Conservation of resources theory

The conservation of resources theory (COR, Hobfoll 1989) is an integrated resource theory that views resources as part of a greater dynamic process to build health and well-being. Resources are entities of central value in their own right (e.g. health) or act as means to achieve such a value (e.g. money) (Hobfoll 2002). Threat of resource loss, net loss of resources, or the inability to rebuild resources after a loss are expected to cause stress and since individuals seek for protection and enhancement of the self, they seek to obtain, retain, grow and protect key resources (Hobfoll 1989). Such key resources are central values, and apart from health, these can be objects (e.g., a car), conditions (e.g., employment), personal assets (e.g., skills, abilities) or energies (e.g., time, information, money) that have no intrinsic own value but are used to gain other resources. In addition, an important implication of COR-theory is that resources link to other resources and interact in their development over time to build stable resource reservoirs, net gains or loss spirals (Hobfoll 1989).

Mapping this onto the work situation of the self-employed, health is a central resource for the self-employed that can decline, be maintained or strengthened over time (see Fig. 1). Using the person-centred approach, we first investigate whether there are subgroups among self-employed with different health developments, for example, stable health, net gains or losses as suggested in COR-theory. Next, we use individual characteristics that are known to influence overall health differences to validate these health trajectories. To further investigate the existence of stable resource reservoirs, net gains or loss spirals, we study how trajectories in general health relate to developments in other key resources. More specifically, we study co-developments in mental well-being (depressive symptoms, sleep disturbances), which—according to COR-theory—should be closely interlinked with physical health (Hobfoll 1989, 2002). For the work domain, we study how different health trajectories are associated with energy resource developments (in time or money), here referred to as business resources (order flow, security of income). We also study links to conditional resources related to employment conditions (change in self-employment status). Finally, we investigate personal resources (physical and mental work ability) as these are important resources that can change along with developments in individuals and their work situations. In Fig. 1, these abilities therefore are portrayed as both an individual and work-domain resource.

Fig. 1

Conceptual model on individual and work-related developments of resources

Central resources: health trajectories over time

To date, there are few studies on health differences and changes in the self-employed. Among the existing studies, many focused on mental health in cross-sectional settings (Stephan 2018), compared health at two-time points only or studied health changes relative to a specific event, such self-employment start or exist (for a recent study, see Nikolova et al. 2021). Evidence on health developments following self-employed workers over extended time periods (e.g. several years) is scarcer and mainly comes from register data. Comparing self-employed and wage workers, mortality and inpatient care rates were found to be higher in the self-employed (Gauffin and Dunlavy 2021). Conversely, other studies reported lower mortality rates in self-employed workers than in wage workers (Goncalves and Martins 2018; Toivanen et al. 2016). It was also found that self-employed workers with limited liability companies had a somewhat lower mortality rate than sole proprietors (Gauffin and Dunlavy 2021), but a less unequivocal picture emerged when sector/industry (Toivanen et al. 2016) and income gradient (Gauffin and Dunlavy 2021) were added to the analyses. In sum, this indicates that heterogeneity in self-employment is worth studying in more detail. However, if the aim is to learn more about health in self-employed workers, these studies have several shortcomings. First, the available register data helps to study proxy measures or consequences of ill-health, which only partly matches the definition of health used by the WHO (WHO, 1948) (being more than the absence of disease). Second, especially measures of mortality can only portray an outcome, that overlooks the importance of health as a resource (Hobfoll 1989), and the agentic role of workers to protect it. Adopting this view, a recent 16-year follow-up study from Australia (Hessels et al. 2020) shows that self-reported general and physical but not mental health were related to higher earnings in self-employed compared to wage workers. However, although self-employed with and without employees were differentiated, Hessels et al. (2020) did not investigate whether health developed differently in subgroups of the self-employed (e.g. health remains stable, fluctuates, increases or decreases over time). Thus, it remains unknown how prevalent potentially different health trajectories in self-employed workers are, and whether investigating such trajectories would help to better understand developments in other important resource domains. This leads to our first research question:

RQ1. Which different health trajectories can be identified in the self-employed, and how prevalent are these?

Validating identified health trajectories with demographic characteristics

In Sweden, as in many other countries, men are overrepresented in self-employment (OECD 2017). Regarding gender differences in health, it has been found that hospitalization rates were lower for self-employed women than men (Goncalves and Martin 2018), but mental health declined more in female compared to male self-employed workers during the pandemic (Caliendo et al. 2023). However, such health differences between self-employed men and women might be explained by the fact that they work in different sector (Toivanen et al. 2015): Whereas men are most likely to be self-employed in manufacturing, logistics and transportation, and thus, blue-collar work, self-employed women are most likely working in the service sector (OECD 2017).

Regarding education, the average Swedish self-employed has a secondary education (Toivanen et al. 2016) which was the dominant educational level in Sweden in the past (SCB 2022). However, in recent years, those with post-secondary (academic) education and business ideas, and those with low educational levels and few entry paths into the labor market have been most likely to enter self-employment (Simoes et al. 2016). Variations in educational level also mean variations in socio-economic status and thus differences in health (Rostila and Toivanen 2018).

Although the average Swedish self-employed is over 50 years of age, recent investigations show that younger individual in Sweden seem to be as likely to start a new business as older ones (GEM 2022), thus age differences may matter when health developments are studied.

In sum, health is likely to be related to demographic characteristics, but it is unclear to what extent different health trajectories over time may be more prevalent for certain groups of self-employed workers. Due to insufficient previous research, we rather formulate a research question than a hypothesis.

RQ2. Are certain subgroups of self-employed (older workers, females, blue-collar workers, workers with lower education) more likely to be found in a specific health trajectory?

Relations of identified health trajectories to other resource developmentsAssociations of health trajectories with resource development in well-being

Given that the work environment of self-employed workers is characterized by many challenges related to uncertainty (Mcmullen and Shepherd 2006), high work load (Bernhard-Oettel et al. 2019), long work hours (Binder and Blankenberg 2021) and a blurred work-non-work boundary (Stephan 2018) mental health and good sleep may be resources that are constantly threatened and may have to be rebuilt after periods of net loss.

Many studies on mental well-being in self-employment found that autonomy was positively related to well-being whereas work demands and uncertainty related to the business were associated with elevated levels of burnout, depressive symptoms, fatigue or stress (Stephan 2018). Differences among the self-employed were also noted; those who saw opportunities with their businesses and those who had stabilized their businesses over time reported less distress (Stephan 2018). Longitudinal studies on job and life satisfaction found that longer time and more persistent careers in self-employment were associated with higher levels in satisfaction (Koch et al. 2021; Litsardopoulos et al. 2021).

Few if any studies have used a person-centred approach to explore the existence of different patterns of well-being in the self-employed. One notable exception comes from a cross-sectional study that identified several distinct profiles in well-being among the self-employed (Bujacz et al. 2020), and workers with happy, satisfied, passionate and flourishing patterns were less depressed and anxious and had a more favorable work and income situation. This aligns with COR theory (Hobfoll 1989) suggesting that resourceful individuals are less vulnerable and more likely to build up further resources.

Little research on self-employed workers has been done on sleep, even though sleep may be an important resource to feel well and energetic for a new work day. Sleep quality and thus well-being may vary between individuals and, depending on stressful challenges, even within individuals over time. A recent study showed that on average, the self-employed may sleep more hours than wage workers; but in periods of psychological distress, their sleep quality deteriorates (Wolfe and Patel 2020). How such sleep disturbances develop over time and link to differences in general health in the self-employed is not known.

RQ3. How do health trajectories in self-employed workers differ regarding developments of depressive symptoms and sleep disturbances?

Associations of health trajectories with business resource development

Business resources, particularly in terms of money but also time are valuable energy resources to protect or increase the individual resource reservoir (Hobfoll 1989). The availability of these resources may vary in the group of self-employed workers as a function of order flow and possibilities to secure a steady income stream. Running a shop, working in seasonal industries or offering consultancy implies varying contract and order distribution over the year. This means that some self-employed may encounter unpredictable changes in demand (Goncalves and Martin 2018). Some may adopt strategies such as taking on more clients than necessary to secure economic success (Grant and Ferris 2012). Thus, workload and income flows are unevenly distributed, which may lead to stress (Goncalves and Martin 2018), since important resources are at stake. In line with this, subjective well-being has been related to business success and economic well-being in the self-employed (Patel & Wolfe 2019) and as mentioned above, better health was found to be associated with higher earnings (Hessels et al. 2020). However, how potential subgroups of self-employed workers with different health trajectories struggle with or succeed in accumulating business resources, e.g., client/order flow and income security, has not been studied yet. Thus, our research question reads:

RQ4. How do health trajectories in self-employed workers differ regarding developments of business resources, i.e., order flow and income security?

Associations of health trajectories with resource development in employment status

Individuals may fully work in self-employment or as combiners, that is, they combine self-employment with wage work (also called hybrid entrepreneurs, see Folta et al. 2010). Since combiners often enter self-employment stepwise or for a test period, they are believed to differ from those solely working as self-employed (Folta et al. 2010). In Sweden, this group has a fair share of entrepreneurial activity. Keeping a paid job in addition to self-employment may add to income security but could also be strenuous and negative for health. Indeed, results show that combiners reported more work-life interference (Hagqvist et al. 2018) and small but statistically significant higher levels of sleep disturbances, exhaustion and depressive symptoms (Bergman et al. 2021) than those solely being self-employed. Remaining self-employed, working as a combiner or taking on wage work instead of self-employment may help individuals maintain, rebuild or stop loss in health resources. A recent study on involuntary and voluntary exit from self-employment found that well-being was more negatively affected than health among those who left self-employment and that effects were stronger for involuntary exit (Nikolova et al. 2021). Thus, developments in health and conditional resources that are tied to the employment status of self-employment, wage work or a combination of both may be related. Still, subgroups with different health trajectories and switches that also include being a combiner have not been studied. Thus, we ask:

RQ5. How do health trajectories in self-employed workers differ regarding changes in employment status over time?

Associations of health trajectories with personal resource development

Next, overall health is often discussed as an important resource in self-employment and a prerequisite to be able to run and develop the business (Shepherd and Patzelt 2015; Torrès and Thurik 2019). Since the self-employed are a heterogeneous group and work in various occupations that may entail either more physical (e.g. for carpenters) or mental work (e.g. for HR consultants) the development of both their mental and physical capabilities may be of importance. Such functional capacities have been termed as physical or mental work ability and its most important determinant is health (Ilmarinen et al. 2008). Assumedly, different health trajectories in the self-employed are related to developments in their work ability, but with few studies at hand, it is difficult to hypothesize potential fluctuations or differences in different subgroups. Thus, we formulate the following research question:

RQ6. How do health trajectories in self-employed workers differ regarding developments in physical and mental work ability?