, decision-making phase. In analogy to the conceptualization 2011). However, over the last decades, an increasing amount of retirement as decision making, in this phase, individuals of research has applied psychological perspectives on the consider and weigh concrete alternatives to reach conclu- retirement process. Wang and Shi (2014) identified three sions about their retirement. The final phase, retirement tran- psychological conceptualizations of retirement based on sition and adjustment, deals with daily changes in retirement their literature review: (1) retirement as decision making, life after the transition from full-time worker to retiree. Daily (2) retirement as an adjustment process, and (3) retirement activities in retirement include leisure activities, caregiving, as a career-development stage. In the conceptualization of and later life work (i.e., paid and/or unpaid forms of work). retirement as decision making, older workers consider and As well as describing the sequential process of retirement, weigh information that they have about themselves, their the temporal process model identifies four factors influenc- work, and their private life to make decisions about their ing the retirement process that we have already mentioned in retirement. Retirement as an adjustment process focuses on the Introduction: individual attributes, family factors, socio- the transition from employment to retirement and the indi- economic factors, and job and organizational factors. vidual development of older peoples’ lives after retirement. The first part of the present study examines the relation- Finally, retirement as a career-development stage stresses ship between psychological empowerment and retirement older people’s ongoing potential to change their own career timing (i.e., desired and actual retirement ages) and thus paths after retirement (Wang & Shi, 2014). These concep- focuses on the first two phases of retirement: retirement tualizations are not completely separate from each other but planning and retirement decision making. The second part of can be viewed as parts of an overarching process (see, e.g., the study examines the relationship between psychological Wöhrmann, 2021). empowerment during working life and later life work (i.e., One framework that helps to integrate the different con- bridge employment and volunteering). Thus, this part of the ceptualizations of retirement is the temporal process model study focuses on the third phase of the retirement process: of retirement (e.g., Wang & Shi, 2014; Wang & Shultz, 1 3 Empowered to Stay Active: Psychological Empowerment, Retirement Timing, and Later Life Work retirement transition and adjustment. Following suggestions the behavior, subjective norms, and perceived behavioral about the correspondence between certain retirement con- control. First, we believe desires constitute a valid proxy ceptualizations and certain theories (see Wang & Shultz, for intentions, as desires have been shown to mediate the 2010), we conceptualize retirement as an adjustment and relationship between attitudes, norms and perceived behav- a decision-making process and focus on continuity theory ioral control, and intentions (see Perugini & Bagozzi, 2001). (Atchley, 1989) and the theory of planned behavior (Ajzen, Second, we argue that in the case of retirement, perceived 1991), respectively. behavioral control plays a particularly important role: Retire- ment timing is, to a great degree, dependent on legal regula- Retirement Planning and Decision Making: The tions. Thus, in countries where pension programs allow for Relationship between Psychological Empowerment, flexible retirement timing—such as Germany—the desire to Desired Retirement Age, and Actual Retirement Age retire should have a major influence on ARA. Based on these arguments, we hypothesize: As suggested by the temporal process model of retirement (e.g., Wang & Shultz, 2010), retirement typically starts with Hypothesis 1 Higher levels of psychological empowerment a phase in which older people gather and weigh information lead to higher levels of DRA, which in turn result in higher about themselves, their work, and their private life to define levels of ARA. when and how to retire. We argue that older individuals use their psychological empowerment experiences as a source of Retirement Transition and Adjustment: The Relationship information from the job and work field and that empowered between Psychological Empowerment and Work older individuals want to retire later than less empowered after Retirement individuals. We base this hypothesis on continuity theory (Atchley, 1989), which is in line with the conceptualization Even for individuals with high levels of psychological of retirement as an adjustment process. Continuity theory empowerment, there comes a time when it is reasonable to focuses on adaptation to change in later life stages, and retire—for example, for financial reasons. However, even it states that people strive to maintain internal continuity after retirement, older people can continue to experience (i.e., their self-concept) and external continuity (i.e., social psychological empowerment by engaging in work after structures) across their lifespan to prevent experiences of retirement. Work after retirement can be paid or nonpaid harmful disruption. A central premise of continuity theory (see definition of later life work; Wöhrmann, 2021). We is that individuals maintain continuity by engaging in types believe that psychological empowerment during working life of activities, behaviors, and relationships similar to those predicts levels of later life work, especially paid work after they have previously experienced. We argue that the more retirement, which is often discussed as bridge employment psychological empowerment older individuals experience in in the literature (Shultz, 2003). their preretirement lives, the more they will wish to continue Bridge employment refers to the pattern of labor force this experience in order to maintain continuity. We hypoth- participation of older employees that spans the period esize that this desire for continuity will be reflected in the between their career jobs and complete withdrawal from desire to delay retirement, as remaining in the workforce the labor force (Shultz, 2003; Wang et al., 2008). A simpler allows them to continue to experience psychological empow- definition of bridge employment is any kind of paid work erment. Previous research has corroborated this assumption. after retirement (e.g., Shultz, 2003). It can be part-time, Schermuly et al. (2017), for example, found that the more full-time, or self-employed, career- or noncareer-related, psychologically empowered employees felt, the later they with the same or a different employer (for a taxonomy, see wanted to retire. Beehr & Bennett, 2015), and it can be dynamic (individuals With the present study, we aim to extend previous findings can enter and leave the workforce several times after retire- on the relationship between psychological empowerment ment; Wang et al., 2009). The most typical categorization of and the delay of retirement by investigating whether psy- bridge employment is that of career bridge employment (i.e., chologically empowered employees not only desire to retire employment in the same industry or field as the career job) later but actually retire later. As stated above, retirement in contrast to bridge employment in a different field (Shultz, can be conceptualized as a motivational decision-making 2003; Wang et al., 2008). process (Shultz & Wang, 2011). In motivational processes, Antecedents of bridge employment can be categorized intentions are the most proximal determinants of actual into personal characteristics (e.g., health), contextual fac- behavior (see theory of planned behavior; Ajzen, 1991). tors (e.g., economy), and organizational factors (e.g., pol- According to the theory of planned behavior, an intention icy incentives; see Beehr & Benett, 2015). It is assumed reflects the extent to which an individual is motivated to per - that organizational factors are indirectly related to bridge form a specific behavior, and it is driven by attitudes toward employment, mediated by subjective attitudes and feelings 1 3 I. Drazic et al. toward the organization. Indeed, cognitions relating to work experiences. We argue that bridge employment constitutes and the organization have been linked to bridge employ- the most proximal setting in which retired individuals can ment. For example, a study by Garcia et al. (2021) found continue to experience psychological empowerment. that organizational support increased bridge employment intentions only among older workers who endorsed a rela- Hypothesis 2a Psychological empowerment before tional psychological contract (vs. a transactional psychologi- retirement is positively related to engagement in bridge cal contract). Other recent studies suggest that experiences employment. of work meaningfulness mediate the relationships between workplace age discrimination, workplace incivility, rela- To further corroborate continuity theory (Atchley, 1989), tional job crafting (i.e., interpersonal crafting at work), we hypothesize that psychological empowerment before and generativity opportunities and bridge employment retirement is more strongly related to bridge employment intentions, respectively (Peng, 2022; Peng et  al., 2020). than to volunteering after retirement. Although a recent Another study focusing on employee’ cognitions found that study has shown that psychological empowerment can be employees with higher outcome expectations regarding stimulated in the volunteering context (Traeger & Alfes, post-retirement work (e.g., they had greater expectations 2019), we believe the context of the original empowerment that post-retirement work would enable them to stay physi- experience to be crucial when it comes to continuity. In line cally and mentally healthy) were more likely to engage in with this argument, a recent longitudinal study by Vangen post-retirement work planning (Wöhrmann et al., 2013). et al. (2021) found that previous engagement in voluntary These findings substantiate the assumption that it is not the work among 62- to 75-year-olds in Norway was strongly organizational factors per se but the subjective interpreta- associated with voluntary work 10 years later. We agree with tions of workers that determine how likely they are to engage Vangen et al. (2021), who state that “being active in older in bridge employment. age may depend on earlier activity patterns” (p. 2). Thus, We hypothesize that psychological empowerment plays we believe that, if psychological empowerment has been such an important mediating role between organizational experienced in the work context, it is more likely that indi- factors and the bridge employment decision making by viduals will seek opportunities in this life arena to continue capturing employees’ subjective experiences of their work. experiencing empowerment. Compared to cognitions of outcome expectations (see Wöhr- mann et al., 2013) or experienced meaningfulness of work Hypothesis 2b Psychological empowerment before retire- (Peng, 2022; Peng et al., 2020), psychological empower- ment is more strongly related to bridge employment than to ment is a broader construct focusing on the present state volunteering after retirement. of employees’ interpretations of their work and therefore may represent a more immediate predictor of bridge employ- Finally, drawing from the model of the bridge employ- ment. Furthermore, in contrast to more general job-related ment decision-making process by Beehr and Bennett (2015), constructs that have been identified as antecedents of bridge we suggest that the relationship between psychological employment, such as job satisfaction, commitment, intrinsic empowerment and bridge employment is influenced by motivation, or work stress (see, e.g., Beehr & Bennett, 2015; boundary conditions. Beehr and Bennett (2015) built on the Wang et  al., 2008), psychological empowerment allows process model of retirement (Wang & Shultz, 2010), add- for more fine-grained insights, as it consists of four facets ing situational constraints (e.g., unexpected factors such as that might serve as potential leverages to promote bridge additional expenses or health problems) as boundary condi- employment. tions to the model. We assume that physical limitations are From a theoretical point of view, we draw from continuity an important boundary condition, as poor health is one of the theory (Atchley, 1989), which we have introduced above. most important reasons for an early exit from the labor force Atchley (1989) argues that “adults employ concepts of their (see, e.g., van Rijn et al., 2014; Vanajan et al., 2020) and past to conceive of their future and structure their choices has been shown to interact with worker characteristics such in response to the changes brought about by normal aging” as age and race (see a review by Fisher et al., 2016). High (p. 183). Retirement constitutes such a normal change. This physical limitations reflect the individual’s inability to per - striving for continuity in one’s self-concept leads older form daily activities such as sitting for more than two hours workers to envision themselves being and acting rather or lifting something, implying a “limited capacity to meet similarly in retirement as in the present moment (Feldman, the requirements of core social, familial, and occupational 2013). Thus, following continuity theory (Atchley, 1989), roles” (Gayman et al., 2008, p. 219). From a theoretical per- we suggest that older workers who have experienced psycho- spective, continuity theory (Atchley, 1989) might explain the logical empowerment during working life will choose cir- potential boundary effect of physical limitations on bridge cumstances after retirement that provide them with similar employment decision making. Continuity theory (Atchley, 1 3 Empowered to Stay Active: Psychological Empowerment, Retirement Timing, and Later Life Work 1989) states that “external continuity is usually an effective data (i.e., 2 SD below or above the mean). Missing values adaptive strategy but can sometimes be maladaptive” (p. were generally accounted for by full-information maximum 189). According to Atchley (1989), external continuity (e.g., likelihood (FIML) estimation or listwise deletion, depending continuing to work after retirement) becomes maladaptive on the model used (see section on statistical analyses below). when the physical and mental abilities required for conti- Participants at T1 were between 54 and 65  years old nuity are seriously impaired. We hypothesize that the rela- (M = 61.97, SD = 1.95); 50% were female, and 77.1% indi- tionship between psychological empowerment and bridge cated being in a committed relationship. According to the employment depends on the level of physical limitations, International Standard Classification of Education 97 clas- such that the relationship between psychological empower- sification, most participants (58.6%) had a high level of edu- ment and bridge employment can only unfold when physical cation (up to a doctoral degree or equivalent). At T1, partici- limitations are low. Consequently, we argue that older people pants worked 37.3 h per week on average (SD = 12.2). Ten are more likely to use adaptive strategies—that is, to retire percent were blue-collar workers, 51.4% white-collar work- when sufficient physical resources for continuing to work ers, 23.3% civil servants, 13.3% self-employed, and 1.4% are no longer available—than to force external continuity, supported their family business. Most participants worked even though they might report high levels of psychological either in small companies (less than 50 employees; 36.7%) or empowerment. large companies (more than 250 employees; 37.1%). Table 1 summarizes the sample characteristics measured at T1. Hypothesis 2c The positive relationship between psychologi- cal empowerment and bridge employment is moderated by Measures physical limitations, such that the relationship is stronger for retirees with low levels of physical limitations and weaker The interview guide for the TOP study was developed and for retirees with high levels of physical limitations. pre-tested by the German Federal Institute for Population Table 1 Sample characteristics at T1 Method n/M %/SD Participants and Procedure Gender Female 105 50 We used data from the TOP study, a longitudinal, popu- Male 105 50 lation-representative panel study, carried out on behalf Age M = 61.97 SD = 1.95 of the German Federal Institute for Population Research Working hours M = 37.29 SD = 12.20 (Mergenthaler et  al., 2017). It investigated older adults’ Relationship status potential in Germany and their transition from working Committed relationship 162 77.14 life to retirement. Standardized and computer-assisted tel- No committed relationship 48 22.86 ephone interviews were conducted by trained interviewers. Education (ISCED) To ensure random sampling of households with a landline Low 4 1.91 number, the Gabler-Häder design was used, which consid- Middle 83 39.53 ers listed and unlisted numbers equally (Sackreuther et al., High 123 58.57 2016). For our research question, we used data from the first Position wave (T1; January–April 2013) and the second wave (T2; Blue-collar 21 10.00 November 2015–February 2016). At T1, a total of 5,002 White-collar 108 51.43 respondents aged 55–70 years completed the interview, 50% Civil servants 49 23.33 of whom also completed the interview at T2 (Mergenthaler Self-employed 28 13.33 et al., 2017). Given our research question, we investigated Support in family business 3 1.43 only respondents who had retired between T1 and T2; thus, Organizational size individuals who (a) were working part- or full-time at T1 Small (less than 50 employees) 77 36.67 and were not yet retired (n = 997) and (b) were retired at Medium-sized (between 50 and 250 37 17.62 T2 (n = 222). We excluded 12 participants who had missing employees) values on the empowerment items, showed logical incon- Large (more than 250 employees) 78 37.14 sistencies regarding their retirement year, or did not indicate Household net income (in euros) M = 2574.82 SD = 2970.37 their retirement year at all. This resulted in a final sample of 210 participants. Furthermore, we replaced the DRA of n = 210; a b c d eight participants with missing values because of extreme n = 203; n = 209; n = 192; n = 207 1 3 I. Drazic et al. Research in cooperation with several universities. Unless reliability, it is increasingly recommended to replace Cron- otherwise indicated, we recoded all items so that higher val- bach’s α with alternative coefficients. This is because of its ues indicated higher scale expressions. unrealistic assumptions, which oftentimes lead to underes- timated reliability scores (McNeish, 2018). Omega coeffi- ARA (T2) cients, for example, are more appropriate measures of reli- ability in case not all items are equally strongly associated ARA was measured with a single item asking participants with the underlying construct (Peters, 2014). This lack of to indicate the year they retired. We then subtracted the par- homogeneity becomes even more likely with shorter scales ticipants’ year of birth from their retirement year. (Berger, 2019). Revelle’s omega (Revelle & Zinbarg, 2009) for the three-faceted empowerment construct in the present DRA (T1) study was 0.61. Although this value is not optimal, it is not surprising: Even in noninterview contexts, lower reliability Participants were asked to indicate the year they would like scores for the empowerment construct are not unusual and to retire. Again, we calculated the difference scores using the deemed acceptable because of the construct’s first-order participants’ year of birth. multidimensionality (Spreitzer, 1995). Schermuly et  al. (2017), who used the same items but with a larger sample Bridge Employment (T2) of the TOP study, conducted exploratory and confirmatory factor analyses, both of which suggested a single empower- Participants were asked whether they were currently engaged ment factor after inspecting the scree plot, Eigenvalues, and in paid work (they were prompted to think of any kind of fit indices. Based on these arguments, we decided to use this paid work after retirement). Of the 210 participants, 78 interview-adapted empowerment scale and its mean score (37.1%) indicated that they were engaged in bridge employ- for subsequent statistical analyses. ment; one participant did not provide the information. Physical Limitations (T2) Volunteering (T2) The level of physical limitations was measured with the Volunteering was measured with the question “Have you question “During the last four weeks, how often were you engaged in any volunteer or honorary activity in the last limited in your activities because of physical problems?” three months, e.g., participating in an association, initiative, Participants assessed this item on a 4-point Likert scale or group?” Of the 210 participants, 100 (47.6%) reported ranging from 1 (very often) to 4 (never). We also recoded volunteering. this item so that higher values reflected greater frequency. Psychological Empowerment (T1) Controls Psychological empowerment was measured with the same Gender (T1) items that were used by Schermuly et al. (2017). The TOP research group adapted the items from Spreitzer’s (1995) In line with other research on bridge employment and retire- original 12-item scale. Especially in the telephone interview ment (see reviews by Beehr & Bennett, 2015 and Fisher context, scales that are designed to be self-administered have et al., 2016), we controlled for gender. been deemed too long and complex (Hughes et al., 2004). Thus, the TOP research group decided to use a selection of Spreitzer’s (1995) empowerment scale and to omit the Internal Locus of Control (T1) impact facet altogether, as prior research suggests a close relationship with the self-determination facet (e.g., Kraimer Meta-analytical findings suggest a strong relationship et al., 1999; Schermuly et al., 2013). Meaning and self-deter- between internal locus of control (a core self-evaluation mination were measured with single items: “My work is very trait) and psychological empowerment (Seibert et al., 2011). important to me” and “During work, I have the opportunity Core self-evaluation traits in turn have been found to be to make decisions autonomously,” respectively. Competence positively related to retirement variables such as retirement was measured with two items: “I face work problems with preparation (Zaniboni et al., 2021). On a 4-point Likert scale calmness because I can always rely on my competencies” ranging from 1 (strongly agree) to 4 (strongly disagree), par- and “Whatever happens in my job, I will master it.” Partici- ticipants were asked to assess the following two statements: pants assessed the items on a 4-point Likert scale ranging “I am in control of my life” and “If I make an effort, I will from 1 (strongly agree) to 4 (strongly disagree). Concerning succeed.” 1 3 Empowered to Stay Active: Psychological Empowerment, Retirement Timing, and Later Life Work Public Sector Employment (T1) bridge employment and volunteering; see Rosseel, 2014). This model was just identified; thus, examining model fit We controlled for employees in the public sector, as they indices to assess model fit is not meaningful. The control have other retirement conditions as private employees in variables gender, locus of control, public sector employee, Germany. For example, the German pension law imposes and equivalized net income of households were included as income restrictions on employees from the public sector who predictors of ARA; in bridge employment and volunteering, have retired (BMAS, 2019). This might alter incentives for we further controlled for temporal latency. bridge employment. Participants in the present study who indicated being civil servants were coded as 1, whereas all other forms of employment were coded as 0. Results Equivalized Net Income of Households (T1) Preliminary Analysis We see pay as a resource provided by the organization, and Table 2 shows descriptive results and intercorrelations. On access to resources has been shown to be positively asso- average, participants wanted to retire at 64.12 years of age ciated with perceptions of psychological empowerment (SD = 2.18) and retired at 63.93 years (SD = 1.83). Psycho- (Seibert et al., 2011). On the other hand, it is well known logical empowerment was positively correlated with all that a person’s financial situation plays an important role dependent variables except for bridge employment and vol- in retirement decisions (see Fisher et al., 2016). We used unteering. Furthermore, psychological empowerment was the modified Organisation for Economic Co-operation and positively correlated with locus of control, gender was posi- Development (OECD) equivalence scale, which adjusts tively correlated with ARA, and volunteering was positively disposable monthly household income by the number and correlated with public sector employment. age of household members (Sackreuther et al., 2016). The median OECD equivalized monthly net income of house- Hypothesis Testing holds was EUR 2076.5. Hypothesis 1 states that higher levels of psychological Temporal Latency (T2) empowerment lead to higher levels of DRA, which in turn result in higher levels of ARA. The observed indirect effect Whether or not someone engages in occupation-related supports this hypothesis (unstandardized indirect effect: activities after retirement may depend on how long ago one 0.82, 95% CI [0.37; 1.24], p < 0.001). Psychological empow- has retired (Fasbender et al., 2016). To calculate temporal erment signic fi antly predicted DRA (β = 0.27, 95% CI [0.73; latency, we subtracted the year of retirement from the sec- 2.20], p < 0.001), which in turn predicted ARA (β = 0.65, ond measurement year. On average, temporal latency was 95% CI [0.44; 0.68], p < 0.001). The direct effect of psycho- 1.75 years (SD = 0.82). logical empowerment on ARA was not significant (β = 0.07, 95% CI [− 0.27, 0.87], p = 0.288), which indicates full medi- Statistical Analyses ation. None of the control variables significantly predicted the ARA (gender, β = 0.07, 95% CI [− 0.11, 0.64], p = 0.174; All hypotheses were tested with the statistical software R locus of control, β = − 0.02, 95% CI [− 0.41, 0.29], p = 0.762; Studio. For Hypothesis 1 (mediation analysis), we conducted public sector employment, β = − 0.07, 95% CI [− 0.75, 0.22], a path analysis using the maximum likelihood estimator with p = 0.246; equivalized net income of households, β = − 0.02, the R package lavaan (Rosseel, 2012). We further specified 95% CI [− 0.00, 0.00], p = 0.838). The model accounts for 5,000 bootstrap samples, a 95% confidence interval (CI), and 7.4% of the variance of DRA and 46.4% of the variance of FIML estimation for missing values. According to Hooper ARA. Following recommendations by Becker et al. (2015), et al. (2008), the model testing Hypothesis 1 showed good fit we also ran the mediation analysis without control variables, indices, χ (4) = 5.571, p = 0.234, CFI = 0.999, TLI = 0.998, which did not change the effects. RMSEA = 0.044, SRMR = 0.032. To test Hypotheses 2a and In Hypothesis 2a, we assumed psychological empower- 2c, we conducted a hierarchical binary logistic regression, ment at T1 to be positively associated with bridge employ- first entering the control variables (Model 1), then entering ment at T2. The results of the binary logistic regression are the main variable (Model 2), and lastly, entering the moder- presented in Table 3. The control variables entered in step ating variable (i.e., physical limitations). For Hypothesis 2b, 1 did not predict engagement in bridge employment. Psy- we conducted a path analysis using the diagonally weighted chological empowerment entered in step 2 significantly pre - least squares (DWLS) estimator to enable the simultane- dicted bridge employment, indicating that individuals with ous inclusion of both categorical dependent variables (i.e., higher levels of psychological empowerment were more 1 3 I. Drazic et al. likely to engage in bridge employment. Model 2 accounts for 4% of the variance in bridge employment. In Hypothesis 2c, we assumed that employees’ physi- cal limitations would moderate the relationship between psychological empowerment and bridge employment. Our data support this hypothesis (see Model 3 in Table 3). The interaction between psychological empowerment and phys- ical limitations was significantly related to the likelihood of engaging in bridge employment. Simple slope analyses revealed that the relationship between psychological empow- erment and bridge employment became stronger when levels of physical limitations were low (i.e., 1 SD below the mean; B = 1.42, p = 0.008, CI [0.36, 2.47]). When physical limita- tions were high (i.e., 1 SD above the mean), the relationship between psychological empowerment and bridge employ- ment was nonsignificant (B = − 0.07, p = 0.910, CI [− 1.22, 1.09]). Figure 2 shows the interaction between psychological empowerment and physical limitations. In Hypothesis 2b, we assumed psychological empower- ment at T1 to be more strongly related to bridge employ- ment than volunteering at T2. The path analysis supports this hypothesis: Psychological empowerment was significant in predicting bridge employment, indicating that individu- als with higher levels of psychological empowerment were more likely to engage in bridge employment. In contrast, psychological empowerment did not significantly predict volunteering. Concerning the set of control variables, none of the variables significantly predicted bridge employment, but volunteering was significantly predicted by public sector employment. The model accounts for 9% of the variance in bridge employment and 12% of the variance in volunteering. The results of the path analysis are summarized in Table 4. We also ran this model without control variables, which slightly decreased the statistical significance of psychologi- cal empowerment, such that it only significantly predicted bridge employment based on a 90% CI [0.05, 0.80]. Additional Analyses To obtain more fine-grained insights, we reran our analy- ses using the three dimensions of psychological empower- ment as individual predictors. For that, we calculated a mean score of the two items measuring the competence facet (self-determination and meaning were measured with single items). Regarding Hypothesis 1, we found that only the competence facet significantly predicted the DRA, which in turn predicted the ARA. The observed indirect effect was significant (unstandardized indirect effect: 0.22, 95% CI [0.04; 0.42], p = 0.025), while t he direct effect was nonsignificant, indicating full mediation. None of the control variables (i.e., gender, locus of con- trol, public sector employee, or equivalized net income 1 3 Table 2 Means (M), standard deviations (SD), and intercorrelations among study variables Variable (unit/scale) M SD 1 2 3 4 5 6 7 8 9 10 1. ARA (years) 63.93 1.83 2. DRA (years) 64.12 2.18 0.67** 3. Bridge employment (0 = no, 1 = yes) 0.37 0.48 0.15* 0.20** 4. Volunteering (0 = no, 1 = yes) 0.48 0.50 − 0.03 − 0.07 − 0.12 5. Psychol. empowerment (scale 1–4) 3.56 0.40 0.23** 0.26** 0.12 -0.06 6. Physical limitations (scale 1–4) 1.68 0.86 − 0.21** − 0.18* − 0.17* − 0.16* − 0.01 7. Gender (1 = female, 2 = male) 1.50 0.50 0.14* 0.11 − 0.10 − 0.06 0.11 0.02 8. Locus of control (scale 1–4) 3.31 0.56 − 0.03 − 0.04 0.02 − 0.01 0.16* − 0.00 − 0.06 9. Public sector (0 = no, 1 = yes) 0.23 0.42 − 0.09 − 0.02 − 0.13 0.24** − 0.03 − 0.10 0.03 − 0.03 10. Equiv. net income (in EUR) 2574.82 2970.37 0.06 0.12 0.00 0.08 0.11 − 0.12 − 0.03 0.00 0.07 11. Latency (years) 1.75 0.82 − 0.11 0.11 0.08 0.07 0.05 0.06 − 0.07 − 0.07 − 0.01 0.14* *p < 0.05. **p < 0.01 a b c n = 210; n = 194; n = 209; n = 207 ARA actual retirement Age, DRA desired retirement age Empowered to Stay Active: Psychological Empowerment, Retirement Timing, and Later Life Work 1 3 Table 3 Results of the binary logistic regression predicting bridge employment Variable Model 1 Model 2 Model 3 B SE B p 95% CI OR B SE B p 95% CI OR B SE B p 95% CI OR Intercept − 0.71 1.00 0.478 [− 2.71, 1.24] − 3.23* 1.64 0.049 [− 6.53, − 0.09] − 0.19 0.23 0.398 [− 0.64, 0.25] Step 1: Control variables  Gender (1 = female, 2 = male) − 0.38 0.30 0.198 [− 0.96, 0.20] 0.68 − 0.45 0.30 0.136 [− 1.04, 0.14] 0.64 − 0.50 0.31 0.108 [− 1.12, 0.11] 0.60  Locus of control 0.06 0.26 0.831 [− 0.46, 0.58] 1.06 − 0.04 0.27 0.885 [− 0.57, 0.49] 0.96 − 0.10 0.28 0.736 [− 0.65, 0.47] 0.91  Public sector − 0.62 0.37 0.092 [− 1.36, 0.08] 0.54 − 0.59 0.37 0.109 [− 1.34, 0.11] 0.55 − 0.70 0.38 0.066 [− 1.48, 0.03] 0.50  Net income − 0.00 0.00 0.957 [− 0.00, 0.00] 1.00 − 0.00 0.00 0.772 [− 0.00, 0.00] 1.00 − 0.00 0.00 0.464 [− 0.00, 0.00] 1.00  Latency 0.18 0.18 0.315 [− 0.17, 0.54] 1.20 0.18 0.18 0.332 [− 0.18, 0.54] 1.20 0.25 0.19 0.201 [− 0.13, 0.62] 1.28 Step 2: Main variable  Psychological empowerment 0.81* 0.41 0.049 [0.03, 1.64] 2.24 0.77 0.43 0.072 [− 0.05, 1.63] 2.15 Step 3: Interaction  Physical limitations − 0.52* 0.20 0.012 [− 0.93, − 0.14] 0.60  Psychological empower- − 0.96* 0.48 0.045 [− 1.92, − 0.01] 0.38 ment × Physical limitations Model Fit:  − 2 log likelihood 264.156 260.055 247.578  Δ − 2 log likelihood (Δdf) 4.101*(1) 12.477**(2)  Pseudo R2 0.02 0.04 0.08 In Model 3, all continuous predictors were mean-centered n = 205 SE Standard error, CI confidence interval, OR odds ratio *p < 0.05, **p < 0.01 I. Drazic et al. Fig. 2 Moderation of the relationship between psycho- logical empowerment and bridge employment by physical limitations Table 4 Results of the path analysis predicting bridge employment and volunteering Variable Bridge employment Volunteering B SE B β p 95% CI B SE B β p 95% CI Control variables  Gender − 0.27 0.19 − 0.13 0.156 [− 0.68, 0.06] − 0.15 0.19 − 0.07 0.437 [− 0.53, 0.21]  Locus of control − 0.03 0.18 − 0.02 0.877 [− 0.35, 0.36] 0.02 0.17 0.01 0.917 [− 0.33, 0.32]  Public sector − 0.36 0.24 − 0.15 0.132 [− 0.90, 0.03] 0.72** 0.23 0.29** 0.002 [0.31, 1.24]  Net income − 0.00 0.00 − 0.03 0.915 [− 0.00, 0.00] 0.00 0.00 0.10 0.547 [− 0.00, 0.00]  Latency 0.11 0.12 0.09 0.350 [− 0.12, 0.35] 0.10 0.12 0.08 0.400 [− 0.13, 0.33] Main variable  Psychological empowerment 0.51* 0.25 0.19* 0.040 [0.04, 1.00] − 0.19 0.25 − 0.07 0.444 [− 0.67, 0.31] N = 205. For gender, 1 = female, 2 = male *p < 0.05, **p < 0.01 CI confidence interval of households) significantly predicted the ARA. Figure 3 Discussion shows the model with standardized parameter estimates. Regarding Hypothesis 2, the results showed that none Especially in aging societies, successfully motivating of the single facets of psychological empowerment sig- older employees to extend their working lives is essen- nificantly predicted bridge employment or volunteering. tial. Using a sample of a longitudinal panel study in Ger- As in the model with psychological empowerment as a many, we found that psychologically empowered employ- unitary construct, none of the control variables signifi- ees not only desired to retire later but indeed retired later cantly predicted bridge employment, and only public (Hypothesis 1). Second, we tested whether psychological sector employment significantly predicted volunteering empowerment predicts engagement in bridge employment (β = 0.30, 95% CI [0.29; 1.22], p = 0.002). All additional (Hypothesis 2a). Our findings also support this hypothesis. analyses are available upon request from the first author. 1 3 Empowered to Stay Active: Psychological Empowerment, Retirement Timing, and Later Life Work Fig. 3 Results of the Mediation Hypothesis with the Individual Facets of Psychological Empowerment as Predictors. Note. *p < 0.05. **p < 0.01. All displayed coefficients are standardized. Control variables are omitted in the figure. Dashed line indicates full mediation Furthermore, we hypothesized that psychological empow- literature, our findings contribute to the literature in three erment better predicts bridge employment than volunteer- unique ways. First, they extend findings from Schermuly ing (Hypothesis 2b). Psychological empowerment was et al. (2017), who found that psychologically empowered indeed significant in predicting bridge employment while employees desired to retire later. By including ARA, we showing a nonsignificant relationship to volunteering; respond to calls for more research on retirement outcomes however, effect sizes were small. Finally, we found that that goes beyond intentions (see Browne et al., 2018). Sec- physical limitations moderated the relationship between ond, we transferred psychological empowerment to the psychological empowerment and bridge employment, such retirement context, suggesting that retirement age, along that the relationship was stronger when physical limita- with constructs such as turnover intention and organizational tions were low and nonexistent when physical limitations commitment, is another important outcome of psychologi- were high. In the next section, we will interpret these find- cal empowerment in the organizational context. Third, our ings in light of existing theory and empirical studies. results support the theory of planned behavior (Ajzen, 1991) as well as its theoretical extension that proposes intentions Theoretical Implications (in our case, DRA) as a mediating variable (see Perugini & Bagozzi, 2001). DRA and ARA were highly correlated in In general, our findings regarding all four hypotheses sup- our sample (r = 0.67), which implies that the desire to retire port the process model of retirement (e.g., Wang & Shultz, at a certain time is realizable. This might be due to the Ger- 2010) in its assumption that job and organizational factors man retirement system, which allows for flexible timing. In might act as antecedents of retirement decisions. Our find- Sweden, where there is also no fixed retirement age, panel ings contribute to the model first by adding psychological studies have found similarly strong relationships between empowerment as a concrete antecedent at the job and organi- DRA and ARA (Örestig et al., 2013; Solem et al., 2016). In zation level. Second, as psychological empowerment was the present study, however, DRA was slightly higher than related not only to retirement timing but also to the bridge ARA, indicating that the participants retired earlier than they employment decision making, our findings suggest that the desired. Solem et al. (2016) identified conditions, such as same job and organizational factors experienced during poor health and education, that increased the risk of retiring working life can influence retirement decisions across phases earlier than preferred. More research is needed to further and have a lasting effect on the retirees’ lives. investigate boundary conditions of the relationship between The results related to Hypothesis 1 are in line with DRA and ARA, as well as the psychological consequences previous findings showing that person-job variables are of not retiring at one’s DRA. important factors for the DRA and the ARA. For exam- Additional analyses of the role of the individual empow- ple, research indicates that job control is associated with erment facets revealed that only the competence facet was later retirement intentions and later actual retirement (see significantly related to the DRA and the ARA, resulting in a review Browne et al., 2018). Going beyond the existing full mediation by the DRA. This finding is interesting and 1 3 I. Drazic et al. might be in line with research on employability, which can activities did not compete for the retirees’ resources in our be defined as “the continuous fulfilling, acquiring or creating study. Vangen et al. (2021) did find a negative relationship of work through the optimal use of competences” (Heijde between paid work and volunteering in a sample of Norwe- & Van Der Heijden, 2006, p. 453). Researchers have argued gian seniors aged 62–75 over a 10-year period. However, that employability is an important factor in the motivation the negative relationship was found only in participants to continue working (see, e.g., Pak et al., 2019). with full-time employment (not among those with part-time The results regarding Hypothesis 2a are in line with pre- employment), and the relationship did not apply for partici- vious findings on antecedents of bridge employment at the pants with former experience in voluntary work. The authors person-job level. For example, a study by Fasbender et al. conclude that, in order to increase older people’s engage- (2016) found that the social meaning of work (e.g., having ment in both paid and unpaid work, it might be beneficial to contact with others) and the personal meaning of work (e.g., stimulate combinations of both forms in late careers (Vangen having a meaningful task) were positively related to the et al., 2021). Thus, while psychological empowerment at likelihood of engaging in bridge employment. Our results the workplace could be used to stimulate paid work after differ from those of Fasbender et al. (2016) in that they do retirement (i.e., bridge employment), informing older indi- not focus on employees’ beliefs about work in general, but viduals early about possibilities for volunteering could help on employees’ cognitions that are shaped by their immedi- to engage them in volunteer work before they retire (e.g., ate work environment. From a theoretical perspective, our Hansen & Slagsvold, 2020). results support continuity theory (Atchley, 1989), as indi- Additional analyses regarding Hypotheses 2a and 2b viduals who were psychologically empowered before retire- revealed that none of the facets of the empowerment con- ment showed more engagement in paid work after retirement struct individually predicted bridge employment. This might (i.e., engagement in bridge employment). This could imply indicate that the relationship between psychological empow- that psychologically empowered individuals are likely to erment and bridge employment is driven by the “Gestalt” seek out contexts in which they can continue to experience of psychological empowerment (i.e., the joint presence of psychological empowerment. This finding is consistent with all facets; see Spreitzer, 1995). However, it might also have empowerment theory, according to which psychologically resulted from methodological limitations. We will discuss empowered employees are more likely to take an active this in the Limitations section. orientation toward work and go the “extra mile” Spreitzer Finally, our results regarding Hypothesis 2c indicate that (2008). Thus, it seems plausible that they are less likely to the level of physical limitations moderates the relationship completely withdraw from work compared to less empow- between psychological empowerment and bridge employ- ered older individuals. In light of continuity theory, one may ment decision making. This finding is in line with previous conclude that psychological empowerment during working research on constraining variables. For example, Zhan et al. life—or its continued experience after retirement—might (2013) identified economic stress as a boundary condition help older individuals to maintain stability during potentially for the relationship between career commitment and career- disruptive life changes such as retirement. Future research based bridge employment decision making and the rela- could test this assumption by investigating the relationship tionship between affective commitment to organization and between psychological empowerment during working life organization-based bridge employment decision making, and post-retirement outcomes such as retirement satisfac- such that high economic stress weakened both relationships. tion, psychological health, and/or physical health. Both these findings and our finding support the model by In Hypothesis 2b, we compared paid and unpaid forms Beehr and Bennett (2015), which stresses the role of situa- of later life work. Our results provide indications that psy- tional constraints in the bridge employment decision-making chological empowerment is more strongly associated with process. Future research could combine our study with the bridge employment than with volunteering. Results regard- study by Zhan et al. (2013) by investigating whether empow- ing Hypothesis 2b might also support continuity theory ered employees are more likely to take bridge employment (Atchley, 1989), according to which middle-aged and older at the same organization and identify constraining variables individuals aim to preserve structures by “applying familiar that could be at play in this relationship. strategies in familiar arenas of life” (p. 183)—context and familiarity have likely played a role in maintaining continu- Practical Implications ity in the present study. Furthermore, the fact that bridge employment and volunteering were not significantly cor - One practical implication of the strong relationship between related might speak against the so-called trade-off hypoth- DRA and ARA could be for policymakers to assess DRA esis (see Vangen et al., 2021): Engaging in bridge employ- among older employees to predict changes in the public ment did not decrease the likelihood of being engaged in pension system. Likewise, organizations might consider volunteering and vice versa. Thus, it seems that the two assessing their older employees’ DRAs for more accurate 1 3 Empowered to Stay Active: Psychological Empowerment, Retirement Timing, and Later Life Work staff planning. Furthermore, organizations wanting to retain and not (exclusively) vice versa. Another potential limitation older employees—whether through delayed retirement or is that most of the constructs were measured by single items bridge employment—could consider enhancing psycho- or reduced scales due to the telephone interview method. For logical empowerment cognitions in their employees. They example, the empowerment construct was limited to four might apply general approaches that have been identified as items. This factor plus the constructs’ multidimensionality contextual antecedents of psychological empowerment, for led to a low omega coefficient. Future research could repli- example, high-performance managerial practices and soci- cate our findings with the standard instrument by Spreitzer opolitical support (see meta-analyses by Llorente-Alonso (1995) using an online questionnaire. Furthermore, future et al., 2023 and Seibert et al., 2011). More specific human research could also test the measurement invariance of resource management measures for older employees should the empowerment scale related to age, in particular when also be considered. For example, measures against age dis- investigating more age-diverse samples (for a brief tutorial, crimination seem important, as research indicates negative see Guo et al., 2023). Another potential limitation, which associations between age discrimination and psychological could be at least partly due to measurement problems, is empowerment, which in turn leads to a lower DRA (Scher- that the effect sizes seemed relatively small. However, recent muly et al., 2014). Optimally, specific measures for empow - research suggests that effect sizes that seem small when ering older employees should be developed. One practi- judged by conventional standards (e.g., by Cohen’s guide- cal suggestion for policymakers thus might be to consider lines) may not be weak at all when more domain-specific funding research in this interdisciplinary research field, as standards are applied (e.g., by evaluating correlations from knowledge in this area could help motivate older workers to multiple relevant meta-analyses and defining cut-offs based remain active longer and thereby reduce the financial burden on the 25th, 50th, and 75th percentiles of reported effect on public pension and health care systems in aging socie- sizes; see Davenport et al., 2022). Moreover, even minor ties. It would be especially important to encourage research individual differences in the timing of retirement and/or on psychological empowerment that investigates workers the choice of bridge employment are likely to have drastic older than 50 years, as a recent meta-analysis by Llorente- effects on social security systems through cumulative effects, Alonso et al. (2023) indicates that few empirical studies have given how many older workers are likely to retire in the included this population. Finally, organizations that want to coming years. The study sample itself is another methodo- retain older employees after retirement should start promot- logical limitation. Participants were mostly highly educated ing the physical health of their employees early because, white-collar workers. This might have inflated the correla- according to our findings, the relationship between psycho- tion between DRA and ARA, as research indicates that a logical empowerment and bridge employment disappears lower education level increases the probability of retiring when employees report high levels of physical limitations. before one’s DRA (Solem et al., 2016). Furthermore, our Research informs us that one way leaders can promote the participants were employed individuals aged between 54 and subjective health of their subordinates is to practice respect- 65 years at the first measurement. Thus, we investigated a ful leadership, which in turn might even increase older work- highly selective sample of the population, excluding individ- ers’ DRA (see Wöhrmann et al., 2017). uals who were unemployed at that time or might have already retired involuntarily. Future research should therefore also Limitations examine workers aged 50 or younger to explore involuntary retirement and ways to prevent it. Another limitation is that The limitations of the present study result mainly from the the findings can be extrapolated to other countries only to a archival nature of the panel data. First, as psychological limited extent due to differing retirement and work regula- empowerment and DRA were collected at the same meas- tions. Finally, it is important for future research to examine urement point, common method bias may have altered their the incremental validity of psychological empowerment for relationship (Podsakoff et al., 2012). At least the criterion predicting retirement timing and engagement in later life in this mediation model, ARA, was measured with a con- work against alternative, conceptually meaningful factors siderable time lag (i.e., three years) and presented a more such as psychological capital, person–environment fit, or objective measure. However, we still cannot draw causal job involvement. conclusions about the relationship between psychological empowerment and engagement in later life work. Although we had two measurement points, the focal variables were Conclusion not measured at both time points. Future researchers should conduct longitudinal or preferably interventional studies to This study aimed to investigate the relationship between substantiate our assumption that psychological empower- psychological empowerment before retirement and later ment increases the likelihood of engaging in later life work life work. The three most important findings of this study 1 3 I. Drazic et al. BMAS. (2019). Hinzuverdienst bei Renten und Pensionen (Addi- are as follows: (1) Older workers with higher levels of tional earnings to pensions). https:// www. einfa ch- teilh aben. psychological empowerment not only wanted to retire later de/ DE/ AS/ Themen/ Finan ziell eHilf en/ Rente nPens ionen/ Hinzu but indeed retired later, (2) older workers with higher lev- v er di e ns tR ente n P ensi onen/ hinzu v er di ens tr ente n pens i onen_ els of psychological empowerment were more likely to node. html# doc11 17827 4body Text3 BMAS. (2020). Pension projections exercise 2021. Federal Ministry engage in bridge employment, and (3) the positive rela- of Labour and Social Affairs. tionship between psychological empowerment and bridge Browne, P., Carr, E., Fleischmann, M., Xue, B., & Stansfeld, S. employment disappeared when older workers experienced A. (2018). The relationship between workplace psychosocial high levels of physical limitations. These findings suggest environment and retirement intentions and actual retirement: A systematic review. European Journal of Ageing, 16(1), 73–82. that psychological empowerment might help to retain older https:// doi. org/ 10. 1007/ s10433- 018- 0473-4 workers in the workforce. At the same time, our study Conger, J. A., & Kanungo, R. N. (1988). The empowerment pro- points to the limitations of psychological empowerment, cess: Integrating theory and practice. Academy of Management as the latter became insignificant in predicting bridge Review, 13(3), 471–482. Czepek, J., & Weber, E. (2015). Flexibilität beim Übergang in die employment when high levels of physical limitations Rente, Aktuelle Berichte (Flexibility during the retirement tran- were reported. Thus, we hope that our study sensitizes sition) (No. 2). Institute for Employment Research (IAB). both researchers and practitioners for the importance of Davenport, M. K., Ruffin, M. A., Oxendahl, T. A., McSpedon, M. R., a holistic approach when studying or promoting longer & Beier, M. E. (2022). “Small” effects, big problems. Work, Aging and Retirement, 8(4), 368–370. https:// doi. org/ 10. 1093/ workar/ working lives. Lastly, we encourage future researchers to waac0 12 replicate our findings outside the interview context. Dorn, D., & Sousa-Poza, A. (2010). “Voluntary” and “Involuntary” early retirement: An international analysis. Applied Economics, 42(4), 427–438. Funding Open Access funding enabled and organized by Projekt DRV. (2020). Altersrentner: So viel können Sie hinzuverdienen (Retir- DEAL. ees: This is how much you can earn additionally). https:// www. deuts c he- r ente nvers ic her ung. de/ Shar e dDocs/ Downl oads/ DE/ Open Access This article is licensed under a Creative Commons Attri- Brosc hueren/ natio nal/ alter srent ner_ hinzu verdi enst. html bution 4.0 International License, which permits use, sharing, adapta- Fasbender, U., Wang, M., Voltmer, J.-B., & Deller, J. (2016). The tion, distribution and reproduction in any medium or format, as long meaning of work for post-retirement employment decisions. as you give appropriate credit to the original author(s) and the source, Work, Aging and Retirement, 2(1), 12–23. https://d oi.o rg/1 0.1 093/ provide a link to the Creative Commons licence, and indicate if changes workar/ wav015 were made. The images or other third party material in this article are Fasbender, U., Wöhrmann, A. M., Wang, M., & Klehe, U. C. (2019). included in the article's Creative Commons licence, unless indicated Is the future still open? The mediating role of occupational future otherwise in a credit line to the material. If material is not included in time perspective in the effects of career adaptability and aging the article's Creative Commons licence and your intended use is not experience on late career planning. Journal of Vocational Behav- permitted by statutory regulation or exceeds the permitted use, you will ior, 111, 24–38. https:// doi. org/ 10. 1016/j. jvb. 2018. 10. 006 need to obtain permission directly from the copyright holder. To view a Feldman, D. C. (2013). Feeling like it’s time to retire: A fit perspec- copy of this licence, visit http://cr eativ ecommons. or g/licen ses/ b y/4.0/ . tive on early retirement decision. In M. Wang (Ed.), The Oxford handbook of retirement (pp. 280–292). Oxford University Press. https:// doi. org/ 10. 1093/ oxfor dhb/ 97801 99746 521. 013. 0109 Fisher, G. G., Chaffee, D. S., & Sonnega, A. (2016). Retirement timing: A review and recommendations for future research. Work, Aging References and Retirement, 2(2), 230–261. https:// doi. org/ 10. 1093/ workar/ waw001 Ajzen, I. (1991). The theory of planned behavior. Organizational Garcia, P. R. J. M., Amarnani, R. K., Bordia, P., & Restubog, S. L. Behavior and Human Decision Processes, 50, 179–211. https:// D. (2021). When support is unwanted: The role of psychological doi. org/ 10. 15288/ jsad. 2011. 72. 322 contract type and perceived organizational support in predicting Atchley, R. C. (1989). A continuity theory of normal aging. The bridge employment intentions. Journal of Vocational Behavior, Gerontologist, 29(2), 183–190. https:// doi. org/ 10. 1093/ geront/ 125, 103525. https:// doi. org/ 10. 1016/j. jvb. 2020. 103525 29.2. 183 Gayman, M. D., Turner, R. J., & Cui, M. (2008). Physical limitations Becker, T. E., Atinc, G., Breaugh, J. A., Carlson, K. D., Edwards, and depressive symptoms: Exploring the nature of the association. J. R., & Spector, P. E. (2015). Statistical control in correla- The Journals of Gerontology Series b: Psychological Sciences tional studies: 10 essential recommendations for organizational and Social Sciences, 63(4), S219–S228. https:// doi. org/ 10. 1093/ researchers. Journal of Organizational Behavior, 37(2), 157– geronb/ 63.4. S219 167. https:// doi. org/ 10. 1002/ job. 2053l Giandrea, M. D., Cahill, K. E., & Quinn, J. (2010). The role of re-entry Beehr, T. A., & Bennett, M. M. (2015). Working after retirement: in the retirement process (Vol. 439). US Department of Labor, US Features of bridge employment and research directions. Work, Bureau of Labor Statistics, Office of Productivity and Technology. Aging and Retirement, 1(1), 112–128. https:// doi. org/ 10. 1093/ Grigoli, F., Koczan, Z., & Topalova, P. W. (2021). Calling older work- workar/ wau007 ers back to work. Applied Economics Letters, 29(6), 559–566. Berger, U. (2019). Interne Konsistenz von Fragebogen-Skalen: Alpha https:// doi. org/ 10. 1080/ 13504 851. 2021. 18762 05 oder Omega (Internal consistency of questionnaire scales: Guo, F., Min, H., Jex, S., & Choi, Y. (2023). Old enough to perceive Alpha or omega). Psychotherapie Psychosomatik Medizinische things differently? Detecting measurement invariance across age Psychologie, 69(08), 348–349. groups using item-focused tree. Work, Aging and Retirement, 9(1), 59–70. https:// doi. org/ 10. 1093/ workar/ waac0 04 1 3 Empowered to Stay Active: Psychological Empowerment, Retirement Timing, and Later Life Work Hansen, T., & Slagsvold, B. (2020). An “army of volunteers”? Engage- Co-operation and Development. https://www .oecd. or g/els/ em p/ ment, motivation, and barriers to volunteering among the baby avera ge- effec tive- age- of- retir ement. htm boomers. Journal of Gerontological Social Work, 63(4), 335–353. OECD. (2019b). Pensions at a glance 2019: OECD and G20 indica- https:// doi. org/ 10. 1080/ 01634 372. 2020. 17582 69 tors. Organisation for Economic Co-Operation and Develop- Heijde, C. M. V. D., & Van Der Heijden, B. I. J. M. (2006). A compe- ment. https:// doi. org/ 10. 1787/ b6d3d cfc- en tence-based and multidimensional operationalization and meas- Örestig, J., Strandh, M., & Stattin, M. (2013). A wish come true? urement of employability. Human Resource Management, 45(3), A longitudinal analysis of the relationship between retire- 449–476. https:// doi. org/ 10. 1002/ hrm. 20119 ment preferences and the timing of retirement. Journal of Hofäcker, D., & Naumann, E. (2015). The emerging trend of work Population Ageing, 6(1–2), 99–118. h ttp s: // do i. or g/ 10 . 1 007/ beyond retirement age in Germany. Zeitschrift Für Gerontolo-s12062- 012- 9075-7 gie Und Geriatrie, 48(5), 473–479. https:// doi. or g/ 10. 1007/ Pak, K., Kooij, D. T. A. M., De Lange, A. H., & Van Veldhoven, s00391- 014- 0669-y M. J. P. M. (2019). Human resource management and the abil- Holtemöller, O., Kooths, S., Michelsen, C., Schmidt, T., & Wollmer- ity, motivation and opportunity to continue working: A review shäuser, T. (2021). Gemeinschaftsdiagnose: Pandemie verzögert of quantitative studies. Human Resource Management Review, Aufschwung – Demografie bremst Wachstum (Pandemic delays 29(3), 336–352. https:// doi. org/ 10. 1016/j. hrmr. 2018. 07. 002 upswing—Demography slows growth). Wirtschaftsdienst, 101(5), Peng, Y. (2022). A two-study examination of age discrimination, 353–357. work meaningfulness, and bridge employment intentions. Hooper, D., Coughlan, J., & Mullen, M. R. (2008). Structural equa- Career Development International, 27(3), 391–407. https:// tion modelling: Guidelines for determining model fit. Electronic doi. org/ 10. 1108/ CDI- 11- 2021- 0295 Journal on Business Research Methods, 6(1), 53–60. Peng, Y., Xu, X., Jex, S. M., & Chen, Y. (2020). The roles of job- Hughes, M. E., Waite, L. J., Hawkley, L. C., & Cacioppo, J. T. (2004). related psychosocial factors and work meaningfulness in A short scale for measuring loneliness in large surveys: Results promoting nurses’ bridge employment intentions. Journal of from two population-based studies. Research on Aging, 26(6), Career Development, 47(6), 701–716. https:// doi. org/ 10. 1177/ 655–672. https:// doi. org/ 10. 1177/ 01640 27504 26857408948 45318 824666 Kanter, R. M. (1977). Men and women of the corporation. Basic Perugini, M., & Bagozzi, R. P. (2001). The role of desires and antici- Books. pated emotions in goal-directed behaviours: Broadening and Kraimer, M. L., Seibert, S. E., & Liden, R. C. (1999). Psychological deepening the theory of planned behaviour. British Journal of empowerment as a multidimensional construct: A test of con- Social Psychology, 40(1), 79–98. https://doi. or g/10. 1348/ 01446 struct validity. Educational and Psychological Measurement, 66011 64704 59(1), 127–142. https:// doi. org/ 10. 1177/ 00131 64499 591009 Peters, G.-J.Y. (2014). The alpha and the omega of scale reliability Lee, M. Y., & Edmondson, A. C. (2017). Self-managing organi- and validity: Why and how to abandon Cronbach’s alpha and the zations: Exploring the limits of less-hierarchical organizing. route towards more comprehensive assessment of scale quality. Research in Organizational Behavior, 37, 35–58. https:// doi. The European Health Psychologist, 16(2), 56–69. https://doi. or g/ org/ 10. 1016/j. riob. 2017. 10. 00210. 31234/ osf. io/ h47fv Li, Y., Turek, K., Henkens, K., & Wang, M. (2022). Retaining retire- Podsakoff, P. M., MacKenzie, S. B., & Podsakoff, N. P. (2012). Sources ment-eligible older workers through training participation: The of method bias in social science research and recommendations on joint implications of individual growth need and organizational how to control it. Annual Review of Psychology, 63(1), 539–569. climates. Journal of Applied Psychology. https:// doi. or g/ 10. https:// doi. org/ 10. 1146/ annur ev- psych- 120710- 100452 1037/ apl00 01065 Revelle, W., & Zinbarg, R. (2009). Coefficientes alpha, beta, omega Liden, R. C., & Arad, S. (1996). A power perspective of empower- and the GLB. Psychometrika, 74(1), 145–154. ment and work groups: Implications for human resources man- Rosseel, Y. (2012). Lavaan: An R package for structural equation mod- agement research. Research in Personnel and Human Manage- eling. Journal of Statistical Software, 48(2), 1–36. https://doi. or g/ ment, 14, 205–251.10. 18637/ jss. v048. i02 Llorente-Alonso, M., García-Ael, C., & Topa, G. (2023). A meta- Rosseel, Y. (2014). Structural equation modeling with categorical vari- analysis of psychological empowerment: Antecedents, organiza- ables (summer school – using R for personality research, August tional outcomes, and moderating variables. Current Psychology. 23–28, Bertinoro, Italy). https://perso nality -pr oject. or g/r/tut orials/ https:// doi. org/ 10. 1007/ s12144- 023- 04369-8summe rscho ol. 14/ rosse el_ sem_ cat. pdf Malik, M., Sarwar, S., & Orr, S. (2021). Agile practices and per- Rouzet, D., Caldera Sánchez, A., Renault, T., & Roehn, O. (2019). formance: Examining the role of psychological empowerment. Fiscal challenges and inclusive growth in ageing societies (OECD International Journal of Project Management, 39(1), 10–20. Economic Policy Papers, No. 27). OECD Publishing. https:// doi. org/ 10. 1016/j. ijpro man. 2020. 09. 002 Sackreuther, I., Schröber, J., Cihlar, V., Mergenthaler, A., Micheel, F., McNeish, D. (2018). Thanks coefficient alpha, we’ll take it from & Schill, G. (2016). TOP - Transitions and Old Age Potential: here. Psychological Methods, 23(3), 412–433. https:// doi. org/ Methodology report on the study. Federal Institute for Population 10. 1037/ met00 00144 Research. Mergenthaler, A., Cihlar, V., Micheel, F., Sackreuther, I., Riedl, S., Schermuly, C. C. (2016). New Work—Gute Arbeit gestalten (New & Maruhn, S. (2017). TOP - Transitions and Old Age Potential: work—Designing good work). Haufe-Lexware. Data and methodology report for the second wave of the study. Schermuly, C. C., Büsch, V., & Graßmann, C. (2017). Psychological Federal Institute for Population Research. empowerment, psychological and physical strain and the desired Naghavi, M. S., Jofreh, M. G., Vaezi, R., & Ghorbanizadeh, V. retirement age. Personnel Review, 46(5), 950–969. https://doi. or g/ (2019). Empowerment of the professional ageing workforce: 10. 1108/ PR- 06- 2015- 0159 A review and development of a model. European Journal of Schermuly, C. C., Deller, J., & Büsch, V. (2014). A research note on International Management, 13(2), 269–286. https://doi. or g/10. age discrimination and the desire to retire. Research on Aging, 1504/ EJIM. 2019. 098185 36(3), 382–393. https:// doi. org/ 10. 1177/ 01640 27513 508288 OECD. (2019a). Ageing and employment policies—Statistics on Schermuly, C. C., Meyer, B., & Dämmer, L. (2013). Leader-mem- average effective age of retirement. Organisation for Economic ber exchange and innovative behavior: The mediating role of 1 3 I. Drazic et al. psychological empowerment. Journal of Personnel Psychology, Vanajan, A., Bültmann, U., & Henkens, K. (2020). Why do older work- 12(3), 132–142. https:// doi. org/ 10. 1027/ 1866- 5888/ a0000 93 ers with chronic health conditions prefer to retire early? Age and Seibert, S. E., Wang, G., & Courtright, S. H. (2011). Antecedents and Ageing, 49(3), 403–410. https:// doi. org/ 10. 1093/ ageing/ afz180 consequences of psychological and team empowerment in organi- Vangen, H., Hellevik, T., & Herlofson, K. (2021). Associations zations: A meta-analytic review. Journal of Applied Psychology, between paid and unpaid work among Norwegian seniors: Com- 96(5), 981–1003. https:// doi. org/ 10. 1037/ a0022 676 petition, complementarity or continuity? European Journal of Shultz, K. S. (2003). Bridge employment: Work after retirement. In G. Ageing. https:// doi. org/ 10. 1007/ s10433- 021- 00615-9 A. Adams & T. A. Beehr (Eds.), Retirement: Reasons, processes, Vignoli, M., Zaniboni, S., Chiesa, R., Alcover, C.-M., Guglielmi, D., and results (pp. 214–241). Springer. & Topa, G. (2021). Maintaining and engaging older workers at Shultz, K. S., & Wang, M. (2011). Psychological perspectives on the work: The trigger role of personal and psychosocial resources. The changing nature of retirement. American Psychologist, 66(3), International Journal of Human Resource Management, 32(8), 170–179. https:// doi. org/ 10. 1037/ a0022 411 1731–1753. https:// doi. org/ 10. 1080/ 09585 192. 2019. 15792 52 Solem, P. E., Syse, A., Furunes, T., Mykletun, R. J., De Lange, A., Wang, M., Adams, G. A., Beehr, T. A., & Shultz, K. S. (2009). Bridge Schaufeli, W., & Ilmarinen, J. (2016). To leave or not to leave: employment and retirement: Issues and opportunities during the Retirement intentions and retirement behaviour. Ageing and Soci- latter part of one’s career. In S. G. Baugh & S. E. Sullivan (Eds.), ety, 36(2), 259–281. https://doi. or g/10. 1017/ S0144 686X1 40011 35 Maintaining focus, energy, and options over the career. Informa- Sousa-Ribeiro, M., Bernhard-Oettel, C., Sverke, M., & Westerlund, H. tion Age Publishing. (2021). Health- and age-related workplace factors as predictors of Wang, M., & Shi, J. (2014). Psychological research on retirement. preferred, expected, and actual retirement timing: Findings from Annual Review of Psychology, 65, 209–233. https:// doi. org/ 10. a Swedish cohort study. International Journal of Environmental 1146/ annur ev- psych- 010213- 115131 Research and Public Health, 18(5), 2746. https://doi. or g/10. 3390/ Wang, M., & Shultz, K. S. (2010). Employee retirement: A review and ijerp h1805 2746 recommendations for future investigation. Journal of Manage- Spreitzer, G. (1995). Psychological empowerment in the workplace: ment, 36(1), 172–206. https://doi. or g/10. 1177/ 01492 06309 347957 Dimensions, measurement, and validation. Academy of Manage- Wang, M., Zhan, Y., Liu, S., & Shultz, K. S. (2008). Antecedents of ment Journal, 38(5), 1442–1465. bridge employment: A longitudinal investigation. Journal of Spreitzer, G. (2008). Taking stock: A review of more than twenty years Applied Psychology, 93(4), 818–830. https:// doi. or g/ 10. 1037/ of research on empowerment at work. In J. Barling & C. L. Cooper 0021- 9010. 93.4. 818 (Eds.), Handbook of organizational behavior (pp. 54–72). Sage. Watermann, H., Fasbender, U., & Klehe, U.-C. (2023). Withdrawing Thomas, K. W., & Velthouse, B. A. (1990). Cognitive elements of from job search: The effect of age discrimination on occupational empowerment: An “interpretive” model of intrinsic task motiva- future time perspective, career exploration, and retirement inten- tion. Academy of Management Review, 15(4), 666–681. tions. Acta Psychologica, 234, 103875. https:// doi. org/ 10. 1016/j. Topa, G., Depolo, M., & Alcover, C.-M. (2018). Early retirement: A actpsy. 2023. 103875 meta-analysis of its antecedent and subsequent correlates. Fron- Wöhrmann, A. M. (2021). Later life work: Individual, job, and organi- tiers in Psychology, 8, 2157. https:// doi. org/ 10. 3389/ fpsyg. 2017. zational factors for retirement timing and work beyond normal 02157 retirement age. Leuphana University of Lüneburg. Topa, G., Moriano, J. A., Depolo, M., Alcover, C.-M., & Morales, Wöhrmann, A. M., Deller, J., & Wang, M. (2013). Outcome expec- J. F. (2009). Antecedents and consequences of retirement plan- tations and work design characteristics in post-retirement work ning and decision-making: A meta-analysis and model. Journal planning. Journal of Vocational Behavior, 83(3), 219–228. https:// of Vocational Behavior, 75(1), 38–55. https:// doi. org/ 10. 1016/j. doi. org/ 10. 1016/j. jvb. 2013. 05. 003 jvb. 2009. 03. 002 Wöhrmann, A. M., Fasbender, U., & Deller, J. (2017). Does more Traeger, C., & Alfes, K. (2019). High-performance human resource respect from leaders postpone the desire to retire? Understand- practices and volunteer engagement: The role of empowerment ing the mechanisms of retirement decision-making. Frontiers in and organizational identification. VOLUNTAS: International Jour - Psychology, 8, 1400. https:// doi. org/ 10. 3389/ fpsyg. 2017. 01400 nal of Voluntary and Nonprofit Organizations, 30(5), 1022–1035. Zaniboni, S., Topa, G., & Balducci, C. (2021). Core self-evaluations https:// doi. org/ 10. 1007/ s11266- 019- 00135-2 affecting retirement-related outcomes. International Journal of Trampusch, C. (2005). Institutional resettlement: The case of early Environmental Research and Public Health, 18(1), 174. retirement in Germany. In W. Streeck & K. Thelen (Eds.), Beyond Zhan, Y., Wang, M., & Yao, X. (2013). Domain specific ee ff cts of com - continuity: Institutional change in advanced political economies mitment on bridge employment decisions: The moderating role of (pp. 203–228). Oxford University Press. economic stress. European Journal of Work and Organizational Truesdale, B. C. (2020). Better jobs, longer working lives: Proposals to Psychology, 22(3), 362–375. https:// doi. org/ 10. 1080/ 13594 32X. improve the low-wage labor market for older workers (Economic 2012. 762763 Studies at Brookings). The Brookings Institution. https:// www. brook ings. edu/ wp- conte nt/ uploa ds/ 2020/ 11/ ES- 11. 16. 20- Trues Publisher's Note Springer Nature remains neutral with regard to dale. pdf jurisdictional claims in published maps and institutional affiliations. van Rijn, R. M., Robroek, S. J. W., Brouwer, S., & Burdorf, A. (2014). Influence of poor health on exit from paid employment: A sys- tematic review. Occupational and Environmental Medicine, 71(4), 295–301. https:// doi. org/ 10. 1136/ oemed- 2013- 101591 1 3 http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Journal of Adult Development Springer Journals

Empowered to Stay Active: Psychological Empowerment, Retirement Timing, and Later Life Work

Loading next page...
 
/lp/springer-journals/empowered-to-stay-active-psychological-empowerment-retirement-timing-PttF3PKrlw

References (96)

Publisher
Springer Journals
Copyright
Copyright © The Author(s) 2023
ISSN
1068-0667
eISSN
1573-3440
DOI
10.1007/s10804-023-09453-8
Publisher site
See Article on Publisher Site

Abstract

Motivating older employees both to prolong their working lives and to stay active even after retirement has become increas- ingly important due to rising old-age dependency ratios. Later life work—including both paid work and volunteering—has thus become an important topic for scholars and practitioners. We aim to extend research on later life work by hypothesizing that psychological empowerment at work increases not only desired and actual retirement ages but also levels of later life work. Second, we test differential effects of psychological empowerment on later life work, expecting it to be more strongly related to paid work after retirement (i.e., bridge employment) than to volunteering. Third, we suggest that the relationship between psychological empowerment and bridge employment depends on the employees’ level of physical limitations. We used data from a longitudinal panel study in Germany in which structured telephone interviews were conducted. A sample of older individuals who had retired between two waves of measurement was drawn (time lag: three years; n = 210). The results of a path analysis support the postulated mediation. Furthermore, as expected, psychological empowerment more accurately predicted bridge employment than volunteering, and physical limitations moderated the relationship between psychological empowerment and bridge employment. Lastly, additional analyses on the individual empowerment facets revealed that only the competence facet played a significant role in the proposed hypotheses. Overall, our findings suggest that psychological empowerment may help to increase older employees’ motivation to delay retirement and to stay active even after retirement. Keywords Aging workforce · Psychological empowerment · Retirement age · Activeness after retirement · Bridge employment The working-age population in many developed countries 2010). It depends not only on official retirement ages but is expected to decrease by more than one-third by 2060 also on individual contexts, needs, and preferences as well (OECD, 2019b). Many aging societies will thus have to as socio-economic factors (Wang & Shultz, 2010). Invol- face an immense financial burden to their public pension untary early retirement—that is, “a retirement that results and healthcare systems (Rouzet et al., 2019). Therefore, in from a situation with (often unexpected) employment con- countries such as Germany, for example, several economic straints” (Dorn & Sousa-Poza, 2010, p. 427)—has increased institutes have recently called for the statutory retirement during the COVID-19 pandemic and is mostly experienced age to be raised to 69 years (Holtemöller et al., 2021). How- by individuals working in low-wage sectors (Hofäcker & ever, even if such increases would provide a legal basis for Naumann, 2015; Truesdale, 2020). The main reasons people longer working lives, the question remains as to whether retire involuntarily are that they work in poor-quality jobs older employees would actually be able and motivated to and experience low employment stability during their 50 s delay retirement. Over the years, individual retirement has (i.e., they are not continuously employed). Thus, it is impor- become a complex and dynamic process (Giandrea et al., tant to note that, although statutory retirement ages are being raised, many older workers are not able to continue working, at least, because delaying retirement requires having a job * Ivana Drazic to retire from (Truesdale, 2020). At the same time, although ivana.drazic@srh.de early retirement (voluntary or involuntary) is widespread Department of Business Psychology, SRH Berlin University in many developed countries (OECD, 2019a), increasing of Applied Sciences, Ernst-Reuter-Platz 10, 10587 Berlin, numbers of older individuals are deciding to prolong their Germany Vol.:(0123456789) 1 3 I. Drazic et al. working lives (Grigoli et al., 2021). This heterogeneity raises that the effect of job satisfaction on early retirement deci- the following question: Which factors influence the retire- sions is negligible. ment timing decision of older employees who could theoreti- A construct at the person-job level that might more con- cally continue working? sistently predict retirement timing by reaching the root of Common variables that have been linked to retirement work-related experiences is psychological empowerment. intentions and, in consequence, retirement decisions are Initially, research on employee empowerment concentrated individual attributes (e.g., health), family factors (e.g., on socio-structural empowerment (see, e.g., Kanter, 1977), marital status), and socio-economic factors (e.g., current which can be defined as the distribution of power across economic conditions; for a review, see Wang & Shultz, organizational participants through empowering policies and 2010). In addition to these factors, job and organizational practices (Liden & Arad, 1996). Although this democratic factors—that is, characteristics of jobs and work organi- approach received great attention, it was found to be lim- zations and interactions between the person and job vari- ited. For example, it could not explain why some employ- ables—also play an important role (see Wang & Shultz, ees did not feel empowered even though they were working 2010). For example, a study by Vignoli et al. (2021) indi- in empowering structures or vice versa (Spreitzer, 2008). cates that higher levels of work ability and lower percep- Because of these limitations, researchers started studying tions of age stereotypes at the workplace are related to the employees’ subjective experiences of empowerment. Conger desire to work longer. Another recent study that inves- and Kanungo (1988) substantially influenced this change in tigated actual retirement decisions found that individual the view of empowerment (Spreitzer, 2008). Thomas and growth need and organizational climates that encourage Velthouse (1990) built on their work by defining empower - older workers’ learning and development were positively ment as an intrinsic motivation, further specifying its cog- related to older workers’ decisions to stay (vs. retire) nitive preconditions. Finally, based on Thomas’ and Velt- regardless of their retirement eligibility (Li et al., 2022). house (1990) work, Spreitzer (1995) proposed a definition Findings from a study by Sousa-Ribeiro et al. (2021) cor- of psychological empowerment that is now used widely in roborate the importance of an age-friendly work environ- empirical research. Spreitzer (1995) defined psychological ment and further showed that feeling positive regarding the empowerment as intrinsic task motivation manifested in future at work was positively related to the participants’ four work-related cognitions: meaning, self-determination, preferred, expected and actually retirement ages. In line competence, and impact. Meaning reflects the degree of with this finding, Fasbender et al. (2019) found that career alignment between employees’ values, beliefs, and behav- adaptability (i.e., the ability and willingness to manage iors and the corresponding requirements of their work roles one’s own career) and levels of personal growth were posi- (Spreitzer, 1995). Competence represents the extent to which tively related to late career planning among a sample of employees believe that they have the skills to perform their older workers, and that both relationships were mediated work well (Spreitzer, 2008). Self-determination comprises by occupational future time perspective. Findings from a “autonomy in the initiation and continuation of work behav- study by Watermann et al. (2023) underscore the impor- iors and processes” experienced by employees (Spreitzer, tance of both older individuals’ future time perspectives 1995, p. 1443). Impact reflects how strongly employees can and perceptions of age discrimination. The authors found influence important outcomes at work (Spreitzer, 1995). that age discrimination was negatively related to retire- Taken together, these four work-related cognitions constitute ment intentions via remaining time (a facet of occupational the “Gestalt” of psychological empowerment, which implies future time perspective). In general, job and organizational that all four cognitions are essential for experiencing psycho- factors seem to be particularly relevant because the aver- logical empowerment (Spreitzer, 1995). Thus, for instance, age employee spends many hours at work before retiring; individuals who perceive a great deal of autonomy but no in Germany, the time spent is approximately 76,500  h meaning at work will not feel psychologically empowered (Schermuly, 2016). Thus, organizations may influence (Spreitzer, 2008). In contrast, individuals who do feel psy- older workers’ retirement decisions by changing certain chologically empowered are more likely to take an active job and organizational factors. Meta-analytical findings orientation toward work (Spreitzer, 1995). suggest that job satisfaction and work involvement are the Meta-analytical findings indicate positive consequences strongest predictors of retirement planning at the person- of psychological empowerment for important work out- job level (see Topa et al., 2009). A more recent system- comes such as job satisfaction and performance (Seibert atic review indicates that high job satisfaction and high et al., 2011). With increasingly complex and dynamic busi- job control are associated with not only later retirement ness environments, research on psychological empower- intentions, but also actual retirement (Browne et al., 2018). ment—as one answer to this challenge—has blossomed over However, a meta-analysis by Topa et al. (2018) suggests the past decades (see Seibert et al., 2011), and the topic still generates considerable research interest (e.g., Malik et al., 1 3 Empowered to Stay Active: Psychological Empowerment, Retirement Timing, and Later Life Work 2021). By enabling employees to use their initiative in their experiences by reflecting how people interpret their job situ- daily work, psychological empowerment can be seen as an ation. It precedes common person-job factors such as job sat- important measure to counteract the challenges of increas- isfaction and represents a more complex antecedent of later ingly complex business environments (Lee & Edmondson, life work, comprising four distinct facets. Finally, from a 2017). However, another major challenge for many com- practical point of view, it is worthwhile to study psychologi- panies is the aging of the workforce, and little research on cal empowerment, as research suggests that it is malleable. empowerment has focused on the specific empowerment Meta-analytical findings indicate that it can be influenced of older employees and its consequences (Naghavi et al., by psychosocial organizational factors such as structural 2019). As one exception, Schermuly et al. (2017) found a empowerment, leadership, and trust in the organization positive relationship between psychological empowerment (Llorente-Alonso et al., 2023; Seibert et al., 2011). To test and desired retirement age (DRA). However, their study our hypotheses, we extended previous research on psycho- was based on cross-sectional data, and it did not investigate logical empowerment and DRA using panel data from the actual retirement age (ARA). Consistent with this limita- project “Transition and Old Age Potential” (TOP). Individu- tion, Browne et al.’s (2018) systematic review calls for more als aged 55 years or over were interviewed via telephone. We studies to examine ARA when investigating the influence of applied a two-wave design with a time lag of three years, psychosocial work conditions on retirement decisions. The investigating individuals who had retired between two waves first part of the present study addresses this research gap by of measurement (n = 210). In the next paragraph, we first examining whether empowered employees not only desire provide information on the German retirement system to to retire later but indeed retire later. contextualize our study. Second, we conceptualize retire- The second part of the study investigates the conse- ment from a psychological perspective and introduce the quences of psychological empowerment on activeness after temporal model of retirement as an overarching framework. retirement. We hypothesize that psychological empower- Lastly, we derive our hypotheses. The conceptual research ment during working life is positively related to levels of model for all hypotheses is depicted in Fig. 1. later life work. Later life work can be defined as “work activities of older adults just before and beyond normal retirement age” (Wöhrmann, 2021, p. 12), including both Background paid work and volunteering. Based on continuity theory (Atchley, 1989), we expect psychological empowerment to The German Retirement System be related to later life work and to better predict engage- ment in paid work after retirement than engagement in vol- Germany has a long history of institutionalized early retire- unteering. Furthermore, as suggested by Beehr and Bennett ment (i.e., the act of retiring before the statutory retirement (2015), we investigate boundary conditions of the relation- age), which was intended, inter-alia, to promote organiza- ship between psychological empowerment and the decision tional restructuring (Trampusch, 2005). Not until the 1990s to engage in bridge employment. We suggest that the posi- did German policymakers gradually begin to close pathways tive relationship between psychological empowerment and to early retirement (Hofäcker & Naumann, 2015). In the bridge employment is moderated by employees’ physical 2000s, Germany further promoted a paradigm shift from limitations. Finally, to advance research on psychological early exit to working longer by introducing labor market empowerment, we respond to calls by Spreitzer (2008) and programs for older employees and a stepwise increase of investigate the roles of the individual facets of psychological the statutory retirement age from 65 to 67 years between empowerment within exploratory analyses. 2011 and 2029 (BMAS, 2020; Hofäcker & Naumann, 2015). In summary, we aim to contribute to the literature on later In general, the current German pension program allows life work by testing whether employees who feel more psy- for flexible retirement timing: Individuals can retire before chologically empowered (a) not only desire to retire later but reaching statutory retirement age by accepting monthly pen- indeed retire later, (b) are more likely to engage in paid work sion claims, which are to be reduced by 0.3% until statu- after retirement than employees who feel less psychologi- tory retirement age is reached. On the other hand, surpass- cally empowered, and (c) are more likely to engage in paid ing statutory retirement age is also possible and even gets work after retirement than in volunteering. Furthermore, we rewarded with a surcharge of 0.5% per month (Czepek & want to shed light on boundary conditions of the proposed Weber, 2015). Furthermore, retirees who do not retire early relationships by proposing physical limitations as a moder- can continue working without income restrictions (DRV, ating variable. We also examine the role of the individual 2020), which further establishes financial incentives to facets of psychological empowerment. From a conceptual prolong working life. Concerning the sample of the present point of view, these investigations are important, because study, the statutory retirement ages for claiming full state psychological empowerment reaches the root of work-related pension ranged from 65 to 66 years. 1 3 I. Drazic et al. Fig. 1 Conceptual research model 2010). This model describes the retirement process heuris- The Temporal Process Model of Retirement tically, capturing its dynamic and complex nature. It outlines three sequential and gradually unfolding retirement stages: Retirement has been conceptualized in many different ways (1) retirement planning, (2) retirement decision making, and in the literature, with the lowest common denominator being (3) retirement transition and adjustment. In the retirement that it is not a single event, but rather a process that takes planning phase, older people begin to envision their still place over a period of time (Shultz & Wang, 2011). For distant retirement lives and make initial plans and prepara- a long time, the focus in retirement research has been on tions for retirement. As retirement nears, people enter the physical and financial factors of retirement (Shultz & Wang, decision-making phase. In analogy to the conceptualization 2011). However, over the last decades, an increasing amount of retirement as decision making, in this phase, individuals of research has applied psychological perspectives on the consider and weigh concrete alternatives to reach conclu- retirement process. Wang and Shi (2014) identified three sions about their retirement. The final phase, retirement tran- psychological conceptualizations of retirement based on sition and adjustment, deals with daily changes in retirement their literature review: (1) retirement as decision making, life after the transition from full-time worker to retiree. Daily (2) retirement as an adjustment process, and (3) retirement activities in retirement include leisure activities, caregiving, as a career-development stage. In the conceptualization of and later life work (i.e., paid and/or unpaid forms of work). retirement as decision making, older workers consider and As well as describing the sequential process of retirement, weigh information that they have about themselves, their the temporal process model identifies four factors influenc- work, and their private life to make decisions about their ing the retirement process that we have already mentioned in retirement. Retirement as an adjustment process focuses on the Introduction: individual attributes, family factors, socio- the transition from employment to retirement and the indi- economic factors, and job and organizational factors. vidual development of older peoples’ lives after retirement. The first part of the present study examines the relation- Finally, retirement as a career-development stage stresses ship between psychological empowerment and retirement older people’s ongoing potential to change their own career timing (i.e., desired and actual retirement ages) and thus paths after retirement (Wang & Shi, 2014). These concep- focuses on the first two phases of retirement: retirement tualizations are not completely separate from each other but planning and retirement decision making. The second part of can be viewed as parts of an overarching process (see, e.g., the study examines the relationship between psychological Wöhrmann, 2021). empowerment during working life and later life work (i.e., One framework that helps to integrate the different con- bridge employment and volunteering). Thus, this part of the ceptualizations of retirement is the temporal process model study focuses on the third phase of the retirement process: of retirement (e.g., Wang & Shi, 2014; Wang & Shultz, 1 3 Empowered to Stay Active: Psychological Empowerment, Retirement Timing, and Later Life Work retirement transition and adjustment. Following suggestions the behavior, subjective norms, and perceived behavioral about the correspondence between certain retirement con- control. First, we believe desires constitute a valid proxy ceptualizations and certain theories (see Wang & Shultz, for intentions, as desires have been shown to mediate the 2010), we conceptualize retirement as an adjustment and relationship between attitudes, norms and perceived behav- a decision-making process and focus on continuity theory ioral control, and intentions (see Perugini & Bagozzi, 2001). (Atchley, 1989) and the theory of planned behavior (Ajzen, Second, we argue that in the case of retirement, perceived 1991), respectively. behavioral control plays a particularly important role: Retire- ment timing is, to a great degree, dependent on legal regula- Retirement Planning and Decision Making: The tions. Thus, in countries where pension programs allow for Relationship between Psychological Empowerment, flexible retirement timing—such as Germany—the desire to Desired Retirement Age, and Actual Retirement Age retire should have a major influence on ARA. Based on these arguments, we hypothesize: As suggested by the temporal process model of retirement (e.g., Wang & Shultz, 2010), retirement typically starts with Hypothesis 1 Higher levels of psychological empowerment a phase in which older people gather and weigh information lead to higher levels of DRA, which in turn result in higher about themselves, their work, and their private life to define levels of ARA. when and how to retire. We argue that older individuals use their psychological empowerment experiences as a source of Retirement Transition and Adjustment: The Relationship information from the job and work field and that empowered between Psychological Empowerment and Work older individuals want to retire later than less empowered after Retirement individuals. We base this hypothesis on continuity theory (Atchley, 1989), which is in line with the conceptualization Even for individuals with high levels of psychological of retirement as an adjustment process. Continuity theory empowerment, there comes a time when it is reasonable to focuses on adaptation to change in later life stages, and retire—for example, for financial reasons. However, even it states that people strive to maintain internal continuity after retirement, older people can continue to experience (i.e., their self-concept) and external continuity (i.e., social psychological empowerment by engaging in work after structures) across their lifespan to prevent experiences of retirement. Work after retirement can be paid or nonpaid harmful disruption. A central premise of continuity theory (see definition of later life work; Wöhrmann, 2021). We is that individuals maintain continuity by engaging in types believe that psychological empowerment during working life of activities, behaviors, and relationships similar to those predicts levels of later life work, especially paid work after they have previously experienced. We argue that the more retirement, which is often discussed as bridge employment psychological empowerment older individuals experience in in the literature (Shultz, 2003). their preretirement lives, the more they will wish to continue Bridge employment refers to the pattern of labor force this experience in order to maintain continuity. We hypoth- participation of older employees that spans the period esize that this desire for continuity will be reflected in the between their career jobs and complete withdrawal from desire to delay retirement, as remaining in the workforce the labor force (Shultz, 2003; Wang et al., 2008). A simpler allows them to continue to experience psychological empow- definition of bridge employment is any kind of paid work erment. Previous research has corroborated this assumption. after retirement (e.g., Shultz, 2003). It can be part-time, Schermuly et al. (2017), for example, found that the more full-time, or self-employed, career- or noncareer-related, psychologically empowered employees felt, the later they with the same or a different employer (for a taxonomy, see wanted to retire. Beehr & Bennett, 2015), and it can be dynamic (individuals With the present study, we aim to extend previous findings can enter and leave the workforce several times after retire- on the relationship between psychological empowerment ment; Wang et al., 2009). The most typical categorization of and the delay of retirement by investigating whether psy- bridge employment is that of career bridge employment (i.e., chologically empowered employees not only desire to retire employment in the same industry or field as the career job) later but actually retire later. As stated above, retirement in contrast to bridge employment in a different field (Shultz, can be conceptualized as a motivational decision-making 2003; Wang et al., 2008). process (Shultz & Wang, 2011). In motivational processes, Antecedents of bridge employment can be categorized intentions are the most proximal determinants of actual into personal characteristics (e.g., health), contextual fac- behavior (see theory of planned behavior; Ajzen, 1991). tors (e.g., economy), and organizational factors (e.g., pol- According to the theory of planned behavior, an intention icy incentives; see Beehr & Benett, 2015). It is assumed reflects the extent to which an individual is motivated to per - that organizational factors are indirectly related to bridge form a specific behavior, and it is driven by attitudes toward employment, mediated by subjective attitudes and feelings 1 3 I. Drazic et al. toward the organization. Indeed, cognitions relating to work experiences. We argue that bridge employment constitutes and the organization have been linked to bridge employ- the most proximal setting in which retired individuals can ment. For example, a study by Garcia et al. (2021) found continue to experience psychological empowerment. that organizational support increased bridge employment intentions only among older workers who endorsed a rela- Hypothesis 2a Psychological empowerment before tional psychological contract (vs. a transactional psychologi- retirement is positively related to engagement in bridge cal contract). Other recent studies suggest that experiences employment. of work meaningfulness mediate the relationships between workplace age discrimination, workplace incivility, rela- To further corroborate continuity theory (Atchley, 1989), tional job crafting (i.e., interpersonal crafting at work), we hypothesize that psychological empowerment before and generativity opportunities and bridge employment retirement is more strongly related to bridge employment intentions, respectively (Peng, 2022; Peng et  al., 2020). than to volunteering after retirement. Although a recent Another study focusing on employee’ cognitions found that study has shown that psychological empowerment can be employees with higher outcome expectations regarding stimulated in the volunteering context (Traeger & Alfes, post-retirement work (e.g., they had greater expectations 2019), we believe the context of the original empowerment that post-retirement work would enable them to stay physi- experience to be crucial when it comes to continuity. In line cally and mentally healthy) were more likely to engage in with this argument, a recent longitudinal study by Vangen post-retirement work planning (Wöhrmann et al., 2013). et al. (2021) found that previous engagement in voluntary These findings substantiate the assumption that it is not the work among 62- to 75-year-olds in Norway was strongly organizational factors per se but the subjective interpreta- associated with voluntary work 10 years later. We agree with tions of workers that determine how likely they are to engage Vangen et al. (2021), who state that “being active in older in bridge employment. age may depend on earlier activity patterns” (p. 2). Thus, We hypothesize that psychological empowerment plays we believe that, if psychological empowerment has been such an important mediating role between organizational experienced in the work context, it is more likely that indi- factors and the bridge employment decision making by viduals will seek opportunities in this life arena to continue capturing employees’ subjective experiences of their work. experiencing empowerment. Compared to cognitions of outcome expectations (see Wöhr- mann et al., 2013) or experienced meaningfulness of work Hypothesis 2b Psychological empowerment before retire- (Peng, 2022; Peng et al., 2020), psychological empower- ment is more strongly related to bridge employment than to ment is a broader construct focusing on the present state volunteering after retirement. of employees’ interpretations of their work and therefore may represent a more immediate predictor of bridge employ- Finally, drawing from the model of the bridge employ- ment. Furthermore, in contrast to more general job-related ment decision-making process by Beehr and Bennett (2015), constructs that have been identified as antecedents of bridge we suggest that the relationship between psychological employment, such as job satisfaction, commitment, intrinsic empowerment and bridge employment is influenced by motivation, or work stress (see, e.g., Beehr & Bennett, 2015; boundary conditions. Beehr and Bennett (2015) built on the Wang et  al., 2008), psychological empowerment allows process model of retirement (Wang & Shultz, 2010), add- for more fine-grained insights, as it consists of four facets ing situational constraints (e.g., unexpected factors such as that might serve as potential leverages to promote bridge additional expenses or health problems) as boundary condi- employment. tions to the model. We assume that physical limitations are From a theoretical point of view, we draw from continuity an important boundary condition, as poor health is one of the theory (Atchley, 1989), which we have introduced above. most important reasons for an early exit from the labor force Atchley (1989) argues that “adults employ concepts of their (see, e.g., van Rijn et al., 2014; Vanajan et al., 2020) and past to conceive of their future and structure their choices has been shown to interact with worker characteristics such in response to the changes brought about by normal aging” as age and race (see a review by Fisher et al., 2016). High (p. 183). Retirement constitutes such a normal change. This physical limitations reflect the individual’s inability to per - striving for continuity in one’s self-concept leads older form daily activities such as sitting for more than two hours workers to envision themselves being and acting rather or lifting something, implying a “limited capacity to meet similarly in retirement as in the present moment (Feldman, the requirements of core social, familial, and occupational 2013). Thus, following continuity theory (Atchley, 1989), roles” (Gayman et al., 2008, p. 219). From a theoretical per- we suggest that older workers who have experienced psycho- spective, continuity theory (Atchley, 1989) might explain the logical empowerment during working life will choose cir- potential boundary effect of physical limitations on bridge cumstances after retirement that provide them with similar employment decision making. Continuity theory (Atchley, 1 3 Empowered to Stay Active: Psychological Empowerment, Retirement Timing, and Later Life Work 1989) states that “external continuity is usually an effective data (i.e., 2 SD below or above the mean). Missing values adaptive strategy but can sometimes be maladaptive” (p. were generally accounted for by full-information maximum 189). According to Atchley (1989), external continuity (e.g., likelihood (FIML) estimation or listwise deletion, depending continuing to work after retirement) becomes maladaptive on the model used (see section on statistical analyses below). when the physical and mental abilities required for conti- Participants at T1 were between 54 and 65  years old nuity are seriously impaired. We hypothesize that the rela- (M = 61.97, SD = 1.95); 50% were female, and 77.1% indi- tionship between psychological empowerment and bridge cated being in a committed relationship. According to the employment depends on the level of physical limitations, International Standard Classification of Education 97 clas- such that the relationship between psychological empower- sification, most participants (58.6%) had a high level of edu- ment and bridge employment can only unfold when physical cation (up to a doctoral degree or equivalent). At T1, partici- limitations are low. Consequently, we argue that older people pants worked 37.3 h per week on average (SD = 12.2). Ten are more likely to use adaptive strategies—that is, to retire percent were blue-collar workers, 51.4% white-collar work- when sufficient physical resources for continuing to work ers, 23.3% civil servants, 13.3% self-employed, and 1.4% are no longer available—than to force external continuity, supported their family business. Most participants worked even though they might report high levels of psychological either in small companies (less than 50 employees; 36.7%) or empowerment. large companies (more than 250 employees; 37.1%). Table 1 summarizes the sample characteristics measured at T1. Hypothesis 2c The positive relationship between psychologi- cal empowerment and bridge employment is moderated by Measures physical limitations, such that the relationship is stronger for retirees with low levels of physical limitations and weaker The interview guide for the TOP study was developed and for retirees with high levels of physical limitations. pre-tested by the German Federal Institute for Population Table 1 Sample characteristics at T1 Method n/M %/SD Participants and Procedure Gender Female 105 50 We used data from the TOP study, a longitudinal, popu- Male 105 50 lation-representative panel study, carried out on behalf Age M = 61.97 SD = 1.95 of the German Federal Institute for Population Research Working hours M = 37.29 SD = 12.20 (Mergenthaler et  al., 2017). It investigated older adults’ Relationship status potential in Germany and their transition from working Committed relationship 162 77.14 life to retirement. Standardized and computer-assisted tel- No committed relationship 48 22.86 ephone interviews were conducted by trained interviewers. Education (ISCED) To ensure random sampling of households with a landline Low 4 1.91 number, the Gabler-Häder design was used, which consid- Middle 83 39.53 ers listed and unlisted numbers equally (Sackreuther et al., High 123 58.57 2016). For our research question, we used data from the first Position wave (T1; January–April 2013) and the second wave (T2; Blue-collar 21 10.00 November 2015–February 2016). At T1, a total of 5,002 White-collar 108 51.43 respondents aged 55–70 years completed the interview, 50% Civil servants 49 23.33 of whom also completed the interview at T2 (Mergenthaler Self-employed 28 13.33 et al., 2017). Given our research question, we investigated Support in family business 3 1.43 only respondents who had retired between T1 and T2; thus, Organizational size individuals who (a) were working part- or full-time at T1 Small (less than 50 employees) 77 36.67 and were not yet retired (n = 997) and (b) were retired at Medium-sized (between 50 and 250 37 17.62 T2 (n = 222). We excluded 12 participants who had missing employees) values on the empowerment items, showed logical incon- Large (more than 250 employees) 78 37.14 sistencies regarding their retirement year, or did not indicate Household net income (in euros) M = 2574.82 SD = 2970.37 their retirement year at all. This resulted in a final sample of 210 participants. Furthermore, we replaced the DRA of n = 210; a b c d eight participants with missing values because of extreme n = 203; n = 209; n = 192; n = 207 1 3 I. Drazic et al. Research in cooperation with several universities. Unless reliability, it is increasingly recommended to replace Cron- otherwise indicated, we recoded all items so that higher val- bach’s α with alternative coefficients. This is because of its ues indicated higher scale expressions. unrealistic assumptions, which oftentimes lead to underes- timated reliability scores (McNeish, 2018). Omega coeffi- ARA (T2) cients, for example, are more appropriate measures of reli- ability in case not all items are equally strongly associated ARA was measured with a single item asking participants with the underlying construct (Peters, 2014). This lack of to indicate the year they retired. We then subtracted the par- homogeneity becomes even more likely with shorter scales ticipants’ year of birth from their retirement year. (Berger, 2019). Revelle’s omega (Revelle & Zinbarg, 2009) for the three-faceted empowerment construct in the present DRA (T1) study was 0.61. Although this value is not optimal, it is not surprising: Even in noninterview contexts, lower reliability Participants were asked to indicate the year they would like scores for the empowerment construct are not unusual and to retire. Again, we calculated the difference scores using the deemed acceptable because of the construct’s first-order participants’ year of birth. multidimensionality (Spreitzer, 1995). Schermuly et  al. (2017), who used the same items but with a larger sample Bridge Employment (T2) of the TOP study, conducted exploratory and confirmatory factor analyses, both of which suggested a single empower- Participants were asked whether they were currently engaged ment factor after inspecting the scree plot, Eigenvalues, and in paid work (they were prompted to think of any kind of fit indices. Based on these arguments, we decided to use this paid work after retirement). Of the 210 participants, 78 interview-adapted empowerment scale and its mean score (37.1%) indicated that they were engaged in bridge employ- for subsequent statistical analyses. ment; one participant did not provide the information. Physical Limitations (T2) Volunteering (T2) The level of physical limitations was measured with the Volunteering was measured with the question “Have you question “During the last four weeks, how often were you engaged in any volunteer or honorary activity in the last limited in your activities because of physical problems?” three months, e.g., participating in an association, initiative, Participants assessed this item on a 4-point Likert scale or group?” Of the 210 participants, 100 (47.6%) reported ranging from 1 (very often) to 4 (never). We also recoded volunteering. this item so that higher values reflected greater frequency. Psychological Empowerment (T1) Controls Psychological empowerment was measured with the same Gender (T1) items that were used by Schermuly et al. (2017). The TOP research group adapted the items from Spreitzer’s (1995) In line with other research on bridge employment and retire- original 12-item scale. Especially in the telephone interview ment (see reviews by Beehr & Bennett, 2015 and Fisher context, scales that are designed to be self-administered have et al., 2016), we controlled for gender. been deemed too long and complex (Hughes et al., 2004). Thus, the TOP research group decided to use a selection of Spreitzer’s (1995) empowerment scale and to omit the Internal Locus of Control (T1) impact facet altogether, as prior research suggests a close relationship with the self-determination facet (e.g., Kraimer Meta-analytical findings suggest a strong relationship et al., 1999; Schermuly et al., 2013). Meaning and self-deter- between internal locus of control (a core self-evaluation mination were measured with single items: “My work is very trait) and psychological empowerment (Seibert et al., 2011). important to me” and “During work, I have the opportunity Core self-evaluation traits in turn have been found to be to make decisions autonomously,” respectively. Competence positively related to retirement variables such as retirement was measured with two items: “I face work problems with preparation (Zaniboni et al., 2021). On a 4-point Likert scale calmness because I can always rely on my competencies” ranging from 1 (strongly agree) to 4 (strongly disagree), par- and “Whatever happens in my job, I will master it.” Partici- ticipants were asked to assess the following two statements: pants assessed the items on a 4-point Likert scale ranging “I am in control of my life” and “If I make an effort, I will from 1 (strongly agree) to 4 (strongly disagree). Concerning succeed.” 1 3 Empowered to Stay Active: Psychological Empowerment, Retirement Timing, and Later Life Work Public Sector Employment (T1) bridge employment and volunteering; see Rosseel, 2014). This model was just identified; thus, examining model fit We controlled for employees in the public sector, as they indices to assess model fit is not meaningful. The control have other retirement conditions as private employees in variables gender, locus of control, public sector employee, Germany. For example, the German pension law imposes and equivalized net income of households were included as income restrictions on employees from the public sector who predictors of ARA; in bridge employment and volunteering, have retired (BMAS, 2019). This might alter incentives for we further controlled for temporal latency. bridge employment. Participants in the present study who indicated being civil servants were coded as 1, whereas all other forms of employment were coded as 0. Results Equivalized Net Income of Households (T1) Preliminary Analysis We see pay as a resource provided by the organization, and Table 2 shows descriptive results and intercorrelations. On access to resources has been shown to be positively asso- average, participants wanted to retire at 64.12 years of age ciated with perceptions of psychological empowerment (SD = 2.18) and retired at 63.93 years (SD = 1.83). Psycho- (Seibert et al., 2011). On the other hand, it is well known logical empowerment was positively correlated with all that a person’s financial situation plays an important role dependent variables except for bridge employment and vol- in retirement decisions (see Fisher et al., 2016). We used unteering. Furthermore, psychological empowerment was the modified Organisation for Economic Co-operation and positively correlated with locus of control, gender was posi- Development (OECD) equivalence scale, which adjusts tively correlated with ARA, and volunteering was positively disposable monthly household income by the number and correlated with public sector employment. age of household members (Sackreuther et al., 2016). The median OECD equivalized monthly net income of house- Hypothesis Testing holds was EUR 2076.5. Hypothesis 1 states that higher levels of psychological Temporal Latency (T2) empowerment lead to higher levels of DRA, which in turn result in higher levels of ARA. The observed indirect effect Whether or not someone engages in occupation-related supports this hypothesis (unstandardized indirect effect: activities after retirement may depend on how long ago one 0.82, 95% CI [0.37; 1.24], p < 0.001). Psychological empow- has retired (Fasbender et al., 2016). To calculate temporal erment signic fi antly predicted DRA (β = 0.27, 95% CI [0.73; latency, we subtracted the year of retirement from the sec- 2.20], p < 0.001), which in turn predicted ARA (β = 0.65, ond measurement year. On average, temporal latency was 95% CI [0.44; 0.68], p < 0.001). The direct effect of psycho- 1.75 years (SD = 0.82). logical empowerment on ARA was not significant (β = 0.07, 95% CI [− 0.27, 0.87], p = 0.288), which indicates full medi- Statistical Analyses ation. None of the control variables significantly predicted the ARA (gender, β = 0.07, 95% CI [− 0.11, 0.64], p = 0.174; All hypotheses were tested with the statistical software R locus of control, β = − 0.02, 95% CI [− 0.41, 0.29], p = 0.762; Studio. For Hypothesis 1 (mediation analysis), we conducted public sector employment, β = − 0.07, 95% CI [− 0.75, 0.22], a path analysis using the maximum likelihood estimator with p = 0.246; equivalized net income of households, β = − 0.02, the R package lavaan (Rosseel, 2012). We further specified 95% CI [− 0.00, 0.00], p = 0.838). The model accounts for 5,000 bootstrap samples, a 95% confidence interval (CI), and 7.4% of the variance of DRA and 46.4% of the variance of FIML estimation for missing values. According to Hooper ARA. Following recommendations by Becker et al. (2015), et al. (2008), the model testing Hypothesis 1 showed good fit we also ran the mediation analysis without control variables, indices, χ (4) = 5.571, p = 0.234, CFI = 0.999, TLI = 0.998, which did not change the effects. RMSEA = 0.044, SRMR = 0.032. To test Hypotheses 2a and In Hypothesis 2a, we assumed psychological empower- 2c, we conducted a hierarchical binary logistic regression, ment at T1 to be positively associated with bridge employ- first entering the control variables (Model 1), then entering ment at T2. The results of the binary logistic regression are the main variable (Model 2), and lastly, entering the moder- presented in Table 3. The control variables entered in step ating variable (i.e., physical limitations). For Hypothesis 2b, 1 did not predict engagement in bridge employment. Psy- we conducted a path analysis using the diagonally weighted chological empowerment entered in step 2 significantly pre - least squares (DWLS) estimator to enable the simultane- dicted bridge employment, indicating that individuals with ous inclusion of both categorical dependent variables (i.e., higher levels of psychological empowerment were more 1 3 I. Drazic et al. likely to engage in bridge employment. Model 2 accounts for 4% of the variance in bridge employment. In Hypothesis 2c, we assumed that employees’ physi- cal limitations would moderate the relationship between psychological empowerment and bridge employment. Our data support this hypothesis (see Model 3 in Table 3). The interaction between psychological empowerment and phys- ical limitations was significantly related to the likelihood of engaging in bridge employment. Simple slope analyses revealed that the relationship between psychological empow- erment and bridge employment became stronger when levels of physical limitations were low (i.e., 1 SD below the mean; B = 1.42, p = 0.008, CI [0.36, 2.47]). When physical limita- tions were high (i.e., 1 SD above the mean), the relationship between psychological empowerment and bridge employ- ment was nonsignificant (B = − 0.07, p = 0.910, CI [− 1.22, 1.09]). Figure 2 shows the interaction between psychological empowerment and physical limitations. In Hypothesis 2b, we assumed psychological empower- ment at T1 to be more strongly related to bridge employ- ment than volunteering at T2. The path analysis supports this hypothesis: Psychological empowerment was significant in predicting bridge employment, indicating that individu- als with higher levels of psychological empowerment were more likely to engage in bridge employment. In contrast, psychological empowerment did not significantly predict volunteering. Concerning the set of control variables, none of the variables significantly predicted bridge employment, but volunteering was significantly predicted by public sector employment. The model accounts for 9% of the variance in bridge employment and 12% of the variance in volunteering. The results of the path analysis are summarized in Table 4. We also ran this model without control variables, which slightly decreased the statistical significance of psychologi- cal empowerment, such that it only significantly predicted bridge employment based on a 90% CI [0.05, 0.80]. Additional Analyses To obtain more fine-grained insights, we reran our analy- ses using the three dimensions of psychological empower- ment as individual predictors. For that, we calculated a mean score of the two items measuring the competence facet (self-determination and meaning were measured with single items). Regarding Hypothesis 1, we found that only the competence facet significantly predicted the DRA, which in turn predicted the ARA. The observed indirect effect was significant (unstandardized indirect effect: 0.22, 95% CI [0.04; 0.42], p = 0.025), while t he direct effect was nonsignificant, indicating full mediation. None of the control variables (i.e., gender, locus of con- trol, public sector employee, or equivalized net income 1 3 Table 2 Means (M), standard deviations (SD), and intercorrelations among study variables Variable (unit/scale) M SD 1 2 3 4 5 6 7 8 9 10 1. ARA (years) 63.93 1.83 2. DRA (years) 64.12 2.18 0.67** 3. Bridge employment (0 = no, 1 = yes) 0.37 0.48 0.15* 0.20** 4. Volunteering (0 = no, 1 = yes) 0.48 0.50 − 0.03 − 0.07 − 0.12 5. Psychol. empowerment (scale 1–4) 3.56 0.40 0.23** 0.26** 0.12 -0.06 6. Physical limitations (scale 1–4) 1.68 0.86 − 0.21** − 0.18* − 0.17* − 0.16* − 0.01 7. Gender (1 = female, 2 = male) 1.50 0.50 0.14* 0.11 − 0.10 − 0.06 0.11 0.02 8. Locus of control (scale 1–4) 3.31 0.56 − 0.03 − 0.04 0.02 − 0.01 0.16* − 0.00 − 0.06 9. Public sector (0 = no, 1 = yes) 0.23 0.42 − 0.09 − 0.02 − 0.13 0.24** − 0.03 − 0.10 0.03 − 0.03 10. Equiv. net income (in EUR) 2574.82 2970.37 0.06 0.12 0.00 0.08 0.11 − 0.12 − 0.03 0.00 0.07 11. Latency (years) 1.75 0.82 − 0.11 0.11 0.08 0.07 0.05 0.06 − 0.07 − 0.07 − 0.01 0.14* *p < 0.05. **p < 0.01 a b c n = 210; n = 194; n = 209; n = 207 ARA actual retirement Age, DRA desired retirement age Empowered to Stay Active: Psychological Empowerment, Retirement Timing, and Later Life Work 1 3 Table 3 Results of the binary logistic regression predicting bridge employment Variable Model 1 Model 2 Model 3 B SE B p 95% CI OR B SE B p 95% CI OR B SE B p 95% CI OR Intercept − 0.71 1.00 0.478 [− 2.71, 1.24] − 3.23* 1.64 0.049 [− 6.53, − 0.09] − 0.19 0.23 0.398 [− 0.64, 0.25] Step 1: Control variables  Gender (1 = female, 2 = male) − 0.38 0.30 0.198 [− 0.96, 0.20] 0.68 − 0.45 0.30 0.136 [− 1.04, 0.14] 0.64 − 0.50 0.31 0.108 [− 1.12, 0.11] 0.60  Locus of control 0.06 0.26 0.831 [− 0.46, 0.58] 1.06 − 0.04 0.27 0.885 [− 0.57, 0.49] 0.96 − 0.10 0.28 0.736 [− 0.65, 0.47] 0.91  Public sector − 0.62 0.37 0.092 [− 1.36, 0.08] 0.54 − 0.59 0.37 0.109 [− 1.34, 0.11] 0.55 − 0.70 0.38 0.066 [− 1.48, 0.03] 0.50  Net income − 0.00 0.00 0.957 [− 0.00, 0.00] 1.00 − 0.00 0.00 0.772 [− 0.00, 0.00] 1.00 − 0.00 0.00 0.464 [− 0.00, 0.00] 1.00  Latency 0.18 0.18 0.315 [− 0.17, 0.54] 1.20 0.18 0.18 0.332 [− 0.18, 0.54] 1.20 0.25 0.19 0.201 [− 0.13, 0.62] 1.28 Step 2: Main variable  Psychological empowerment 0.81* 0.41 0.049 [0.03, 1.64] 2.24 0.77 0.43 0.072 [− 0.05, 1.63] 2.15 Step 3: Interaction  Physical limitations − 0.52* 0.20 0.012 [− 0.93, − 0.14] 0.60  Psychological empower- − 0.96* 0.48 0.045 [− 1.92, − 0.01] 0.38 ment × Physical limitations Model Fit:  − 2 log likelihood 264.156 260.055 247.578  Δ − 2 log likelihood (Δdf) 4.101*(1) 12.477**(2)  Pseudo R2 0.02 0.04 0.08 In Model 3, all continuous predictors were mean-centered n = 205 SE Standard error, CI confidence interval, OR odds ratio *p < 0.05, **p < 0.01 I. Drazic et al. Fig. 2 Moderation of the relationship between psycho- logical empowerment and bridge employment by physical limitations Table 4 Results of the path analysis predicting bridge employment and volunteering Variable Bridge employment Volunteering B SE B β p 95% CI B SE B β p 95% CI Control variables  Gender − 0.27 0.19 − 0.13 0.156 [− 0.68, 0.06] − 0.15 0.19 − 0.07 0.437 [− 0.53, 0.21]  Locus of control − 0.03 0.18 − 0.02 0.877 [− 0.35, 0.36] 0.02 0.17 0.01 0.917 [− 0.33, 0.32]  Public sector − 0.36 0.24 − 0.15 0.132 [− 0.90, 0.03] 0.72** 0.23 0.29** 0.002 [0.31, 1.24]  Net income − 0.00 0.00 − 0.03 0.915 [− 0.00, 0.00] 0.00 0.00 0.10 0.547 [− 0.00, 0.00]  Latency 0.11 0.12 0.09 0.350 [− 0.12, 0.35] 0.10 0.12 0.08 0.400 [− 0.13, 0.33] Main variable  Psychological empowerment 0.51* 0.25 0.19* 0.040 [0.04, 1.00] − 0.19 0.25 − 0.07 0.444 [− 0.67, 0.31] N = 205. For gender, 1 = female, 2 = male *p < 0.05, **p < 0.01 CI confidence interval of households) significantly predicted the ARA. Figure 3 Discussion shows the model with standardized parameter estimates. Regarding Hypothesis 2, the results showed that none Especially in aging societies, successfully motivating of the single facets of psychological empowerment sig- older employees to extend their working lives is essen- nificantly predicted bridge employment or volunteering. tial. Using a sample of a longitudinal panel study in Ger- As in the model with psychological empowerment as a many, we found that psychologically empowered employ- unitary construct, none of the control variables signifi- ees not only desired to retire later but indeed retired later cantly predicted bridge employment, and only public (Hypothesis 1). Second, we tested whether psychological sector employment significantly predicted volunteering empowerment predicts engagement in bridge employment (β = 0.30, 95% CI [0.29; 1.22], p = 0.002). All additional (Hypothesis 2a). Our findings also support this hypothesis. analyses are available upon request from the first author. 1 3 Empowered to Stay Active: Psychological Empowerment, Retirement Timing, and Later Life Work Fig. 3 Results of the Mediation Hypothesis with the Individual Facets of Psychological Empowerment as Predictors. Note. *p < 0.05. **p < 0.01. All displayed coefficients are standardized. Control variables are omitted in the figure. Dashed line indicates full mediation Furthermore, we hypothesized that psychological empow- literature, our findings contribute to the literature in three erment better predicts bridge employment than volunteer- unique ways. First, they extend findings from Schermuly ing (Hypothesis 2b). Psychological empowerment was et al. (2017), who found that psychologically empowered indeed significant in predicting bridge employment while employees desired to retire later. By including ARA, we showing a nonsignificant relationship to volunteering; respond to calls for more research on retirement outcomes however, effect sizes were small. Finally, we found that that goes beyond intentions (see Browne et al., 2018). Sec- physical limitations moderated the relationship between ond, we transferred psychological empowerment to the psychological empowerment and bridge employment, such retirement context, suggesting that retirement age, along that the relationship was stronger when physical limita- with constructs such as turnover intention and organizational tions were low and nonexistent when physical limitations commitment, is another important outcome of psychologi- were high. In the next section, we will interpret these find- cal empowerment in the organizational context. Third, our ings in light of existing theory and empirical studies. results support the theory of planned behavior (Ajzen, 1991) as well as its theoretical extension that proposes intentions Theoretical Implications (in our case, DRA) as a mediating variable (see Perugini & Bagozzi, 2001). DRA and ARA were highly correlated in In general, our findings regarding all four hypotheses sup- our sample (r = 0.67), which implies that the desire to retire port the process model of retirement (e.g., Wang & Shultz, at a certain time is realizable. This might be due to the Ger- 2010) in its assumption that job and organizational factors man retirement system, which allows for flexible timing. In might act as antecedents of retirement decisions. Our find- Sweden, where there is also no fixed retirement age, panel ings contribute to the model first by adding psychological studies have found similarly strong relationships between empowerment as a concrete antecedent at the job and organi- DRA and ARA (Örestig et al., 2013; Solem et al., 2016). In zation level. Second, as psychological empowerment was the present study, however, DRA was slightly higher than related not only to retirement timing but also to the bridge ARA, indicating that the participants retired earlier than they employment decision making, our findings suggest that the desired. Solem et al. (2016) identified conditions, such as same job and organizational factors experienced during poor health and education, that increased the risk of retiring working life can influence retirement decisions across phases earlier than preferred. More research is needed to further and have a lasting effect on the retirees’ lives. investigate boundary conditions of the relationship between The results related to Hypothesis 1 are in line with DRA and ARA, as well as the psychological consequences previous findings showing that person-job variables are of not retiring at one’s DRA. important factors for the DRA and the ARA. For exam- Additional analyses of the role of the individual empow- ple, research indicates that job control is associated with erment facets revealed that only the competence facet was later retirement intentions and later actual retirement (see significantly related to the DRA and the ARA, resulting in a review Browne et al., 2018). Going beyond the existing full mediation by the DRA. This finding is interesting and 1 3 I. Drazic et al. might be in line with research on employability, which can activities did not compete for the retirees’ resources in our be defined as “the continuous fulfilling, acquiring or creating study. Vangen et al. (2021) did find a negative relationship of work through the optimal use of competences” (Heijde between paid work and volunteering in a sample of Norwe- & Van Der Heijden, 2006, p. 453). Researchers have argued gian seniors aged 62–75 over a 10-year period. However, that employability is an important factor in the motivation the negative relationship was found only in participants to continue working (see, e.g., Pak et al., 2019). with full-time employment (not among those with part-time The results regarding Hypothesis 2a are in line with pre- employment), and the relationship did not apply for partici- vious findings on antecedents of bridge employment at the pants with former experience in voluntary work. The authors person-job level. For example, a study by Fasbender et al. conclude that, in order to increase older people’s engage- (2016) found that the social meaning of work (e.g., having ment in both paid and unpaid work, it might be beneficial to contact with others) and the personal meaning of work (e.g., stimulate combinations of both forms in late careers (Vangen having a meaningful task) were positively related to the et al., 2021). Thus, while psychological empowerment at likelihood of engaging in bridge employment. Our results the workplace could be used to stimulate paid work after differ from those of Fasbender et al. (2016) in that they do retirement (i.e., bridge employment), informing older indi- not focus on employees’ beliefs about work in general, but viduals early about possibilities for volunteering could help on employees’ cognitions that are shaped by their immedi- to engage them in volunteer work before they retire (e.g., ate work environment. From a theoretical perspective, our Hansen & Slagsvold, 2020). results support continuity theory (Atchley, 1989), as indi- Additional analyses regarding Hypotheses 2a and 2b viduals who were psychologically empowered before retire- revealed that none of the facets of the empowerment con- ment showed more engagement in paid work after retirement struct individually predicted bridge employment. This might (i.e., engagement in bridge employment). This could imply indicate that the relationship between psychological empow- that psychologically empowered individuals are likely to erment and bridge employment is driven by the “Gestalt” seek out contexts in which they can continue to experience of psychological empowerment (i.e., the joint presence of psychological empowerment. This finding is consistent with all facets; see Spreitzer, 1995). However, it might also have empowerment theory, according to which psychologically resulted from methodological limitations. We will discuss empowered employees are more likely to take an active this in the Limitations section. orientation toward work and go the “extra mile” Spreitzer Finally, our results regarding Hypothesis 2c indicate that (2008). Thus, it seems plausible that they are less likely to the level of physical limitations moderates the relationship completely withdraw from work compared to less empow- between psychological empowerment and bridge employ- ered older individuals. In light of continuity theory, one may ment decision making. This finding is in line with previous conclude that psychological empowerment during working research on constraining variables. For example, Zhan et al. life—or its continued experience after retirement—might (2013) identified economic stress as a boundary condition help older individuals to maintain stability during potentially for the relationship between career commitment and career- disruptive life changes such as retirement. Future research based bridge employment decision making and the rela- could test this assumption by investigating the relationship tionship between affective commitment to organization and between psychological empowerment during working life organization-based bridge employment decision making, and post-retirement outcomes such as retirement satisfac- such that high economic stress weakened both relationships. tion, psychological health, and/or physical health. Both these findings and our finding support the model by In Hypothesis 2b, we compared paid and unpaid forms Beehr and Bennett (2015), which stresses the role of situa- of later life work. Our results provide indications that psy- tional constraints in the bridge employment decision-making chological empowerment is more strongly associated with process. Future research could combine our study with the bridge employment than with volunteering. Results regard- study by Zhan et al. (2013) by investigating whether empow- ing Hypothesis 2b might also support continuity theory ered employees are more likely to take bridge employment (Atchley, 1989), according to which middle-aged and older at the same organization and identify constraining variables individuals aim to preserve structures by “applying familiar that could be at play in this relationship. strategies in familiar arenas of life” (p. 183)—context and familiarity have likely played a role in maintaining continu- Practical Implications ity in the present study. Furthermore, the fact that bridge employment and volunteering were not significantly cor - One practical implication of the strong relationship between related might speak against the so-called trade-off hypoth- DRA and ARA could be for policymakers to assess DRA esis (see Vangen et al., 2021): Engaging in bridge employ- among older employees to predict changes in the public ment did not decrease the likelihood of being engaged in pension system. Likewise, organizations might consider volunteering and vice versa. Thus, it seems that the two assessing their older employees’ DRAs for more accurate 1 3 Empowered to Stay Active: Psychological Empowerment, Retirement Timing, and Later Life Work staff planning. Furthermore, organizations wanting to retain and not (exclusively) vice versa. Another potential limitation older employees—whether through delayed retirement or is that most of the constructs were measured by single items bridge employment—could consider enhancing psycho- or reduced scales due to the telephone interview method. For logical empowerment cognitions in their employees. They example, the empowerment construct was limited to four might apply general approaches that have been identified as items. This factor plus the constructs’ multidimensionality contextual antecedents of psychological empowerment, for led to a low omega coefficient. Future research could repli- example, high-performance managerial practices and soci- cate our findings with the standard instrument by Spreitzer opolitical support (see meta-analyses by Llorente-Alonso (1995) using an online questionnaire. Furthermore, future et al., 2023 and Seibert et al., 2011). More specific human research could also test the measurement invariance of resource management measures for older employees should the empowerment scale related to age, in particular when also be considered. For example, measures against age dis- investigating more age-diverse samples (for a brief tutorial, crimination seem important, as research indicates negative see Guo et al., 2023). Another potential limitation, which associations between age discrimination and psychological could be at least partly due to measurement problems, is empowerment, which in turn leads to a lower DRA (Scher- that the effect sizes seemed relatively small. However, recent muly et al., 2014). Optimally, specific measures for empow - research suggests that effect sizes that seem small when ering older employees should be developed. One practi- judged by conventional standards (e.g., by Cohen’s guide- cal suggestion for policymakers thus might be to consider lines) may not be weak at all when more domain-specific funding research in this interdisciplinary research field, as standards are applied (e.g., by evaluating correlations from knowledge in this area could help motivate older workers to multiple relevant meta-analyses and defining cut-offs based remain active longer and thereby reduce the financial burden on the 25th, 50th, and 75th percentiles of reported effect on public pension and health care systems in aging socie- sizes; see Davenport et al., 2022). Moreover, even minor ties. It would be especially important to encourage research individual differences in the timing of retirement and/or on psychological empowerment that investigates workers the choice of bridge employment are likely to have drastic older than 50 years, as a recent meta-analysis by Llorente- effects on social security systems through cumulative effects, Alonso et al. (2023) indicates that few empirical studies have given how many older workers are likely to retire in the included this population. Finally, organizations that want to coming years. The study sample itself is another methodo- retain older employees after retirement should start promot- logical limitation. Participants were mostly highly educated ing the physical health of their employees early because, white-collar workers. This might have inflated the correla- according to our findings, the relationship between psycho- tion between DRA and ARA, as research indicates that a logical empowerment and bridge employment disappears lower education level increases the probability of retiring when employees report high levels of physical limitations. before one’s DRA (Solem et al., 2016). Furthermore, our Research informs us that one way leaders can promote the participants were employed individuals aged between 54 and subjective health of their subordinates is to practice respect- 65 years at the first measurement. Thus, we investigated a ful leadership, which in turn might even increase older work- highly selective sample of the population, excluding individ- ers’ DRA (see Wöhrmann et al., 2017). uals who were unemployed at that time or might have already retired involuntarily. Future research should therefore also Limitations examine workers aged 50 or younger to explore involuntary retirement and ways to prevent it. Another limitation is that The limitations of the present study result mainly from the the findings can be extrapolated to other countries only to a archival nature of the panel data. First, as psychological limited extent due to differing retirement and work regula- empowerment and DRA were collected at the same meas- tions. Finally, it is important for future research to examine urement point, common method bias may have altered their the incremental validity of psychological empowerment for relationship (Podsakoff et al., 2012). At least the criterion predicting retirement timing and engagement in later life in this mediation model, ARA, was measured with a con- work against alternative, conceptually meaningful factors siderable time lag (i.e., three years) and presented a more such as psychological capital, person–environment fit, or objective measure. However, we still cannot draw causal job involvement. conclusions about the relationship between psychological empowerment and engagement in later life work. Although we had two measurement points, the focal variables were Conclusion not measured at both time points. Future researchers should conduct longitudinal or preferably interventional studies to This study aimed to investigate the relationship between substantiate our assumption that psychological empower- psychological empowerment before retirement and later ment increases the likelihood of engaging in later life work life work. The three most important findings of this study 1 3 I. Drazic et al. BMAS. (2019). Hinzuverdienst bei Renten und Pensionen (Addi- are as follows: (1) Older workers with higher levels of tional earnings to pensions). https:// www. einfa ch- teilh aben. psychological empowerment not only wanted to retire later de/ DE/ AS/ Themen/ Finan ziell eHilf en/ Rente nPens ionen/ Hinzu but indeed retired later, (2) older workers with higher lev- v er di e ns tR ente n P ensi onen/ hinzu v er di ens tr ente n pens i onen_ els of psychological empowerment were more likely to node. html# doc11 17827 4body Text3 BMAS. (2020). Pension projections exercise 2021. Federal Ministry engage in bridge employment, and (3) the positive rela- of Labour and Social Affairs. tionship between psychological empowerment and bridge Browne, P., Carr, E., Fleischmann, M., Xue, B., & Stansfeld, S. employment disappeared when older workers experienced A. (2018). The relationship between workplace psychosocial high levels of physical limitations. These findings suggest environment and retirement intentions and actual retirement: A systematic review. European Journal of Ageing, 16(1), 73–82. that psychological empowerment might help to retain older https:// doi. org/ 10. 1007/ s10433- 018- 0473-4 workers in the workforce. At the same time, our study Conger, J. A., & Kanungo, R. N. (1988). The empowerment pro- points to the limitations of psychological empowerment, cess: Integrating theory and practice. Academy of Management as the latter became insignificant in predicting bridge Review, 13(3), 471–482. Czepek, J., & Weber, E. (2015). Flexibilität beim Übergang in die employment when high levels of physical limitations Rente, Aktuelle Berichte (Flexibility during the retirement tran- were reported. Thus, we hope that our study sensitizes sition) (No. 2). Institute for Employment Research (IAB). both researchers and practitioners for the importance of Davenport, M. K., Ruffin, M. A., Oxendahl, T. A., McSpedon, M. R., a holistic approach when studying or promoting longer & Beier, M. E. (2022). “Small” effects, big problems. Work, Aging and Retirement, 8(4), 368–370. https:// doi. org/ 10. 1093/ workar/ working lives. Lastly, we encourage future researchers to waac0 12 replicate our findings outside the interview context. Dorn, D., & Sousa-Poza, A. (2010). “Voluntary” and “Involuntary” early retirement: An international analysis. Applied Economics, 42(4), 427–438. Funding Open Access funding enabled and organized by Projekt DRV. (2020). Altersrentner: So viel können Sie hinzuverdienen (Retir- DEAL. ees: This is how much you can earn additionally). https:// www. deuts c he- r ente nvers ic her ung. de/ Shar e dDocs/ Downl oads/ DE/ Open Access This article is licensed under a Creative Commons Attri- Brosc hueren/ natio nal/ alter srent ner_ hinzu verdi enst. html bution 4.0 International License, which permits use, sharing, adapta- Fasbender, U., Wang, M., Voltmer, J.-B., & Deller, J. (2016). The tion, distribution and reproduction in any medium or format, as long meaning of work for post-retirement employment decisions. as you give appropriate credit to the original author(s) and the source, Work, Aging and Retirement, 2(1), 12–23. https://d oi.o rg/1 0.1 093/ provide a link to the Creative Commons licence, and indicate if changes workar/ wav015 were made. The images or other third party material in this article are Fasbender, U., Wöhrmann, A. M., Wang, M., & Klehe, U. C. (2019). included in the article's Creative Commons licence, unless indicated Is the future still open? The mediating role of occupational future otherwise in a credit line to the material. If material is not included in time perspective in the effects of career adaptability and aging the article's Creative Commons licence and your intended use is not experience on late career planning. Journal of Vocational Behav- permitted by statutory regulation or exceeds the permitted use, you will ior, 111, 24–38. https:// doi. org/ 10. 1016/j. jvb. 2018. 10. 006 need to obtain permission directly from the copyright holder. To view a Feldman, D. C. (2013). Feeling like it’s time to retire: A fit perspec- copy of this licence, visit http://cr eativ ecommons. or g/licen ses/ b y/4.0/ . tive on early retirement decision. In M. Wang (Ed.), The Oxford handbook of retirement (pp. 280–292). Oxford University Press. https:// doi. org/ 10. 1093/ oxfor dhb/ 97801 99746 521. 013. 0109 Fisher, G. G., Chaffee, D. S., & Sonnega, A. (2016). Retirement timing: A review and recommendations for future research. Work, Aging References and Retirement, 2(2), 230–261. https:// doi. org/ 10. 1093/ workar/ waw001 Ajzen, I. (1991). The theory of planned behavior. Organizational Garcia, P. R. J. M., Amarnani, R. K., Bordia, P., & Restubog, S. L. Behavior and Human Decision Processes, 50, 179–211. https:// D. (2021). When support is unwanted: The role of psychological doi. org/ 10. 15288/ jsad. 2011. 72. 322 contract type and perceived organizational support in predicting Atchley, R. C. (1989). A continuity theory of normal aging. The bridge employment intentions. Journal of Vocational Behavior, Gerontologist, 29(2), 183–190. https:// doi. org/ 10. 1093/ geront/ 125, 103525. https:// doi. org/ 10. 1016/j. jvb. 2020. 103525 29.2. 183 Gayman, M. D., Turner, R. J., & Cui, M. (2008). Physical limitations Becker, T. E., Atinc, G., Breaugh, J. A., Carlson, K. D., Edwards, and depressive symptoms: Exploring the nature of the association. J. R., & Spector, P. E. (2015). Statistical control in correla- The Journals of Gerontology Series b: Psychological Sciences tional studies: 10 essential recommendations for organizational and Social Sciences, 63(4), S219–S228. https:// doi. org/ 10. 1093/ researchers. Journal of Organizational Behavior, 37(2), 157– geronb/ 63.4. S219 167. https:// doi. org/ 10. 1002/ job. 2053l Giandrea, M. D., Cahill, K. E., & Quinn, J. (2010). The role of re-entry Beehr, T. A., & Bennett, M. M. (2015). Working after retirement: in the retirement process (Vol. 439). US Department of Labor, US Features of bridge employment and research directions. Work, Bureau of Labor Statistics, Office of Productivity and Technology. Aging and Retirement, 1(1), 112–128. https:// doi. org/ 10. 1093/ Grigoli, F., Koczan, Z., & Topalova, P. W. (2021). Calling older work- workar/ wau007 ers back to work. Applied Economics Letters, 29(6), 559–566. Berger, U. (2019). Interne Konsistenz von Fragebogen-Skalen: Alpha https:// doi. org/ 10. 1080/ 13504 851. 2021. 18762 05 oder Omega (Internal consistency of questionnaire scales: Guo, F., Min, H., Jex, S., & Choi, Y. (2023). Old enough to perceive Alpha or omega). Psychotherapie Psychosomatik Medizinische things differently? Detecting measurement invariance across age Psychologie, 69(08), 348–349. groups using item-focused tree. Work, Aging and Retirement, 9(1), 59–70. https:// doi. org/ 10. 1093/ workar/ waac0 04 1 3 Empowered to Stay Active: Psychological Empowerment, Retirement Timing, and Later Life Work Hansen, T., & Slagsvold, B. (2020). An “army of volunteers”? Engage- Co-operation and Development. https://www .oecd. or g/els/ em p/ ment, motivation, and barriers to volunteering among the baby avera ge- effec tive- age- of- retir ement. htm boomers. Journal of Gerontological Social Work, 63(4), 335–353. OECD. (2019b). Pensions at a glance 2019: OECD and G20 indica- https:// doi. org/ 10. 1080/ 01634 372. 2020. 17582 69 tors. Organisation for Economic Co-Operation and Develop- Heijde, C. M. V. D., & Van Der Heijden, B. I. J. M. (2006). A compe- ment. https:// doi. org/ 10. 1787/ b6d3d cfc- en tence-based and multidimensional operationalization and meas- Örestig, J., Strandh, M., & Stattin, M. (2013). A wish come true? urement of employability. Human Resource Management, 45(3), A longitudinal analysis of the relationship between retire- 449–476. https:// doi. org/ 10. 1002/ hrm. 20119 ment preferences and the timing of retirement. Journal of Hofäcker, D., & Naumann, E. (2015). The emerging trend of work Population Ageing, 6(1–2), 99–118. h ttp s: // do i. or g/ 10 . 1 007/ beyond retirement age in Germany. Zeitschrift Für Gerontolo-s12062- 012- 9075-7 gie Und Geriatrie, 48(5), 473–479. https:// doi. or g/ 10. 1007/ Pak, K., Kooij, D. T. A. M., De Lange, A. H., & Van Veldhoven, s00391- 014- 0669-y M. J. P. M. (2019). Human resource management and the abil- Holtemöller, O., Kooths, S., Michelsen, C., Schmidt, T., & Wollmer- ity, motivation and opportunity to continue working: A review shäuser, T. (2021). Gemeinschaftsdiagnose: Pandemie verzögert of quantitative studies. Human Resource Management Review, Aufschwung – Demografie bremst Wachstum (Pandemic delays 29(3), 336–352. https:// doi. org/ 10. 1016/j. hrmr. 2018. 07. 002 upswing—Demography slows growth). Wirtschaftsdienst, 101(5), Peng, Y. (2022). A two-study examination of age discrimination, 353–357. work meaningfulness, and bridge employment intentions. Hooper, D., Coughlan, J., & Mullen, M. R. (2008). Structural equa- Career Development International, 27(3), 391–407. https:// tion modelling: Guidelines for determining model fit. Electronic doi. org/ 10. 1108/ CDI- 11- 2021- 0295 Journal on Business Research Methods, 6(1), 53–60. Peng, Y., Xu, X., Jex, S. M., & Chen, Y. (2020). The roles of job- Hughes, M. E., Waite, L. J., Hawkley, L. C., & Cacioppo, J. T. (2004). related psychosocial factors and work meaningfulness in A short scale for measuring loneliness in large surveys: Results promoting nurses’ bridge employment intentions. Journal of from two population-based studies. Research on Aging, 26(6), Career Development, 47(6), 701–716. https:// doi. org/ 10. 1177/ 655–672. https:// doi. org/ 10. 1177/ 01640 27504 26857408948 45318 824666 Kanter, R. M. (1977). Men and women of the corporation. Basic Perugini, M., & Bagozzi, R. P. (2001). The role of desires and antici- Books. pated emotions in goal-directed behaviours: Broadening and Kraimer, M. L., Seibert, S. E., & Liden, R. C. (1999). Psychological deepening the theory of planned behaviour. British Journal of empowerment as a multidimensional construct: A test of con- Social Psychology, 40(1), 79–98. https://doi. or g/10. 1348/ 01446 struct validity. Educational and Psychological Measurement, 66011 64704 59(1), 127–142. https:// doi. org/ 10. 1177/ 00131 64499 591009 Peters, G.-J.Y. (2014). The alpha and the omega of scale reliability Lee, M. Y., & Edmondson, A. C. (2017). Self-managing organi- and validity: Why and how to abandon Cronbach’s alpha and the zations: Exploring the limits of less-hierarchical organizing. route towards more comprehensive assessment of scale quality. Research in Organizational Behavior, 37, 35–58. https:// doi. The European Health Psychologist, 16(2), 56–69. https://doi. or g/ org/ 10. 1016/j. riob. 2017. 10. 00210. 31234/ osf. io/ h47fv Li, Y., Turek, K., Henkens, K., & Wang, M. (2022). Retaining retire- Podsakoff, P. M., MacKenzie, S. B., & Podsakoff, N. P. (2012). Sources ment-eligible older workers through training participation: The of method bias in social science research and recommendations on joint implications of individual growth need and organizational how to control it. Annual Review of Psychology, 63(1), 539–569. climates. Journal of Applied Psychology. https:// doi. or g/ 10. https:// doi. org/ 10. 1146/ annur ev- psych- 120710- 100452 1037/ apl00 01065 Revelle, W., & Zinbarg, R. (2009). Coefficientes alpha, beta, omega Liden, R. C., & Arad, S. (1996). A power perspective of empower- and the GLB. Psychometrika, 74(1), 145–154. ment and work groups: Implications for human resources man- Rosseel, Y. (2012). Lavaan: An R package for structural equation mod- agement research. Research in Personnel and Human Manage- eling. Journal of Statistical Software, 48(2), 1–36. https://doi. or g/ ment, 14, 205–251.10. 18637/ jss. v048. i02 Llorente-Alonso, M., García-Ael, C., & Topa, G. (2023). A meta- Rosseel, Y. (2014). Structural equation modeling with categorical vari- analysis of psychological empowerment: Antecedents, organiza- ables (summer school – using R for personality research, August tional outcomes, and moderating variables. Current Psychology. 23–28, Bertinoro, Italy). https://perso nality -pr oject. or g/r/tut orials/ https:// doi. org/ 10. 1007/ s12144- 023- 04369-8summe rscho ol. 14/ rosse el_ sem_ cat. pdf Malik, M., Sarwar, S., & Orr, S. (2021). Agile practices and per- Rouzet, D., Caldera Sánchez, A., Renault, T., & Roehn, O. (2019). formance: Examining the role of psychological empowerment. Fiscal challenges and inclusive growth in ageing societies (OECD International Journal of Project Management, 39(1), 10–20. Economic Policy Papers, No. 27). OECD Publishing. https:// doi. org/ 10. 1016/j. ijpro man. 2020. 09. 002 Sackreuther, I., Schröber, J., Cihlar, V., Mergenthaler, A., Micheel, F., McNeish, D. (2018). Thanks coefficient alpha, we’ll take it from & Schill, G. (2016). TOP - Transitions and Old Age Potential: here. Psychological Methods, 23(3), 412–433. https:// doi. org/ Methodology report on the study. Federal Institute for Population 10. 1037/ met00 00144 Research. Mergenthaler, A., Cihlar, V., Micheel, F., Sackreuther, I., Riedl, S., Schermuly, C. C. (2016). New Work—Gute Arbeit gestalten (New & Maruhn, S. (2017). TOP - Transitions and Old Age Potential: work—Designing good work). Haufe-Lexware. Data and methodology report for the second wave of the study. Schermuly, C. C., Büsch, V., & Graßmann, C. (2017). Psychological Federal Institute for Population Research. empowerment, psychological and physical strain and the desired Naghavi, M. S., Jofreh, M. G., Vaezi, R., & Ghorbanizadeh, V. retirement age. Personnel Review, 46(5), 950–969. https://doi. or g/ (2019). Empowerment of the professional ageing workforce: 10. 1108/ PR- 06- 2015- 0159 A review and development of a model. European Journal of Schermuly, C. C., Deller, J., & Büsch, V. (2014). A research note on International Management, 13(2), 269–286. https://doi. or g/10. age discrimination and the desire to retire. Research on Aging, 1504/ EJIM. 2019. 098185 36(3), 382–393. https:// doi. org/ 10. 1177/ 01640 27513 508288 OECD. (2019a). Ageing and employment policies—Statistics on Schermuly, C. C., Meyer, B., & Dämmer, L. (2013). Leader-mem- average effective age of retirement. Organisation for Economic ber exchange and innovative behavior: The mediating role of 1 3 I. Drazic et al. psychological empowerment. Journal of Personnel Psychology, Vanajan, A., Bültmann, U., & Henkens, K. (2020). Why do older work- 12(3), 132–142. https:// doi. org/ 10. 1027/ 1866- 5888/ a0000 93 ers with chronic health conditions prefer to retire early? Age and Seibert, S. E., Wang, G., & Courtright, S. H. (2011). Antecedents and Ageing, 49(3), 403–410. https:// doi. org/ 10. 1093/ ageing/ afz180 consequences of psychological and team empowerment in organi- Vangen, H., Hellevik, T., & Herlofson, K. (2021). Associations zations: A meta-analytic review. Journal of Applied Psychology, between paid and unpaid work among Norwegian seniors: Com- 96(5), 981–1003. https:// doi. org/ 10. 1037/ a0022 676 petition, complementarity or continuity? European Journal of Shultz, K. S. (2003). Bridge employment: Work after retirement. In G. Ageing. https:// doi. org/ 10. 1007/ s10433- 021- 00615-9 A. Adams & T. A. Beehr (Eds.), Retirement: Reasons, processes, Vignoli, M., Zaniboni, S., Chiesa, R., Alcover, C.-M., Guglielmi, D., and results (pp. 214–241). Springer. & Topa, G. (2021). Maintaining and engaging older workers at Shultz, K. S., & Wang, M. (2011). Psychological perspectives on the work: The trigger role of personal and psychosocial resources. The changing nature of retirement. American Psychologist, 66(3), International Journal of Human Resource Management, 32(8), 170–179. https:// doi. org/ 10. 1037/ a0022 411 1731–1753. https:// doi. org/ 10. 1080/ 09585 192. 2019. 15792 52 Solem, P. E., Syse, A., Furunes, T., Mykletun, R. J., De Lange, A., Wang, M., Adams, G. A., Beehr, T. A., & Shultz, K. S. (2009). Bridge Schaufeli, W., & Ilmarinen, J. (2016). To leave or not to leave: employment and retirement: Issues and opportunities during the Retirement intentions and retirement behaviour. Ageing and Soci- latter part of one’s career. In S. G. Baugh & S. E. Sullivan (Eds.), ety, 36(2), 259–281. https://doi. or g/10. 1017/ S0144 686X1 40011 35 Maintaining focus, energy, and options over the career. Informa- Sousa-Ribeiro, M., Bernhard-Oettel, C., Sverke, M., & Westerlund, H. tion Age Publishing. (2021). Health- and age-related workplace factors as predictors of Wang, M., & Shi, J. (2014). Psychological research on retirement. preferred, expected, and actual retirement timing: Findings from Annual Review of Psychology, 65, 209–233. https:// doi. org/ 10. a Swedish cohort study. International Journal of Environmental 1146/ annur ev- psych- 010213- 115131 Research and Public Health, 18(5), 2746. https://doi. or g/10. 3390/ Wang, M., & Shultz, K. S. (2010). Employee retirement: A review and ijerp h1805 2746 recommendations for future investigation. Journal of Manage- Spreitzer, G. (1995). Psychological empowerment in the workplace: ment, 36(1), 172–206. https://doi. or g/10. 1177/ 01492 06309 347957 Dimensions, measurement, and validation. Academy of Manage- Wang, M., Zhan, Y., Liu, S., & Shultz, K. S. (2008). Antecedents of ment Journal, 38(5), 1442–1465. bridge employment: A longitudinal investigation. Journal of Spreitzer, G. (2008). Taking stock: A review of more than twenty years Applied Psychology, 93(4), 818–830. https:// doi. or g/ 10. 1037/ of research on empowerment at work. In J. Barling & C. L. Cooper 0021- 9010. 93.4. 818 (Eds.), Handbook of organizational behavior (pp. 54–72). Sage. Watermann, H., Fasbender, U., & Klehe, U.-C. (2023). Withdrawing Thomas, K. W., & Velthouse, B. A. (1990). Cognitive elements of from job search: The effect of age discrimination on occupational empowerment: An “interpretive” model of intrinsic task motiva- future time perspective, career exploration, and retirement inten- tion. Academy of Management Review, 15(4), 666–681. tions. Acta Psychologica, 234, 103875. https:// doi. org/ 10. 1016/j. Topa, G., Depolo, M., & Alcover, C.-M. (2018). Early retirement: A actpsy. 2023. 103875 meta-analysis of its antecedent and subsequent correlates. Fron- Wöhrmann, A. M. (2021). Later life work: Individual, job, and organi- tiers in Psychology, 8, 2157. https:// doi. org/ 10. 3389/ fpsyg. 2017. zational factors for retirement timing and work beyond normal 02157 retirement age. Leuphana University of Lüneburg. Topa, G., Moriano, J. A., Depolo, M., Alcover, C.-M., & Morales, Wöhrmann, A. M., Deller, J., & Wang, M. (2013). Outcome expec- J. F. (2009). Antecedents and consequences of retirement plan- tations and work design characteristics in post-retirement work ning and decision-making: A meta-analysis and model. Journal planning. Journal of Vocational Behavior, 83(3), 219–228. https:// of Vocational Behavior, 75(1), 38–55. https:// doi. org/ 10. 1016/j. doi. org/ 10. 1016/j. jvb. 2013. 05. 003 jvb. 2009. 03. 002 Wöhrmann, A. M., Fasbender, U., & Deller, J. (2017). Does more Traeger, C., & Alfes, K. (2019). High-performance human resource respect from leaders postpone the desire to retire? Understand- practices and volunteer engagement: The role of empowerment ing the mechanisms of retirement decision-making. Frontiers in and organizational identification. VOLUNTAS: International Jour - Psychology, 8, 1400. https:// doi. org/ 10. 3389/ fpsyg. 2017. 01400 nal of Voluntary and Nonprofit Organizations, 30(5), 1022–1035. Zaniboni, S., Topa, G., & Balducci, C. (2021). Core self-evaluations https:// doi. org/ 10. 1007/ s11266- 019- 00135-2 affecting retirement-related outcomes. International Journal of Trampusch, C. (2005). Institutional resettlement: The case of early Environmental Research and Public Health, 18(1), 174. retirement in Germany. In W. Streeck & K. Thelen (Eds.), Beyond Zhan, Y., Wang, M., & Yao, X. (2013). Domain specific ee ff cts of com - continuity: Institutional change in advanced political economies mitment on bridge employment decisions: The moderating role of (pp. 203–228). Oxford University Press. economic stress. European Journal of Work and Organizational Truesdale, B. C. (2020). Better jobs, longer working lives: Proposals to Psychology, 22(3), 362–375. https:// doi. org/ 10. 1080/ 13594 32X. improve the low-wage labor market for older workers (Economic 2012. 762763 Studies at Brookings). The Brookings Institution. https:// www. brook ings. edu/ wp- conte nt/ uploa ds/ 2020/ 11/ ES- 11. 16. 20- Trues Publisher's Note Springer Nature remains neutral with regard to dale. pdf jurisdictional claims in published maps and institutional affiliations. van Rijn, R. M., Robroek, S. J. W., Brouwer, S., & Burdorf, A. (2014). Influence of poor health on exit from paid employment: A sys- tematic review. Occupational and Environmental Medicine, 71(4), 295–301. https:// doi. org/ 10. 1136/ oemed- 2013- 101591 1 3

Journal

Journal of Adult DevelopmentSpringer Journals

Published: May 25, 2023

Keywords: Aging workforce; Psychological empowerment; Retirement age; Activeness after retirement; Bridge employment

There are no references for this article.