From business owner to retiree: understanding life satisfaction's impact on entrepreneurial exit

Abstract

Purpose

Entrepreneurial exit research has overlooked the unique context of exits at retirement age when an exit marks the end of an entrepreneurial career (i.e. retirement). To better understand retirement exit decisions and transitions, this study introduces the concept of work ability (i.e. an individual’s ability to meet work demands) into the entrepreneurial exit literature and, based on role theory, hypothesises its effect and interaction with general life satisfaction in explaining the entrepreneurial exits to retirement. The study clarifies the dynamics between the voluntary and non-voluntary aspects behind exit to retirement.

Design/methodology/approach

The authors use mixed-effects ordinal logistic regression with four-wave panel data on 198 Finnish small business entrepreneurs who intend to retire to test hypotheses on the relationship between work ability, general life satisfaction and entrepreneurial exit to retirement.

Findings

The study provides partial support for the hypothesis that work ability negatively impacts entrepreneurial exit to retirement and strong support for the idea that this effect is affected by general life satisfaction. Entrepreneurs who experience higher life satisfaction are likely to retire on their own terms, whereas those less satisfied continue working until declining work ability forces them to retire.

Originality/value

The study contributes to the entrepreneurial exit literature by showing how the exit dynamics unfold in the unique context of entrepreneurial exit to retirement. The theoretical discussion opens up the potential psychological mechanisms behind such dynamics.

Citation

Rönkkö, M., von Bonsdorff, M.E. and Mansikkamäki, S. (2024), "From business owner to retiree: understanding life satisfaction's impact on entrepreneurial exit", International Journal of Entrepreneurial Behavior & Research, Vol. 30 No. 11, pp. 261-278. https://doi.org/10.1108/IJEBR-12-2023-1243

Various personal, firm-related, and environmental factors influence entrepreneurs’ decisions to exit or continue business (Caliendo et al., 2020; DeTienne et al., 2008; Lin et al., 2022). The exit research has addressed why founders leave their companies, but the exit choices of ageing entrepreneurs are still underexplored (Backman et al., 2019, p. 9; Morris et al., 2020). Likewise, research on entrepreneurial careers has covered transitions to and from entrepreneurship (Burton et al., 2016) but has overlooked the end of entrepreneurial careers, often marked by retirement. As more entrepreneurs approach retirement age, understanding the late-career dynamics and the role of ageing in entrepreneurial exits is crucial (von Bonsdorff et al., 2019; Kautonen et al., 2017; Maâlaoui and Razgallah, 2019).

The authors draw on retirement and entrepreneurship literature to understand the understudied (Morris et al., 2020) topic of entrepreneurial exit to retirement. Ageing brings physical, psychological, and social changes, which may lead to a decline in the ability to meet job demands (work ability; Ilmarinen and Ilmarinen, 2015) in individuals 50 years and older (Ilmarinen, 2009). This can lead to workforce withdrawal among individuals at or approaching retirement age (Harada, 2007; Winter et al., 2004). However, physiological changes are not the only factors driving retirement. Using the role theory (Ashforth, 2001), a well-established social-psychological framework in the ageing in work-life research (Wang and Shi, 2014; Wang and Shultz, 2010), the authors propose that overall life satisfaction, a vital psychological resource, plays a central part in entrepreneurial exits to retirement.

The authors use four-wave survey data of Finnish entrepreneurs who intend to retire within the next two years, between 2017 and 2021, to test a set of hypotheses. The results show that declining work ability is associated with exits only for entrepreneurs who are dissatisfied with their lives. This supports the theorising that satisfied entrepreneurs view retirement as the closure of a successful career. Thus, the retirement and exit processes are mutually supportive, leading to a timely retirement. In contrast, dissatisfied entrepreneurs persist with their ventures until forced out by declining health. The central contribution is shedding light on the lifespan-specific characteristics of entrepreneurial exit. The study adds to the understanding of how delaying the exit to retirement due to an emotional connection to the company plays out. Furthermore, by incorporating well-being into role identity and transition, it complements the discussion on work and non-work identities in retirement transition. The paper continues with the theoretical background and hypotheses, followed by methods, results, and discussion.

1. Theoretical foundations

1.1 Retirement as entrepreneurial exit and the drivers of exit

Entrepreneurial exit to retirement combines two distinct exits: the entrepreneur’s exit from the firm (DeTienne, 2010) and the individual’s exit from work life to retirement. Entrepreneurial exits are either voluntary or forced (Coad, 2014; Justo et al., 2015). Alternative opportunities, economic motivation, and future prospects of the business, among others, may drive voluntary exits. An exit can be forced by financial failure or other firm stakeholders (DeTienne, 2010). While financial reasons often drive exits, entrepreneurs also exit financially viable firms and persist with struggling ones (DeTienne et al., 2008; Wennberg et al., 2010). Indeed, various personal or person-related, firm-related, and environment-related reasons (e.g. Caliendo et al., 2020; DeTienne et al., 2008; Lin et al., 2022) can affect the decision to exit or continue.

Entrepreneurial exits are often delayed. Entrepreneurs may continue with failing firms due to overconfidence, believing they can still reverse their fortunes (Lin et al., 2022). However, this explanation may be oversimplified (Chen et al., 2018). Jenkins and Byrne (2021) reviewed psychological theories of exit and failure and identified two reasons entrepreneurs might stay with an inevitably failing firm. First, escalation of commitment (e.g. Yamakawa and Cardon, 2017) refers to persisting with a decision or action despite worsening prospects. Escalation of commitment can arise from sunk-cost fallacy, over-optimism, preference for risk-taking to avoid losses, and many other sources (Sleesman et al., 2018). Second, delaying exiting a failing business can make the entrepreneur more emotionally prepared for the failure (anticipatory grief; Shepherd et al., 2009). Even harvest exits, where entrepreneurs cash in on their success, may get delayed or declined because of the entrepreneur’s emotional attachment to the firm (Jenkins and Byrne, 2021; Rouse, 2016).

The occupational health literature recognises health as crucial for an individual’s capacity to work. Specifically, work ability (i.e. an individual’s functional capacity to respond to work-related demands) has been identified as a critical predictor of various work outcomes, including retirement (Cadiz et al., 2019) and career sustainability (Stuer et al., 2019). Whereas prior entrepreneurship literature has recognised that health problems may force exits (Morris et al., 2020, p. 4), the effects of health and work ability on exits before one is forced have received little attention. Instead, studies have focused on topics such as how entrepreneurship affects health (see Stephan et al., 2020, for a recent summary) or how health affects different entrepreneurial outcomes (e.g. Hatak and Zhou, 2021). Only a few studies use health to explain exits: Descriptive studies by Harada (2007) and Winter et al. (2004) have shown that declining health drives entrepreneurial exits more often when approaching retirement age than within younger groups. Hessels et al. (2018) and Shahid and Kundi (2022) link the entrepreneur’s mental health to an increased likelihood of exit. However, the link between health and entrepreneurial exit remains understudied (Hessels et al., 2018; Shepherd and Patzelt, 2015).

1.2 A role theory perspective on retirement transition

Retirement is currently understood as a gradual process of four phases: planning, decision-making, transitioning, and adjustment to retirement (Shultz and Wang, 2011). Within the retirement literature, role theory is one of the main theories used to understand the retirement transition (see, e.g. Wang and Wanberg, 2017). Role theorists view retirement as a transition process that involves weakening or losing work roles and strengthening family and community roles (Wang et al., 2011; Wang and Shultz, 2010). This transition can be daunting and complex (Beehr, 2014), and its consequences can be either beneficial or detrimental, depending on how aligned the role transition is with an individual’s values and objectives (Wang and Shi, 2014).

Roles are patterns of behaviour and expectations that accompany a specific position or situation. Role theory assumes that individuals possess multiple roles, such as gender, profession, and family roles, that shape their behaviour, self-perception, and how others perceive their actions (Anglin et al., 2022; Ashforth, 2001, pp. 3–4). Role identity theory builds on role theory and argues that individuals build their self-concept based on the roles they occupy (Ashforth, 2001, chap. 2; Sluss et al., 2011). Roles differ from identities, a difference often lost in entrepreneurship research (Wagenschwanz, 2021). “Role is external; it is linked to social positions within the social structure. Identity is internal, consisting of internalized meanings and expectations associated with a role.” (Stryker and Burke, 2000, p. 289).

Role salience and role centrality refer to the importance of a role for an individual (Anglin et al., 2022, pp. 1474–1475). For entrepreneurs, the entrepreneur role often plays a central part in their identities (e.g. Shepherd and Patzelt, 2018, chap. 5). For example, Murnieks et al. (2014) demonstrated that entrepreneur role centrality strongly affected entrepreneurial passion, affecting both entrepreneurial self-efficacy and behaviour. Role identities can also affect who becomes an entrepreneur. Focusing on individuals who are currently not entrepreneurs, Seibert et al. (2021) demonstrated that entrepreneurial identity aspirations affected entrepreneurial activities.

Strong entrepreneurial roles may lead to adverse outcomes during the role identity change during retirement, such as crisis and continuity outcomes (Ashforth, 2001). Retirement involves a role transition away from the work role to occupy other roles (e.g. the family role) to a greater extent (Wang et al., 2011, p. 2). The role theory suggests that with high investments in a dominant current role (e.g. the work role), one’s perception of self-worth may be strongly associated with that role (Ashforth, 2001). Thus, for individuals highly invested in their jobs, a drift away from the pre-retirement role can lead to an absence of role identity (Quick and Moen, 1998, pp. 44–45), and retirement may be experienced as disruptive and stressful (Burke, 1991). This may be particularly salient for entrepreneurs who identify strongly with the entrepreneur role (Kleine and Schmitt, 2021; Murnieks et al., 2014; Radu-Lefebvre et al., 2021). Indeed, entrepreneurs often perceive retiring as undesirable, unwilling to relinquish their identities as entrepreneurs (Cesaroni et al., 2021; Morris et al., 2020). Similar behaviour has been observed with wage-earners who can extend their work role by part-time working, consulting, or other bridge work (von Bonsdorff et al., 2017; Bordia et al., 2020).

Retirement transition can be particularly problematic for individuals with weak non-role identities, causing a crisis and disruption (Bordia et al., 2020). In contrast, individuals with developed non-work role identities could experience smooth retirement role transitions. Moreover, strong work role identities can act as resources for building non-work role identities, which Bordia et al. (2020) demonstrated with an example of a project manager who developed a non-work role identity around managing church projects. In a family firm context, such retainment of an entrepreneurial role could arise from maintaining an advisory role after retirement (Cesaroni et al., 2021). Liu and Allan (2023) supported this idea by pointing out that individuals can reflect and reminisce about their work roles after retirement. As such, the work role identity can persist, albeit in a different form, and ease the role transition through a less disruptive role change. These arguments align with the role enhancement hypothesis, which suggests that individuals with multiple roles experience greater psychological well-being (e.g. Adelmann, 1994).

In summary, role theory examines how individuals behave based on the roles they have. How work and non-work role identities are constructed affects retirement adjustment in both employees (Bordia et al., 2020) and entrepreneurs (Morris et al., 2020). Because entrepreneurs often build strong work identities (Murnieks et al., 2014), role theory seems particularly useful for understanding retirement transition in the entrepreneurial context. However, while we know that strong work identities and weak non-work identities can complicate retirement adjustment, how they affect entrepreneurial exit to retirement is still unclear.

2. Hypothesis development

The theoretical argument builds on the idea of retirement as a process: when an entrepreneur’s work ability starts to decline, their propensity to exit voluntarily increases. When work ability decreases sufficiently, an exit is forced. The authors also argue that when entrepreneurs are satisfied with their life, the exit process plays out smoothly before work ability starts to decline substantially. In contrast, dissatisfied entrepreneurs resist retirement for psychological reasons, delaying exit that may ultimately become forced by declining work ability.

2.1 Work ability and entrepreneurial exit to retirement

Health is an important driver of the retirement decision (see, e.g. Wang and Shi, 2014) and an important indicator of career sustainability (De Vos et al., 2020). Job demands, such as physical workload or time pressure that exhaust individuals’ mental and physical resources, may cause health problems (Demerouti et al., 2001). The work ability concept implements this idea by focusing on the balance between an individual’s physical and mental resources and job demands (Cadiz et al., 2019; Ilmarinen, 2009; Ilmarinen and Ilmarinen, 2015). Work ability starts to decline in old age (Cadiz et al., 2019; Ilmarinen, 2009), and this influences many work-related outcomes, such as retirement behaviour (von Bonsdorff et al., 2010; Brady et al., 2020; McGonagle et al., 2015).

Employees’ work ability is negatively related to withdrawal from the workforce (e.g. Salonen, 2003). Compared to employees, entrepreneurs often face greater job demands and workloads (Stephan and Roesler, 2010) because of longer work days and greater responsibility (Patzelt and Shepherd, 2011, p. 229). Considering the high job demands and the heightened risk of health problems that may challenge the balance between job demands and available resources when approaching retirement age, work ability is expected to be associated with entrepreneurs’ retirement:

H1.

Work ability is negatively related to entrepreneurial exit to retirement among retirement-age entrepreneurs.

Entrepreneurs tend to retire older and be more attached to their firms than employees (Morris et al., 2020). This suggests that entrepreneurs may persist with their firms even when their work ability declines. Indeed, often, “the exit process will become inevitable only when they are faced with, e.g. severe illnesses” (von Bonsdorff et al., 2019, p. 71). However, entrepreneurs likely vary in their level of persistence. Next, the authors use role theory to argue that entrepreneurs most satisfied with their lives are least likely to wait until poor health forces them out.

2.2 Work ability, general life satisfaction, and exit to retirement

General life satisfaction refers to an individual’s overall subjective evaluation of their life. It is a broad assessment of one’s happiness, contentment, and fulfilment in various domains of life, such as relationships, health, work, personal achievements, and leisure activities. The experiences in the entrepreneur role and non-work role affect the overall level of an individual’s life satisfaction (Erdogan et al., 2012; Kleine and Schmitt, 2021). Generally, satisfying roles improve overall well-being (Kulik et al., 2015). Satisfaction with the entrepreneur and non-work roles are not separate as the two roles interact: A role conflict occurs when the roles one occupies have conflicting requirements. However, the roles may also be mutually enriching, such as using the work role as a retirement resource through reminiscence (Liu and Allan, 2023).

When dissatisfaction stems from the non-work role, increasing the salience of that role through retirement may seem like an undesirable option (role exit vs role entry; Jolles et al., 2023). While adopting more satisfying non-work role identities might help (Bordia et al., 2020), the outcomes of these efforts may be uncertain. Drawing on prospect theory (Kahneman and Tversky, 1979), Wennberg et al. (2010) have shown that entrepreneurs tend to delay exits when framed as a loss. Sustainable careers research suggests that the same logic likely applies to exits to retirement by showing that executives with a “retirement script” are more ready to retire (Hallpike et al., 2024). Thus, if entrepreneurs perceive retirement into non-satisfactory non-work role(s) as a failure (i.e. loss), they might choose to stay out of retirement even if this risks a forced retirement later on.

Just like dissatisfaction with the non-work role, dissatisfaction with the entrepreneurial role can also contribute to a delayed retirement. Exits to retirement can be successful not only financially (Morris et al., 2020) but also in non-financial terms, such as personal reputation and firm mission persistence after exit (Strese et al., 2018). Indeed, entrepreneurs can retire as respected community members (Bordia et al., 2020), having built companies that outlive their founders. An entrepreneur may see exiting an underperforming firm (financially or otherwise) as a failure, thus framing it as a loss (Hsu et al., 2017, p. 20). Both prior research on delayed exits (e.g. Jenkins and Byrne, 2021; Lin et al., 2022) and prospect theory indicate that entrepreneurs may prefer to delay retirement despite the potentially greater risk of a forced retirement further in the future.

These considerations suggest that dissatisfied entrepreneurs are less inclined to retire. This seems counterintuitive, as one might expect those satisfied with their lives to resist change (Erdogan et al., 2012). However, for ageing entrepreneurs, two factors suggest otherwise. First, their diminishing perceived time in the workplace and life more generally (Fung and Isaacowitz, 2016; Zacher and Frese, 2009) makes maintaining the status quo unsustainable. Second, those with higher life satisfaction are better prepared for the retirement transition: A satisfying work-role identity aids the transition (Bordia et al., 2020), and a satisfying entrepreneurial role reduces the likelihood of delays due to escalation of commitment or anticipatory grief (Jenkins and Byrne, 2021) and may also be a resource for the transition, e.g. through reminiscence (Liu and Allan, 2023). Therefore, the authors hypothesise:

H2.

Satisfaction with life is positively related to entrepreneurial exit to retirement among retirement-age entrepreneurs.

Life satisfaction not only directly affects the decision to retire for entrepreneurs but also moderates this process. The firm often represents their life’s work. Building on the role theory, the authors argue that when an entrepreneur is satisfied with their life’s work, retirement presents a closure of a successful career. In contrast, when entrepreneurs are unsatisfied with their life’s work, they do not view exit to retirement as a closure but as a failure. In this case, entrepreneurs face an escalation of commitment and thus persist with the venture longer than they should and may ultimately be forced to exit to retirement by their declining work ability.

H3.

Satisfaction with life moderates the negative relationship between work ability and entrepreneurial exit to retirement among retirement-age entrepreneurs, such that the negative relationship is weaker when satisfaction with life is higher.

3. Methods and data

The hypotheses were tested using a four-wave panel collected during a broader study of entrepreneurs’ retirement. The population consists of Finnish small business entrepreneurs who intend to retire within two years from an entrepreneurial career and employ at least one person in addition to themselves. The participants were recruited with the help of the Federation of Finnish Entrepreneurs (FFE). FFE first advertised the study in their weekly newspaper and other communications, providing the first 79 participants. The remaining participants were recruited with the help of FFE’s local agencies to identify 123 more participants. Four participants were dropped because their companies employed more than 250 people and thus fell outside the scope of the study. The final sample size for the initial wave was 198.

Baseline data were collected between 05/2017 and 06/2018 (T1) using telephone interviews that lasted about 1–1.5 h. The participants were asked about their work, health, company, retirement plans, family, and exit plans from the business. Two follow-ups were done by postal mail, with the first follow-up between 05/2019–01/2020 (T2, N = 153) and the second between 06–11/2020 (T3, N = 151). The third follow-up took place during the COVID pandemic between 9–10/2021 (T4, N = 171) using telephone interviews. T-tests of the demographic variables indicated no differences between respondents and nonrespondents, suggesting that attrition does not affect the results.

3.1 Measures

Exit status was measured using a multiple-choice question where the authors asked if there had been any changes to the firm’s ownership structure. In line with the view of retirement as a gradual process (Shultz and Wang, 2011), this variable was operationalised as a three-level ordered category consisting of 1) no plans to exit, 2) started exit, and 3) exited. The full scale, shown in the Appendix, includes one item showing no current exit plans, three items indicating the initiation of various exit types, and four items marking the completion of different exits.

Perceived work ability. Entrepreneurs’ self-reported work ability was assessed by the first item of the Work Ability Index (WAI; Tuomi et al., 1998). The respondents were asked to evaluate their current work ability against their lifetime best on a scale of 0–10. The complete WAI instrument consists of 7 parts with different scale formats. However, because of its complexity, the single-item version is often favoured. The single item is highly correlated with the full scale and has comparable predictive validity (Ahlstrom et al., 2010). The full WAI was collected in the baseline wave and subsets in the follow-up waves. The single-item version was used for its simplicity and availability in all waves.

General life satisfaction was measured with the Satisfaction with Life Scale (SWLS), which assesses a person’s overall judgement of their life (Diener et al., 1985). The scale has five items, including “I am satisfied with my life” and “In most ways my life is close to my ideal”, each rated on a seven-point Likert scale (1 = strongly disagree, 7 = strongly agree). (ω_Η = 0.82, 0.79, 0.85, 0.84, ω_T = 0.91, 0.91, 0.89, 0.88 for T1-T4).

Several control variables were used. The first firm-level control variable is firm performance because it is likely to affect general life satisfaction, and successful companies are easier to exit from than less successful ones. Performance was measured using the ten-item perceived relative performance measure developed by Delaney and Huselid (1996) because a relative measure allowed us to compare companies with different goals and business characteristics. The authors also controlled for firm size, measured as a category (1, 2–4, 5–10, 11–20, 21–50, and 51–250 personnel) because larger companies might be easier to exit from than smaller ones. The final firm-level control was family business, measured with a question asking whether the entrepreneur identified the company as a family business.

Three entrepreneur-related characteristics were controlled. Age was controlled because it can be expected to influence both exit intentions and work ability. Gender is an important role, and prior research has shown that men and women develop different work and non-work role identities. College education was controlled because college-educated and non-college-educated entrepreneurs likely face different work demands and resources, influencing work ability. The demographic controls (age, gender, and college education of the entrepreneur; personnel and family business status of the firm) were collected during the first wave. All other variables were collected longitudinally.

3.2 Analytical strategy

The data were analysed with mixed-effects ordinal logistic regression because the authors expect work ability to have a nonlinear effect such that there are little differences between high levels of work ability (work ability is sufficient and little changes make no difference) and between low levels of work ability (exit becomes a dominant outcome well before reaching complete inability to work). Further, while exit status is a continuum from not considering an exit to starting an exit to completing an exit, considering these as equally spaced is unrealistic. Starting an exit requires a mental commitment and is thus closer to being exited than not considering an exit. All independent variables were lagged by one year to consider the possible reverse causality. Cluster robust standard errors were used to account for the possibility that the error term might be autocorrelated over time and year fixed-effects to control for the effects of the COVID pandemic and other global trends.

Testing the third hypothesis requires a model where perceived work ability and general life satisfaction interact. However, the authors suspected that the effects of negative ends of satisfaction would be stronger than positive ends. Because theory provides little explicit guidance on choosing a functional form, it was discovered empirically (Rönkkö et al., 2022). To do so, the authors rounded the life satisfaction variable to whole numbers and estimated a categorical interaction model providing a separate effect of work ability for each life satisfaction level. This analysis revealed little differences between the high values of life satisfaction, but the effect started to change more radically in the lower values of life satisfaction. Thus, the life satisfaction variable was log-transformed to model large effects for smaller values.

4. Results

Table 1 shows the descriptive statistics and correlations. The mean age was about 64.5 years, which means that, on average, the informants are past the standard retirement age in Finland, which is 63 for a person born in 1954 or earlier. About 80% of the entrepreneurs are men, which is expected in these age cohorts. Exit correlates negatively with work ability, as expected. The correlation is small, but this masks an important finding explained later. Firm performance, work ability, and satisfaction with life are all positively correlated, which is expected. The relatively high correlation between life satisfaction and firm performance aligns with the theory that firms are essential parts of entrepreneurs’ lives. Men are slightly less satisfied with life and have somewhat lower firm performance, which the authors found interesting. Despite planning to retire in two years from the start of the study, just 33% of the respondents had exited by the final wave, while 40% were in the process of doing so.

Table 2 shows the ordinal logistic regression results. Model 1 is the control variable model. Model 2 adds the variables of interest. The effect of work ability on exit is negative, and the effect of life satisfaction is positive, as expected based on the theory. However, neither of these effects is statistically significant, and thus, Model 2 does not provide sufficient evidence to support the hypotheses. However, this model omits the interaction between the two and consequently provides an incomplete picture of the relationship between the variables. Model 3 adds an interaction between work ability and life satisfaction. The interaction effect is positive (i.e. the negative effect diminishes) and statistically significant, supporting Hypothesis 3.

Because the ordinal logistic regression model is non-linear, moderation was interpreted by plotting (Rönkkö et al., 2022). Figure 1 shows adjusted prediction plots for three levels of general life satisfaction: low (life satisfaction = 2), average (life satisfaction = 5), and high (life satisfaction = 7). These values were chosen to represent the scale minimum (just one person had life satisfaction less than 2, so 2 instead of the scale minimum 1 was used), mean, and maximum. The graphs show the expected probability of each outcome as a function of work satisfaction ranging from “work ability at its best” to “cannot work at all” (i.e. from 10 to 0).

Figure 1 shows a remarkable result. For the highest levels of life satisfaction, work ability seems mostly unrelated to exit status. In contrast, when life satisfaction is low, work ability strongly negatively affects exit outcomes. In other words, more satisfied entrepreneurs tend to exit at high rates (predicted probabilities of started exit at about 50% and exited at about 35%) even when they can still work. In contrast, less satisfied entrepreneurs tend to keep working if they can (predicted probability of not planning an exit at about 65% when work ability is at its best) and only exit when their work ability declines. This result provides explicit support for Hypothesis 3. Thus, while, on average, satisfaction with life or work ability does not seem to affect retirement, the effects of work ability are substantial for the less satisfied entrepreneurs, as the theory predicts. As such, Model 3 partially supports Hypothesis 1 and Hypothesis 2.

The rug plots on the horizontal axis show that most cases had good or only slightly declined work ability, and minimal work ability was rare. Moreover, the cases of poor work ability were relatively more common in the weak life satisfaction subsample compared to the other subsamples. This raises a concern that these few outlier cases might drive the results. To address this concern, the models were re-estimated using a subsample that included only cases where work ability was 5 or higher. The results were very similar to the reported results, thus providing evidence of the robustness of the findings.

5. Discussion

Exit to retirement differs from other entrepreneurial exits in two key ways. First, the exit of a retirement-age entrepreneur is also a career exit involving a significant life change. This may create a greater sense of loss than exits at earlier career stages. Second, although retirement is commonly viewed as a voluntary choice, delaying it will inevitably make it a forced one because of the inability to work or death.

The study found that entrepreneurs who are satisfied with their lives do not exit the business because of declining work ability. However, the same is not the case for less satisfied entrepreneurs, whose exit strongly depends on their work ability. In other words, dissatisfied entrepreneurs keep working until forced to exit because of declining health, whereas satisfied entrepreneurs exit on their own terms.

5.1 Implications for theory

The study contributes to two more general discussions on entrepreneurial role transitions, role identities, and delayed exits. The exit literature is largely based on the assumption that entrepreneurial exit results from rational decision-making and that financial gain opportunities often drive harvest exits (Jenkins and Byrne, 2021). From this perspective, the authors propose that whether entrepreneurs see retirement as a success or a failure depends on pre-retirement life satisfaction. Exiting from an unsatisfactory career state is perceived as a failing end for one’s career. This highlights the importance of recognising the significance of the perceived success of an entrepreneurial exit (Strese et al., 2018) not just as an outcome but also as an antecedent of the exit.

Entrepreneurs often delay exiting unsuccessful businesses and harvesting successful ones due to an emotional connection to the company (Jenkins and Byrne, 2021; Rouse, 2016). Delays in distress exits may be due to factors such as escalation of commitment (Yamakawa and Cardon, 2017) or anticipatory grief to decrease the psychological costs (Shepherd et al., 2009). However, while entrepreneurial role identity has been discussed in the exit literature (Rouse, 2016), its effect in delaying exits has yet to be addressed. This may be because prior research has assumed that entrepreneurs can maintain their entrepreneurial identity by pursuing new venture ideas (Ahmed et al., 2022; Rouse, 2016). Exit to retirement presents an interesting phenomenon because it denies this assumption. As such, the role transition that occurs in entrepreneurial exit to retirement may present an opposite to the difficult role transition process of becoming a founder (Hoang and Gimeno, 2010) because of its permanent nature. Thus, like becoming a founder, ceasing to be an entrepreneur might present an identity transition many would prefer not to take.

A retiring entrepreneur may lose their entrepreneurial identity. What this means for an individual is mainly unknown (Radu-Lefebvre et al., 2021). Recent research has shown that identity complexity (identities with several roles that may or may not overlap) is associated with increased satisfaction with life through identity integration, where individuals combine their various roles into a cohesive self-concept (Ng et al., 2023): the present study and Bordia et al. (2020) provide indirect support for this idea. This means that entrepreneurs with complex identities are better positioned to handle a role loss of their entrepreneurial identity.

Beyond the entrepreneurship literature, the study also links to the broader discussion on role identities and well-being (Anglin et al., 2022) and sustainable careers (De Vos et al., 2020). Prior work has investigated the differences in life satisfaction and overall positive well-being between entrepreneurs and employees (Stephan et al., 2022), the effects of entrepreneurial passion on well-being (Chen et al., 2022), and the link between career satisfaction and retirement (Hallpike et al., 2024). Whereas research on sustainable careers shows that employees who have both strong work and non-work roles are less likely to experience role conflict (Hirschi et al., 2020), little is known about the relationship between role transitions and well-being, both in the context of the entrepreneurial role identity (Radu-Lefebvre et al., 2021) and more generally (see, Anglin et al., 2022, p. 1489). Building on role theory, the authors link life satisfaction, an indicator of well-being, to the retirement exit decision. The authors suggest that individuals’ well-being, influenced by their perception of transitioning away from an entrepreneurial role to retirement, determines their readiness for this change. Incorporating well-being into the role identity and role transition discussion complements and extends the recent insights of Bordia et al. (2020) on the strength of work and non-work identities as defining factors for the adaptation of retirement transition and potential delays.

5.2 Implications for practice

How the transition into retirement and withdrawal from work plays out has a substantial effect on satisfaction with retirement (Shultz and Wang, 2011; Wang et al., 2011). As demonstrated by previous studies, the transition may be challenging for entrepreneurs (Morris et al., 2020). Especially in cases where resources have declined before transitioning into retirement (Zhan et al., 2022), it would be essential to find ways to support ageing individuals – here entrepreneurs – to facilitate their timely transition to retirement. The study suggests that strong role identities and the satisfying life they provide make the retirement transition smoother. For individual entrepreneurs, this means pre-emptively strengthening their non-work role identities before retirement. This can be done, for example, by engaging in non-work activities or fostering social networks outside the entrepreneurial context. Another alternative is developing the entrepreneurial role to provide satisfaction and allow for using it as a resource through reminiscence.

A smooth and timely retirement exit benefits not only the entrepreneurs but also their firms and the broader society (Morris et al., 2020). For example, CEO’s age has been shown to possibly harm firm growth and performance in closely held firms (Belenzon et al., 2019). Delaying an entrepreneurial exit until work ability starts to suffer can be risky for a firm. Also, after a successful exit, the company’s new owner can make investments and bring their experience and knowledge to the company. This will likely boost the company’s performance and may also help create new jobs. Timely retirement exits are thus important for maximising societal benefits from entrepreneurship.

5.3 Limitations

Like all studies, the study is not without its limitations. The first and most important limitation of the present study is that the authors have conducted the study in a single country context, and the cultural norms related to retirement and the features of the pension and healthcare systems may affect retirement behaviour. The Finnish pension system is considered reliable and provides an earnings-related pension for entrepreneurs in contrast to, e.g. the US system, which is more savings-based (Jensen et al., 2020). Finnish entrepreneurs are required by law to take out pension insurance and receive pension benefits based on the contributions they have made during their time as entrepreneurs and based on their possible careers as employees. At the time of data collection, employees and entrepreneurs were eligible for retirement at age 63, after which it was possible to continue working (but the earnings-based pension will no longer accumulate after age 68) (Finnish Centre for Pensions, 2023). The stable pension system and the somewhat strong normative pressure to retire after reaching the official retirement age may push entrepreneurs to retire around the official age. However, during the telephone interviews (T1), we found that the retirement-aged entrepreneurs were highly committed to their businesses and felt strongly about continuing to work despite indicating their plans to retire in a few years. We are unaware of any international comparisons of entrepreneurs' work identities and non-work identities, but prior research suggests this is not unique to the Finnish context (Morris et al., 2020).

As the second limitation, although the survey data are longitudinal, the data are provided by a single respondent. This raises the question of common method variance (Podsakoff et al., 2012). However, given that all variables are measured with different scale formats and because the exit variable focused on concrete plans or actions instead of perceptions and was measured at a different time than the explanatory variables, common method variance does not seem a significant concern.

A third limitation is that, whereas the study builds on role theory, role identities or role transitions were not measured. Instead, predictions from role theory were used to theorise potential causal mechanisms between life satisfaction and entrepreneurial exit to retirement. Thus, it is left to future research to test the theorised causal mechanism more comprehensively.

6. Conclusions

The authors introduce the concept of work ability—individuals' capacity to meet work demands—into the study of entrepreneurial exit, examining its interaction with life satisfaction in influencing entrepreneurial retirement decisions. Drawing from retirement literature and role theory, the study found partial evidence that work ability negatively affects entrepreneurial retirement and strong evidence that this is influenced by overall life satisfaction. The research indicates that entrepreneurs with high life satisfaction tend to retire voluntarily, while those less satisfied work until declining work ability necessitates retirement. This highlights that retirement decisions involve both voluntary and involuntary aspects and shows that life satisfaction plays a dual part in the entrepreneurial retirement process.

Figures

Figure 1

Adjusted prediction plots of exit status at different values of work ability over three levels of life satisfaction

Table 1

Descriptive statistics and correlations

	Mean	SD	Min	Max	1	2	3	4	5	6	7	8	9
1. Exit status (t + 1)	1.10	0.67	0	2	1
2. Life satisfaction	5.35	0.97	1	7	0.09	1
3. Work ability	7.53	1.32	1	10	−0.09	0.27**	1
4. Firm performance	3.57	0.54	1.5	5	−0.03	0.39**	0.17**	1
5. Firm size	2.68	0.92	1	5	0.22**	0.04	−0.07	0.13**	1
6. Family business	0.78	0.41	0	1	−0.17**	0.05	−0.03	0.00	−0.06	1
7. Age	64.52	3.47	54	77	−0.08	−0.06	0.06	−0.10	−0.11*	0.18**	1
8. Gender	1.82	0.39	1	2	0.04	−0.15**	−0.12*	−0.14**	0.16**	−0.06	−0.01	1
9. College education	0.27	0.45	0	1	0.02	0.02	0.12*	0.09	0.04	−0.15**	0.08	−0.08	1

Note(s): Gender: 1 = female, 2 = male. Demographic variables were measured at wave 1, Life satisfaction, Work ability, and Firm performance were measured at waves 1–3, and Exit status (t+1) was measured at waves 2–4. Statistics by wave are reported online at https://osf.io/6cmzs

**p < 0.01, *p < 0.05

Source(s): Authors’ own work

Table 2

Mixed effects ordinal logistic regression models of exit status

	1	2	3
Firm performance	−0.302	−0.396	−0.392
	(0.353)	(0.355)	(0.356)
Firm size	0.723**	0.701**	0.715**
	(0.249)	(0.241)	(0.239)
Family business	−1.315*	−1.357**	−1.421**
	(0.526)	(0.518)	(0.521)
Age	−0.033	−0.028	−0.027
	(0.055)	(0.055)	(0.054)
Gender	−0.224	−0.192	−0.120
	(0.629)	(0.629)	(0.628)
College education	−0.208	−0.168	−0.259
	(0.529)	(0.523)	(0.527)
Time
Wave 2	0.549*	0.624*	0.646*
	(0.255)	(0.253)	(0.258)
Wave 3	0.006	0.034	0.122
	(0.374)	(0.379)	(0.379)
Life satisfaction (log)		1.168	−2.650*
		(1.149)	(1.231)
Work ability		−0.136	−1.180***
		(0.105)	(0.336)
Life satisfaction (log) × Work ability			0.641**
			(0.206)
Threshold: Started exit	−5.466	−4.531	−10.459
	(4.087)	(4.425)	(4.511)
Threshold: Exited	−0.919	−0.002	−5.877
	(3.987)	(4.332)	(4.407)
Variance of random intercept	5.126	4.891	4.908
	(1.401)	(1.373)	(1.368)
Log pseudolikelihood	−495.75	−489.58	−488.52

Note(s): ***p < 0.001, **p < 0.01, *p < 0.05. Cluster robust standard errors in parentheses. All independent variables lagged by one year

Regression models without controls are available online at https://osf.io/6cmzs

Source(s): Authors’ own work

Appendix The exit status measures

Have you recently implemented changes to the ownership structure of your firm? Choose the option that best matches the current situation:

• (1)

  No, and there are no plans for them in the coming years

• (2)

  Not yet, but I am planning to terminate the company’s operations and/or liquidate the company’s assets

• (3)

  Not yet, but succession arrangements are currently underway

• (4)

  Not yet, but owner change arrangements are currently underway

• (5)

  Yes, a generational change has been implemented in the company

• (6)

  Yes, I have sold all or most of the company’s operations

• (7)

  Yes, I have sold all or most of the company’s share capital

• (8)

  I have terminated the company’s operations and/or realized the company’s assets