Skip to main content
Erschienen in:

Open Access 01.08.2024 | Research Paper

The Paradox of Job Retention Schemes: A Latent Growth Curve Modeling Approach to Immediate and Prolonged Effects of Short-Time Work on Job Insecurity and Employee Well-Being

verfasst von: Katharina Klug, Claudia Bernhard-Oettel, Magnus Sverke

Erschienen in: Journal of Happiness Studies | Ausgabe 6/2024

Aktivieren Sie unsere intelligente Suche, um passende Fachinhalte oder Patente zu finden.

search-config
loading …

Abstract

Many countries rely on short-time work to prevent mass layoffs in economic crises. Despite serving to protect jobs, short-time work may trigger job insecurity perceptions, which may impair employee well-being. Moreover, past experiences of unemployment may increase susceptibility to job insecurity in response to short-time work. Drawing on Conservation of Resources (COR) Theory, Appraisal Theory and temporal stressor-strain models, this study investigates effects of short-time work on the development of job satisfaction, life satisfaction and affective well-being via perceived job insecurity across 6 years, considering previous unemployment as a moderator. We used propensity score matching to account for selectivity into short-time work and calculated latent growth curve models with N = 1211 employees in Germany (619 affected by short-time work, 592 controls). Short-time work predicted higher levels and an immediate increase in job insecurity, followed by a decrease over time. Both levels and changes in job insecurity were associated with levels and changes in well-being. Indirect effects of short-time work on well-being via job insecurity persisted 2 years after short-time work. We found no difference between previously unemployed respondents and others in their reactions to short-time work. The findings support COR theory and a prolonged stress-reaction model, showing lingering effects on well-being via job insecurity even after short-time work ends. The study supports short-time work as an antecedent of job insecurity and reveals temporal dynamics between job insecurity, its antecedents and outcomes over time. When implementing short-time work, employers should aim to mitigate concerns about job security to protect employee well-being.
Hinweise

Supplementary Information

The online version contains supplementary material available at https://​doi.​org/​10.​1007/​s10902-024-00787-y.

Publisher's Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

1 Introduction

Many OECD countries rely on “short-time work” or comparable job retention schemes to prevent mass layoffs in economic crises. These schemes enable firms to temporarily reduce their employees’ working hours, while governments subsidize compensatory income support (European Commission, 2020, p. 2; OECD, 2020a, 2020b). During the Covid-19 crisis, the prevalence of short-time work reached unprecedented heights, affecting over 38 million European employees (Drahokoupil & Müller, 2021). In the literature, short-time work has mainly been discussed as a policy instrument, and studies have focused on its effectiveness in cushioning economic shocks, investigating the benefits and risks for businesses and labor markets (Brenke et al., 2013; Cahuc & Carcillo, 2011; Drahokoupil & Müller, 2021).
In contrast, the psychological consequences for individual employees are largely unknown so far (for exceptions, see e.g. Rauvola et al., 2022; Tušl et al., 2021; Wels et al., 2022). Although short-time work is explicitly designed to prevent job loss, it may induce job insecurity, that is, the perception of an involuntary threat to the continuity of the job (Sverke et al., 2002). From a stress-theoretical perspective, short-time work may signal economic uncertainty which employees appraise as threatening (Lazarus & Folkman, 1984), since reduced hours and income entail a loss of resources (Hobfoll, 1989). The resulting job insecurity perceptions can negatively affect employee well-being (Green et al., 2016; Silla et al., 2009; Taht et al., 2020).
Because short-time work reflects an organization-induced change that has a clear onset and is by definition temporary, it lends itself to analyzing the causal dynamics between job insecurity, its organizational antecedents and employee reactions over time, for example regarding the development and persistence of psychological effects: It is possible that employees immediately recover once work goes back to normal. It is also plausible that short-time work triggers lingering perceptions of job insecurity, since the future of the organization may be uncertain—indeed, some firms making use of short-time work eventually have to downsize in the long run (OECD, 2020b). Stress is a process that unfolds within persons over time (Garst et al., 2000), and there is still a lot to uncover about the temporal dynamics between job insecurity, its antecedents and consequences (Klug et al., 2020, 2024a). For example, people’s past experiences inform their present stress appraisals: Previous unemployment has been linked to heightened job insecurity, but it is yet unclear why. One explanation could be perceptual bias such that employees who have lost a job in the past may react to the same cue (e. g., short-time work) with stronger job insecurity perceptions than others (Ellonen & Nätti, 2015; Shoss, 2017).
This study aims to investigate the short- and long-term effects of employment disruptions as in short-time work on employee well-being via perceived job insecurity, taking into account the moderating role of past unemployment. Drawing on classic stress theories (Hobfoll, 1989; Lazarus & Folkman, 1984) and models of temporal stressor–strain dynamics (Frese & Zapf, 1988), we conceptualize short-time work as an antecedent of job insecurity which, in turn, is assumed to affect employee well-being. Using a sample of employees who experienced short-time work in the 2008 recession and its aftermath, we trace the development of job insecurity and well-being across 6 years from the onset of short-time work in comparison to a matched control group. More specifically, we investigate effects of between-person differences (i.e., affected by short-time work or not) on within-person change in job insecurity and well-being over time.
We contribute to the literature in several ways: First, we address open questions regarding the temporal relationships between job insecurity, its antecedents and consequences, such as when effects first occur and when they recede (Klug et al., 2020; Watson & Osberg, 2017). Second, taking previous unemployment into account as a moderator, we address the questions of how and why past labor market experiences shape future perceptions of job insecurity and, in turn, vulnerability to the negative consequences for well-being (Shoss, 2017). Third, from a practical perspective, evidence from before the pandemic can be informative regarding potential long-term effects for employee well-being. Although short-time work is certainly preferable to layoffs (Stuart et al., 2021), studying its effects over time renders important insights for organizations and policy makers in future crises.

1.1 Short-time Work as an Antecedent of Job Insecurity and Well-being over Time

Short-time work, also known as “furlough”,1 refers to job retention schemes “that allow firms experiencing economic difficulties to temporarily reduce the hours worked while providing their employees with income support from the State for the hours not worked.” (European Commission, 2020, p. 2). While specific implementations vary between countries, such as eligibility criteria, the level of income allowance or maximum duration, the defining feature is that employees are not laid off—the employment contract and associated benefits are preserved (Müller & Schulten, 2020; OECD, 2020b). As a policy instrument, short-time work has been evaluated regarding its effectiveness in preventing unemployment (Balleer et al., 2016). To date, empirical analyses of psychological consequences for employees are limited (see Möhring et al., 2021; Rauvola et al., 2022; Tušl et al., 2021; and Wels et al., 2022).
We argue that short-time work can have the paradoxical effect of increasing employees’ job insecurity. Job insecurity is a subjective perceptual phenomenon reflecting an involuntary threat to the continuity of one’s job (Lee et al., 2018; Sverke et al., 2002). Job insecurity is a multifaceted construct: Quantitative job insecurity (i.e., concerns about losing the job as such) can be distinguished from qualitative job insecurity (i.e., concerns about losing valued job features), as well as cognitive job insecurity (i.e., evaluative thoughts) from affective job insecurity in terms of negative feelings related to the future of the job (Chirumbolo et al., 2020; Hellgren et al., 1999; Jiang & Lavaysse, 2018; Sverke & Hellgren, 2002).
In this study, we focus on quantitative job insecurity, because short-time work is used when employees may otherwise be laid off (i.e., the job as such is at risk). We operationalize job insecurity in terms of employees’ worries about the future of their job, without distinguishing between cognitive and affective facets (by definition, worrying is a cognitive process closely intertwined with negative affect; Borkovec et al., 1998; Xie et al., 2019). This approach is supported by empirical evidence: While cognitive and affective job insecurity may be distinguished from one another, both tend to correlate negatively with well-being, and positively to one another (Jiang & Lavaysse, 2018). Although (quantitative) job insecurity is sometimes conflated with unemployment in the literature, work psychological definitions typically understand job insecurity as concerns about continuity in one’s current job which, as such, can only be experienced by employees (De Witte, 1999; Lee et al., 2018). This is also the perspective taken in this article.
From a stress-theoretical perspective, employees are expected to react to short-time work with job insecurity perceptions for two reasons. First, transactional stress theory explains stress mainly as a function of subjective appraisals (Lazarus & Folkman, 1984) suggesting that employees would interpret short-time work as a cue of organizational change, economic difficulties, and a reduced demand for their labor (Shoss, 2017). Consequently, they would appraise the situation as threatening and perceive their job at risk. Second, conservation of resources (COR; Hobfoll, 1989) theory explains stress in terms of resource depletion. From this perspective, the reduction in hours and income that accompany short-time work to varying degrees (Müller & Schulten, 2020) reflect resource loss (Baranik et al., 2019; Halbesleben et al., 2013). As resource loss sensitizes people to further loss (Hobfoll et al., 2018), employees in short-time work may react with increased worries whether they can keep their job in the long run.
In line with this reasoning, organizational change has been found to predict job insecurity (Keim et al., 2014). Organizational change also entails role ambiguity and rumors, which provide fertile grounds for job insecurity (Blackmore & Kuntz, 2011; Keim et al., 2014; Smet et al., 2016). Moreover, involuntary and uncontrollable changes in one’s job, like short-time work, relate to perceived job insecurity (Hofer et al., 2021), as well as a difficult financial situation and staff changes (Ellonen & Nätti, 2015; Lee & Sanders, 2013). Based on theoretical and empirical considerations, we therefore expect that employees in short-time work generally experience more job insecurity than those whose work is not disrupted:
Hypothesis 1:
Short-time work is positively associated with job insecurity.

1.2 Short- and Long-term Developments of Job Insecurity in Response to Short-time Work

The stress process unfolds within persons over time: Stressors can accumulate or decrease, appraisals can change, and resource losses and gains can reinforce each other in a spiraling fashion (Ford et al., 2014; Garst et al., 2000). Because short-time work has a clear onset and is by definition temporary (in Europe typically 6–24 months), it poses a unique opportunity to investigate these temporal dynamics between job insecurity, its organizational antecedents and well-being among employees (see Klug et al., 2020, 2024a). The question arises how employee reactions develop over time and what happens after short-time work ends. Different models have been proposed in the literature to describe how stress reactions develop over time. For example, the “stress reaction model” holds that stress responses occur immediately and begin to disappear once the source of stress is removed, whereas the “accumulation model” states that stress reactions accumulate over time and remain even after the source is removed (Frese & Zapf, 1988; Garst et al., 2000).
The stress reaction model would predict job insecurity to increase during short-time work and decrease thereafter. Research shows that job insecurity perceptions are susceptible to short-term changes (Garrido Vásquez et al., 2018; Smet et al., 2016), such that insecurity likely increases immediately with the onset of short-time work. It is, however, difficult to predict when job insecurity decreases again. The primacy of loss principle in COR theory states that resource loss impacts people longer and more severely than equivalent resource gain (Hobfoll et al., 2018), such that employees may feel stressed and less secure even after they return to their previous working conditions (Baranik et al., 2019). Once the uncertainty associated with the onset of short-time work has begun, employees might worry whether their employer will recover economically, whether short-time work will be repeated, or whether layoffs may happen eventually. This is in line with research on downsizing survivors, who show elevated job insecurity levels even after keeping their jobs (Dlouhy & Casper, 2021; Maertz et al., 2010). However, it also seems unlikely that these effects will be unlimited. Therefore, we expect a non-linear development of job insecurity over time because of the temporary nature of short-time work, while refraining from predictions about the exact time of recovery.
Hypothesis 2:
Short-time work is associated with an immediate increase in job insecurity, followed by a decrease over time.
If short-time work elicits job insecurity, it will likely have negative consequences for employee well-being. Several meta-analyses support job insecurity as a stressor with negative effects on employee well-being (Cheng & Chan, 2008; Jiang & Lavaysse, 2018; Llosa et al., 2018; Sverke et al., 2002), which can be comparable to effects of unemployment (Kim & Knesebeck, 2015). Studies show that the effect sizes of job insecurity on employee reactions may depend on the welfare regime, employment protection legislation or union density in a given country (Fullerton et al., 2020; Probst & Jiang, 2017; Sverke et al., 2019). At the same time, the comprehensive evidence from a range of countries suggests that, overall, a general negative effect of job insecurity on employees’ well-being is rather universal across modern economies (Cheng & Chan, 2008; Jiang & Lavaysse, 2018; Llosa et al., 2018; Sverke et al., 2002). In this study, we focus on job satisfaction, life satisfaction and affective well-being to capture both cognitive and affective aspects, as well as job-specific and general well-being (Schimmack et al., 2008; Warr, 1994).
A growing body of longitudinal research suggests that within-person changes in job insecurity also relate to changes in well-being over time (e.g., Bernhard‐Oettel et al., 2020; Kinnunen et al., 2014; Klug et al., 2019; Smet et al., 2016; Vander Elst et al., 2014). These studies show immediate reactions to job insecurity, rendering synchronous changes likely (Bernhard‐Oettel et al., 2020; Klug, 2020). Long-term changes and patterns of accumulation in job insecurity have also been found to relate to long-term changes in employee well-being (Kinnunen et al., 2014; Klug et al., 2019). In line with previous research on the development of stressors and strain over time (Garst et al., 2000), we therefore expect:
Hypothesis 3:
The level of job insecurity is negatively associated with the respective levels of job satisfaction (H3a), life satisfaction (H3b) and affective well-being (H3c).
Hypothesis 4:
Changes in job insecurity are associated with synchronous changes in job satisfaction (H4a), life satisfaction (H4b) and affective well-being (H4c), such that when job insecurity increases, well-being decreases and vice versa.

1.3 Linking Short-time Work to Well-being via Job Insecurity

From the previous hypotheses follows that short-time work may affect employee well-being indirectly via job insecurity. So far, a few studies have investigated the potential impact of short-time work on well-being during the Covid-19 pandemic. In a study from Germany, Möhring et al. (2021) investigated the differential impact of short-time work on employee well-being by gender and parenthood during lockdown and found a negative effect on job satisfaction, especially for mothers (Möhring et al., 2021). Another study with employees in Germany found that short-term work alleviated the effects of the first lockdown on work fatigue, but this effect was not found in the second lockdown (Rauvola et al., 2022). A cross-sectional study in Germany and Switzerland showed that employees in short-time work simultaneously perceived negative effects on their working life and positive effects on their private life; however, short-time work with a complete reduction to zero hours was associated with worse mental well-being and self-rated health (Tušl et al., 2021; see also Halbesleben et al., 2013). Wels et al. (2022) investigated employees in Great Britain during lockdown and found that employees in short-time work reported worse mental health than those with uninterrupted employment. Building on these studies, our theoretical considerations suggest that perceived job insecurity could be part of the explanation why short-time work can have negative effects on well-being:
Hypothesis 5:
Job insecurity mediates the relationship between short-time work and well-being over time.

1.3.1 Exploring Temporal Dynamics Between Short-time Work and Employee Reactions

It is possible that the effects of short-time work on job insecurity and, in turn, well-being linger beyond the duration of short-time work itself (Halbesleben et al., 2013). The accumulation model of stress suggests that some stress reactions can persist even after the stressor has disappeared. While the stress reaction model predicts a decrease of strain once the stressor subsides, the decrease may take some time (Frese & Zapf, 1988). In line with the primacy of resource loss principle in COR theory (Hobfoll et al., 2018), elevated insecurity can persist even after downsizing (Dlouhy & Casper, 2021; Maertz et al., 2010), and mid-life job insecurity has been shown to affect well-being even after retirement (Barrech et al., 2011). In contrast, Watson and Osberg (2017) found that employees’ mental health recovered after one-time experiences of insecurity while only repeated experiences had lasting effects. Still, other studies suggest that employee well-being appears not to improve immediately after job insecurity decreases (Ferrie et al., 2002; Klug et al., 2019).
Processes of timing, accumulation and recovery between job insecurity, its antecedents and outcomes are underexplored in the literature (Klug et al., 2020). Generally, we would expect employees to recover from short-time work as the source of insecurity subsides and their access to resources is restored. However, given the state of the literature, it is impossible to make an informed prediction as to when exactly effects would decrease. We therefore take an exploratory approach to this issue:
Research question: When do effects of short-time work on job insecurity and well-being recede?

1.4 The Role of Previous Unemployment

Because job insecurity involves subjective appraisals (Sverke et al., 2002), employees likely differ in the extent to which they perceive short-time work as threatening their job. One important difference relates to people’s past labor market experiences concerning unemployment. Unemployment experiences may have scarring effects such that job loss affects people throughout their careers, beyond the duration of unemployment itself (Clark et al., 2001; Knabe & Rätzel, 2011). Accordingly, previous unemployment has been linked to later perceptions of job insecurity (Ellonen & Nätti, 2015; Erlinghagen, 2008; Kinnunen & Nätti, 1994). It is yet unclear whether this effect depends on selection effects (i.e., being more likely to be re-employed in less secure jobs), or whether the experience of job loss changes one’s susceptibility to job insecurity (Shoss, 2017), which would be in line with the transactional stress model.
In the transactional stress model, the appraisal of a stressor is informed by previous experiences and coping efforts (Lazarus & Folkman, 1984). Hence, we expect previously unemployed individuals to react to the same objective cue of threat (short-time work) with stronger subjective insecurity, because they have already experienced once that a loss could not be prevented. COR theory also suggests that previous loss heightens people’s fear of future loss (Hobfoll et al., 2018). Employees who have already experienced job loss may thus feel less capable of preventing a second loss and perceive short-time work as more threatening than others. This could show both in higher levels of job insecurity, but also a steeper pattern of change over time. For example, compared to their peers, young adults who have been previously unemployed have shown trajectories with initially higher job insecurity upon labor market entry (Klug et al., 2019). We expect that, due to past loss experiences, the development of job insecurity in response to short-time work is particularly pronounced among previously unemployed persons:
Hypothesis 6:
Previous unemployment experiences moderate the relationships between short-time work and job insecurity over time, such that employees who have experienced unemployment respond with more job insecurity to short-time work in terms of levels (H6a) and change over time (H6b).

2 Method

2.1 Context of the Study

The study was conducted in Germany and, taking the 2008 recession as a starting point, included from the years 2009 to 2018. The German welfare state closely resembles the Bismarckian prototype with a high degree of contribution-based social protection and stratification along occupational groups (Arts & Gelissen, 2002; Esping-Andersen, 1990). Germany also has comparably strong employment protection regulations and generous unemployment assistance (OECD, 2020a, 2022), although the labor market reforms in the early 2000s reflected far-reaching cuts in the social security system and a shift toward activating labor market policies (Zohlnhöfer, 2016).
Dating back to the early 1900s, short-time work has a long tradition in Germany (Brenke et al., 2013). Companies experiencing substantial economic difficulties can reduce employees’ hours and apply for short-time work at the Federal Employment Agency, which pays the short-time work allowance (currently 60% of the previous net pay, 67% for employees with children; Schulten & Müller, 2020). Short-time work serves to alleviate the impact of economic shocks by protecting employees’ jobs and associated benefits, at the same time relieving companies of personnel costs, but also of costs related to firing and rehiring employees once the business is resumed. The instrument thus supports firms’ internal flexibility in the face of cyclical fluctuations in demand (Brenke et al., 2013; Schulten & Müller, 2020).
During acute crises, such as the 2008 recession, and most recently the pandemic, regulations regarding short-time work have been adjusted: During the 2008 crisis, for example, the maximum duration was expanded from 12 to 18, then 24 months, and then reset to 12 months again in 2010 (Brenke et al., 2013). Especially when regulations are expanded, concerns regarding potential risks arise, such as abuse by employers, keeping employees in (repeated) short-time work for too long, or artificially sustaining businesses which are ultimately not viable (Boeri & Bruecker, 2011; Brenke et al., 2013; Cahuc & Carcillo, 2011; Drahokoupil & Müller, 2021; OECD, 2020b). Accordingly, the stabilizing effect of short-time work on labor markets has been evaluated positively with regard to short-term, cyclical rather than structural crises (Drahokoupil & Müller, 2021; OECD, 2020b).

2.2 Sample and Procedure

Our study was based on a secondary analysis of available survey data from the German Socio-Economic Panel study (SOEP), a longitudinal population study which has been conducted annually since 1984. The SOEP provides representative panel data on living and working conditions from over 30,000 individuals (Wagner et al., 2007). Perceived job insecurity and well-being are regularly surveyed in the SOEP. Short-time work has been occasionally included in recurring intervals (1991–1995, 2010–2014 and 2017–2018).
Focusing on the most recent economic crisis before the pandemic, we included employees affected by short-time work 2009–2013. The timeframe enabled us to follow employees affected by short-time work for 5 years until the latest available data wave. We selected employed respondents aged 18–64 and followed them for 6 years from 1 year before the onset of short-time work (t − 1) up to 5 years later (t5). Pooling the data across survey waves, we defined the year in which respondents reported being in short-time work as t1.
To account for selection effects into short-time work, we used propensity score matching to create a matched control group of respondents not affected by short-time work (Rubin, 1973). Propensity score matching has been suggested to strengthen causal inference in observational studies of developmental trajectories following life events (Infurna et al., 2016). With a 1:1 matching approach, we selected a control with a comparable socio-demographic profile for each respondent affected by short-time work. Because subgroup differences between respondents with versus without prior unemployment experience were of substantive interest, we followed recommendations for stratified matching and calculated separate propensity scores for the two subgroups (Green & Stuart, 2014), using the following steps: For each survey wave from 2009 to 2013, we calculated two cross-sectional logistic regressions of being in short-time work on age, gender, migration background, education, job tenure, hours per week, company size, part-time employment, temporary employment, occupational class and earnings to estimate the respective propensity scores for respondents with prior unemployment and for those without. In both groups, respondents in short-time work were then matched to control respondents not affected in the given year using a nearest-neighbor algorithm with replacement (Rosenbaum & Rubin, 1983). Matching with replacement (i.e., matching some controls to more than one respondent in the treatment group) may result in unbalanced group sizes, but increases matching quality when using nearest-neighbor algorithms (Caliendo & Kopeinig, 2008). As a result of the matching procedure, there were no significant differences between respondents with and without short-time work in any of the covariates used for the matching, both in the overall sample and within the respective subgroups of respondents with and without prior unemployment.
We did not impose strict inclusion criteria regarding missing data and included all participants with data on short-time work and at least one valid observation on one of the other variables. To ensure that a linear growth model is identified, three observations per person should be present for a “a sizeable portion of the cases” (Curran et al., 2010, p. 125). The proportion of respondents with three or more observations over time was high on all study variables (90% each on job insecurity, job satisfaction and life satisfaction; 85% on affective well-being).
Because only a small share of the SOEP sample had experienced short-time work, we determined the sample size by identifying those respondents who were affected at some point during the observation period, met the selection criteria and could be successfully matched to a control. Out of over 35,000 employees in the SOEP sample, 680 respondents who were in short-time work at some point between 2009 and 2013 were identified. A total of 619 short-time workers could be included in the matching process, and were matched to 592 control respondents, resulting in a final sample size of N = 1211 (61 respondents were dropped due to missing values on matching variables). The control group was slightly smaller due to matching with replacement (i.e., some respondents served as controls for more than one short-time worker).
About two thirds of the sample were men (70%), respondents were on average 43 years old at t1 (SD = 10), and 18% had a migration background (i.e., respondents themselves or either of their parents were born outside of Germany). Most respondents (72%) had vocational training, 19% had an academic degree and 9% had no professional degree beyond high school. At t1, respondents worked an average of 40 h a week (SD = 9), 14% had a part-time job, 9% had a temporary contract, 52% worked in organizations with 100 employees or more, and the average job tenure was 12 years (SD = 10). More than a third (39%) had been unemployed at some point prior to t1.
Table 1 shows the sociodemographic characteristics of the short-time workers and the control group at t1. As can be seen in the table, there are no differences between the short-time workers and the control group as a result of the matching. For comparative purposes we also provide the same sociodemographic characteristics for the entire group of all employees aged 18–64 before matching, although direct comparisons with this group are less relevant since short-term work may be more/less applicable based on certain sociodemographic characteristics.2
Table 1
Sociodemographic profiles of short-time workers, the control group, and all employees
 
Short-time workers t1
Control group t1
Short-time work vs. control group
All employees 2009–2013
M (SD) / Mdn / %
M (SD) / Mdn / %
M (SD) / Mdn / %
Age
42.94 (9.95)
42.65 (9.79)
t (1209) = 0.51, ns
43.61 (11.33)
Gender: woman
30%
30%
χ2 = 0.02, ns
51%
Migration background
18%
17%
χ2 = 0.13, ns
32%
Education
  
χ2 = 0.15, ns
 
High school or lower
9%
9%
17%
Vocational training
73%
72%
58%
Tertiary degree
18%
19%
25%
Job tenure
12.36 (9.96)
12.04 (10.04)
t (1209) = 0.56, ns
10.7 (9.61)
Hours per week
39.74 (10.19)
39.47 (8.56)
t (1209) = 0.62, ns
36.60 (12.56)
Company sizea
6
6
z = 0.610, ns
5
Part-time employment
14%
14%
χ2 = 0.04, ns
42%
Temporary contract
10%
9%
χ2 = 0.41, ns
27%
Occupational class
  
χ2 = 0.98, ns
 
Manual worker
37%
37%
30%
Non-manual worker
16%
14%
25%
Service class employee
47%
49%
45%
Earnings
34 703.51 (49 523.08)
34 753.10 (49 444.57)
t (1209) = − 0.02, ns
32 139.52
Nshort-time work = 619; Ncontrol group = 592; Nall employees = 35 254. Company size as number of employees in categories: 1 =  < 5, 2 = 5–10, 3 = 11– ≤ 20, 4 = 20– ≤ 100, 5 = 100– ≤ 200, 6 = 200– ≤ 2000, 6 =  ≥ 2000. ns not significant
Furthermore, as a result of the matching procedure, the sample was not completely representative of the active workforce in Germany, but rather of employees likely affected by short-time work (see Brenke et al., 2013; Tarullo & Desiere, 2023): Compared to the general workforce, men, employees with a vocational degree, those working in larger companies and manual workers were over-represented, and working hours and job tenure were slightly above average. Employees with no professional degree, those working part-time and on temporary contracts, as well as employees with a migration background were underrepresented. The average age was comparable to the general workforce (www.​destatis.​de). These sample characteristics are in line with reports that short-time workers are typically labor market insiders (i.e., prime age, qualified, in permanent full-time employment), and that short-time work in the 2010s was largely driven by male-dominated sectors such as manufacturing and construction (Brenke et al., 2013; Tarullo & Desiere, 2023).

2.3 Measures

Unless stated otherwise, all variables were defined as continuous, time-varying and measured at all time points from t − 1 to t5.
Short-time work at t1. Short-time work at t1 was assessed with a dichotomous variable (0 = no, 1 = yes) and defined as a time-invariant predictor. The SOEP survey asks respondents whether they have been affected by short-time work in the year prior to the survey. To assess synchronous effects on the other variables, we created a leading variable to record the response for a given year (e.g., for a respondent reporting short-time work for the previous year in 2011, the leading variable assigns a 1 in the year 2010).
Perceived job insecurity. Job insecurity was assessed with an item asking participants how much they worry about the security of their job, thus reflecting quantitative job insecurity and capturing both cognitive and affective elements (by definition, worries are cognitions, but with a negative valence, thus not affectively neutral, see Borkovec et al., 1998). The item was rated on a 3-point scale (1 = very worried, 2 = somewhat worried, 3 = not worried at all; reverse-coded in the analyses). Similar single-item measures of job insecurity have been successfully used in previous research (e. g., Ferrie et al., 2002), and shown meaningful relationships with employee well-being, albeit with somewhat lower associations compared to studies using multi-item measures (Sverke et al., 2002). Recent analyses further support the validity and reliability of single item measures of job insecurity (Matthews et al., 2022).
Job satisfaction. Job satisfaction was measured with an item asking respondents to rate how satisfied they are with their job on an 11-point-scale from 0 = absolutely dissatisfied to 10 = absolutely satisfied. Previous research supports the validity and reliability of single-item measures for job satisfaction (Fisher et al., 2016).
Life satisfaction. Life satisfaction was measured with a five-item scale asking respondents to rate their satisfaction with life in general as well as with different aspects of their life on an 11-point-scale from 0 = absolutely dissatisfied to 10 = absolutely satisfied (income, health, leisure time and housing; see Richter et al., 2017; Schimmack et al., 2010 for evidence of the scale’s validity). Reliabilities in the sample ranged from α = 0.71 at t1 to α = 0.76 at t4.
Affective well-being. Affective well-being was measured with a four-item scale asking respondents how often they have felt happy, angry, sad and worried, respectively, in the past four weeks on a scale from 1 = very rarely to 5 = very often (see Richter et al., 2017; Schimmack et al., 2008 for evidence of validity). Reliabilities in the sample ranged from α = 0.63 at t2 to α = 0.72 at t5.
Previous unemployment at t − 1. A dichotomous variable indicated whether respondents had ever been unemployed before t1 (coded 0 = no, 1 = yes), defined as a time-invariant variable.
Table 2 shows descriptive statistics and bivariate correlations between all study variables across the six time points.
Table 2
Descriptive statistics and bivariate correlations among the study variables across time points
 
M (SD)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
1 Short-time work
0.51 (–)
                         
2 Prev. unemployment
0.40 (–)
0.00
                        
3 Job insecurity t− 1
1.81 (0.71)
0.15
0.11
                       
4 Job insecurity t1
1.92 (0.76)
0.26
0.12
0.59
                      
5 Job insecurity t2
1.81 (0.73)
0.24
0.10
0.52
0.60
                     
6 Job insecurity t3
1.65 (0.68)
0.15
0.10
0.50
0.47
0.55
                    
7 Job insecurity t4
1.64 (0.68)
0.11
0.09
0.48
0.45
0.58
0.62
                   
8 Job insecurity t5
1.61 (0.68)
0.10
0.07
0.44
0.46
0.50
0.59
0.61
                  
9 Job satisfaction t− 1
6.97 (1.99)
− 0.02
− 0.04
− 0.24
− 0.22
− 0.21
− 0.21
− 0.15
− 0.13
                 
10 Job satisfaction t1
6.79 (2.11)
− 0.09
− 0.03
− 0.21
− 0.30
− 0.25
− 0.20
− 0.16
− 0.14
0.53
                
11 Job satisfaction t2
6.80 (2.09)
− 0.15
− 0.02
− 0.23
− 0.27
− 0.32
− 0.24
− 0.21
− 0.21
0.52
0.58
               
12 Job satisfaction t3
6.79 (2.08)
− 0.06
− 0.03
− 0.23
− 0.24
− 0.24
− 0.31
− 0.24
− 0.18
0.48
0.50
0.60
              
13 Job satisfaction t4
6.84 (2.02)
− 0.02
0.03
− 0.19
− 0.15
− 0.17
− 0.22
− 0.19
− 0.16
0.44
0.33
0.47
0.57
             
14 Job satisfaction t5
6.92 (1.96)
− 0.04
− 0.02
− 0.18
− 0.17
− 0.19
− 0.25
− 0.20
− 0.25
0.44
0.35
0.47
0.50
0.58
            
15 Life satisfaction t-1
7.00 (1.31)
− 0.05
− 0.16
− 0.27
− 0.25
− 0.23
− 0.29
− 0.21
− 0.16
0.52
0.33
0.39
0.36
0.37
0.45
           
16 Life satisfaction t1
6.99 (1.32)
− 0.05
− 0.15
− 0.24
− 0.23
− 0.23
− 0.22
− 0.20
− 0.18
0.42
0.47
0.41
0.34
0.34
0.40
0.72
          
17 Life satisfaction t2
6.98 (1.36)
− 0.11
− 0.18
− 0.29
− 0.26
− 0.29
− 0.26
− 0.26
− 0.22
0.43
0.41
0.53
0.43
0.40
0.43
0.71
0.71
         
18 Life satisfaction t3
6.91 (1.36)
− 0.06
− 0.14
− 0.24
− 0.24
− 0.27
− 0.27
− 0.25
− 0.22
0.40
0.37
0.46
0.52
0.46
0.46
0.64
0.63
0.74
        
19 Life satisfaction t4
7.01 (1.34)
− 0.04
− 0.12
− 0.28
− 0.19
− 0.23
− 0.25
− 0.24
− 0.18
0.40
0.31
0.37
0.37
0.55
0.43
0.60
0.63
0.69
0.71
       
20 Life satisfaction t5
7.02 (1.30)
− 0.04
− 0.13
− 0.25
− 0.19
− 0.24
− 0.26
− 0.20
− 0.19
0.43
0.32
0.40
0.39
0.40
0.54
0.65
0.64
0.69
0.69
0.72
      
21 Affect. well-being t-1
3.64 (0.64)
− 0.05
− 0.11
− 0.21
− 0.17
− 0.13
− 0.19
− 0.08
− 0.09
0.35
0.24
0.30
0.26
0.22
0.25
0.48
0.36
0.41
0.37
0.30
0.33
     
22 Affect. well-being t1
3.62 (0.64)
− 0.04
− 0.08
− 0.19
− 0.20
− 0.16
− 0.18
− 0.10
− 0.14
0.29
0.33
0.26
0.22
0.22
0.25
0.37
0.43
0.37
0.38
0.29
0.31
0.54
    
23 Affect. well-being t2
3.65 (0.66)
− 0.08
− 0.05
− 0.20
− 0.16
− 0.18
− 0.20
− 0.13
− 0.13
0.26
0.28
0.32
0.27
0.19
0.26
0.37
0.36
0.46
0.41
0.31
0.40
0.55
0.57
   
24 Affect. well-being t3
3.60 (0.66)
− 0.04
− 0.03
− 0.19
− 0.16
− 0.17
− 0.25
− 0.18
− 0.17
0.27
0.27
0.32
0.37
0.35
0.32
0.34
0.33
0.42
0.49
0.38
0.35
0.52
0.48
0.58
  
25 Affect. well-being t4
3.66 (0.64)
− 0.02
− 0.06
− 0.21
− 0.12
− 0.16
− 0.26
− 0.20
− 0.19
0.27
0.18
0.24
0.27
0.38
0.30
0.33
0.33
0.35
0.40
0.47
0.39
0.44
0.44
0.48
0.55
 
26 Affect. well-being t5
3.67 (0.66)
− 0.08
− 0.07
− 0.18
− 0.13
− 0.17
− 0.23
− 0.18
− 0.19
0.32
0.21
0.28
0.29
0.32
0.39
0.35
0.33
0.37
0.39
0.38
0.50
0.49
0.46
0.59
0.55
0.60
N = 549–1211 due to pairwise deletion of missing data. Scale range 1–3 for job insecurity, 0–10 for job and life satisfaction, 1–5 for affective well-being. Significant correlations (p < 0.05) in boldface. Prev. previous, Affect. affective

2.4 Analysis

We used Stata 13 for data preparation, descriptive statistics and propensity score matching, and Mplus 8 (Muthén & Muthén, 19982017) for the analyses. The analysis scripts are available at https://​osf.​io/​tebz2/​?​view_​only=​89f0451ac33b4af2​8cfa9fa6b611678b​. To capture between-person differences in within-person changes in job insecurity and well-being over time, we calculated latent growth curve models (LGCM; Bollen & Curran, 2006) with robust maximum likelihood estimation. Specifically, to test hypotheses 1–5, we calculated parallel process models (Cheong et al., 2003) for job insecurity and well-being with short-time work as a time-invariant predictor. We modeled, for both job insecurity and well-being, 1) a latent intercept factor representing the initial level, 2) a latent slope factor representing linear change and 3) a quadratic growth factor, because we expected non-linear change in the variables of interest over time (cf. Bollen & Curran, 2006). We specified the parallel process model such that the latent growth factors of both job insecurity and well-being were each regressed on short-time work, whereas the intercept and slope for well-being were each regressed on the intercept and slope for job insecurity (cf. Cheong et al., 2003). The quadratic term for well-being was additionally regressed on the quadratic term for job insecurity. To control for effects of baseline levels in job insecurity or well-being on within-person change in the respective other variable over time, we specified paths from the well-being intercept to the job insecurity slope and quadratic term, and vice versa.
To test for mediation, we first calculated indirect effects of short-time work on the well-being intercepts and slopes via the job insecurity intercept and slope, respectively. While an indirect effect via two slopes can provide general evidence of a linear mediation over time, it does not indicate whether the indirect effect is present at each time point (Bollen & Curran, 2006). Additionally, to the best of our knowledge, there seems to be no established standard to account for nonlinear mediation in parallel process models (see Liu et al., 2009; von Soest & Hagtvet, 2011). We therefore calculated additional linear growth models for the well-being variables with job insecurity as a time-varying covariate to probe for indirect effects at each time point, with Bonferroni-adjustment for multiple tests. That is, we set the threshold at p < 0.008 (p < 0.05 divided by 6 tests at 6 time points) and calculated 99%-bootstrapped confidence intervals instead of 95%. This approach allowed accounting for nonlinearity in the mediation over time while simultaneously exploring when effects first occur and when they recede to answer our research question. In all models, we accounted for missing data with full-information maximum likelihood (FIML) estimation (Enders, 2001).

3 Results

Before we tested our hypotheses, we conducted several preliminary analyses (documented in Table S1 and S2, respectively, in the supplementary materials): First, we calculated confirmatory factor analyses (CFA) to test the structure and invariance of our measures across groups and time points. Following established guidelines and common practice, we inspected (changes in) χ2 values, CFI, RMSEA and SRMR as indices of model fit (see Hu & Bentler, 1999). The results at t1 supported our hypothesized four-factor model (job insecurity, job satisfaction, life satisfaction and affective well-being) which showed acceptable fit (χ2 (40) = 223.14, p < 0.001, CFI = 0.91, RMSEA = 0.07, SRMR = 0.05; ∆ χ2 (3) = 86.78, p < 0.001, compared to a two-factor model; see Table S1).
Second, we tested the configural (equal factor patterns), metric (equal patterns and item loadings) and scalar invariance (equal patterns, item loadings and intercepts) of our measures from the four-factor model a) between employees affected by short-time work and the control group at t1, b) between employees with and without previous unemployment experience at t1, and c) the respective invariance levels of the multi-item measures (life satisfaction and affective well-being) across the six time points. The models supported metric invariance (partial invariance for affective well-being; see Table S1), suggesting that comparing associations between the variables across groups and time points was justified (Vandenberg & Lance, 2000). Additionally, scalar invariance was supported for the comparison of employees in short-time work and the control group, suggesting that comparing also the mean levels of variables between these two groups was justified (Vandenberg & Lance, 2000).
Third, we calculated univariate LGCMs to assess the degree and shape of change in the variables of interest over time. The results showed overall acceptable model fit and suggested that adding a nonlinear, quadratic growth term was justified as it improved the fit compared to a model with only an intercept and a linear slope for job insecurity, job satisfaction, life satisfaction and affective well-being (see Table S2).

3.1 Effects of Short-time Work on Perceived Job Insecurity and Well-being Over Time

Because the variables were measured on different scales, Fig. 1 shows the standardized path coefficients from the parallel process LGCMs for the three different outcomes: Short-time work as a dichotomous indicator was modeled as a manifest variable (as represented by the rectangle), whereas intercepts and slopes for job insecurity and well-being were modeled as latent variables from the six observations of each respective variable, and, accordingly, represented as elliptical shapes in the figure. Since the job insecurity intercept reflects the initial level of job insecurity, the positive association with short-time work shown in the figure (β = 0.47, p < 0.001 in all three models) supported our hypothesis that short-time work would be associated with higher levels of job insecurity (H1). Short-time work was also positively associated with the slope (β = 0.58–0.62, p = 0.030–0.042) and negatively with the quadratic growth term of job insecurity (β =  − 1.53– − 1.91, p < 0.001). In a quadratic LGCM, the slope can be interpreted as the immediate rate of linear change, while the quadratic term reflects the rate of change in the slope at subsequent time points (Bollen & Curran, 2006). The results thus indicate an inverted U-shaped growth trajectory. Figure 2 illustrates the development of job insecurity among employees affected by short-time work compared to the control group over time. This pattern was in line with our hypothesis that over time, job insecurity would increase, and then decrease again in response to short-time work (H2).
Regarding effects on the outcome variables, the level of job insecurity was negatively associated with the level of job satisfaction (β =  − 0.42, p < 0.001), life satisfaction (β =  − 0.35, p < 0.001) and affective well-being (β =  − 0.36, p < 0.001), respectively (see paths between the intercepts in Fig. 1). Hypothesis 3, which stated that job insecurity would be associated with lower levels of well-being, was supported. Negative relationships also emerged between the job insecurity slope and the slopes of job satisfaction (β =  − 0.41, p < 0.001), and affective well-being (β =  − 0.75, p < 0.001), but not life satisfaction. Our hypothesis that changes in job insecurity over time would be negatively associated with changes in well-being over time was partially supported (H4).

3.2 Job Insecurity as a Mediator Between Short-time Work and Well-being

As a first step, we ran general tests for the mediating role of job insecurity in the relationship between short-time work and well-being (H5) by inspecting the indirect relationships between short-time work and well-being via the job insecurity intercept and slope, respectively. Indirect effects via the job insecurity intercepts were significant on all outcomes (β =  − 0.19, SE = 0.04, p < 0.001; 95%-CI [− 0.26; − 0.13] for job satisfaction; β =  − 0.17; SE = 0.04; p < 0.001; 95%-CI [− 0.17; − 0.07] for life satisfaction; β =  − 0.14; SE = 0.03; p < 0.001; 95%-CI [− 0.19; − 0.08] for affective well-being). Indirect effects via the job insecurity slope were closely non-significant for job satisfaction (β =  − 0.25; SE = 0.14; p = 0.070; 95%-CI [− 0.47; − 0.02]) and affective well-being (β =  − 0.37; SE = 0.20; p = 0.061; 95%-CI [− 69; − 0.05]), and not significant for life satisfaction.
Second, we calculated more specific tests for mediation at each time point. The results are shown in Table 3: The indirect relationships over time mirrored the non-linear growth trajectory in job insecurity that followed the onset of short-time work: Indirect effects of short-time work via job insecurity were significant on all three outcomes at t1, t2 and t3, and additionally at t4 for job satisfaction while from t4 onwards (t5 for job satisfaction), the indirect effects were no longer significant. Hypothesis 5 did not receive strong support regarding linear trends across the whole observation period, but it was supported in terms of general levels of job insecurity and at specific time points. From t − 1 to t4, short-time work had moderate effects on job insecurity, whereas job insecurity showed consistent negative relationships of small to moderate size on all outcomes across the six time points.
Table 3
Standardized path coefficients, indirect and total effects of short-time work on well-being via job insecurity across time points, with bootstrapped 99% confidence intervals
 
STW → job insecurity → job satisfaction
STW → job insecurity → life satisfaction
STW → job insecurity → affective well-being
 
aa
b
c’
indirect
total
b
c’
indirect
total
b
c’
indirect
total
t− 1
             
 
0.31**
− 0.18**
0.03
− 0.05**
[− 0.06; − 0.001]
− 0.03
[− 0.22; 0.11]
− 0.09**
− 0.06
− 0.03*
[− 0.05; − 0.002]
− 0.08
[− 0.24; 0.07]
− 0.11**
− 0.04
− 0.03*
[− 0.06; − 0.01]
− 0.07
[− 0.23; 0.09]
t1
             
 
0.55**
− 0.20**
− 0.07
− 0.11**
[− 0.14; − 0.06]
− 0.18*
[− 0.30; 0.001]
− 0.09**
− 0.07
− 0.05**
[− 0.08; − 0.02]
− 0.12
[− 0.27; 0.02]
− 0.13**
− 0.01
− 0.07**
[− 0.11; − 0.03]
− 0.08
[− 0.23; 0.08]
t2
             
 
0.49**
− 0.18**
− 0.19*
− 0.09**
[− 0.12; − 0.04]
− 0.28**
[− 0.38; − 0.08]
− 0.08**
− 0.16*
− 0.04**
[− 0.07; − 0.02]
− 0.20*
[− 0.35; − 0.05]
− 0.11**
− 0.10
− 0.05**
[− 0.08; − 0.02]
− 0.15
[− 0.32; 0.08]
t3
             
 
0.28**
− 0.21**
− 0.08
− 0.06**
[− 0.08; − 0.01]
− 0.14
[− 0.31; 0.002]
− 0.10**
− 0.11
− 0.03*
[− 0.05; − 0.01]
− 0.14
[− 0.30; 0.02]
− 0.17**
− 0.02
− 0.05**
[− 0.08; − 0.02]
− 0.07
[− 0.24; 0.10]
t4
             
 
0.21*
− 0.17**
− 0.08
− 0.04*
[− 0.05; 0.002]
− 0.11
[− 0.22; 0.11]
− 0.07*
− 0.09
− 0.02
[− 0.03; 0.003]
− 0.10
[− 0.27; 0.06]
− 0.14**
0.00
− 0.03
[− 0.06; 0.00]
− 0.03
[− 0.20; 0.14]
t5
             
 
0.19
− 0.16**
− 0.04
− 0.03
[− 0.05; 0.02]
0.07
[− 0.16; 0.20]
− 0.07*
0.07
− 0.01
[− 0.03; 0.01]
− 0.09
[− 0.26; 0.08]
− 0.11**
− 0.11
− 0.02
[− 0.05; 0.01]
− 0.14
[− 0.31; 0.04]
N = 1211. Based on latent growth curve models with job insecurity as a time-varying covariate, calculated with maximum likelihood estimation and 5000 bootstrap samples. STW = short-time work. a = effect of short-time work on job insecurity; b = effect of job insecurity on outcome variable; c’ = remaining direct effect of short-time work on outcome after accounting for indirect effect via a and b. a The a-path is equal across all models and thus omitted from the results on life satisfaction and affective well-being. Bonferroni-adjusted p-values for six tests: *p < 0.008; **p < 0.001
As for our exploratory research question, the pattern of results suggests that the effect of short-time work via job insecurity on job satisfaction lasted for 4 years while it lasted for 3 years on life satisfaction and affective well-being before it receded: From t4 onwards (t5 for job satisfaction), employees affected by short-time work no longer differed from the control group, which is likely why indirect effects via the slopes did not reach statistical significance.
As a robustness check, we also calculated our models with an additional covariate indicating repeated exposure to short-time work (0 = no, 1 = yes), because 26 percent of the short-time workers were affected more than once beyond t1. After accounting for repeated exposure, the effects of short-time work on the job insecurity intercept and quadratic term were still significant, but not as concerns the slope (additionally, trends of p < 0.10 were visible for effects of repeated short-time work on the slope and quadratic term). At t4, the p-value for the indirect effect of short-time work on job satisfaction increased to p = 0.053 Other results did not differ from the models without repeated exposure. Overall, we found a similar pattern of results with the exception that effects of short-time work on job insecurity, and indirect effects on job satisfaction disappeared 1 year earlier.

3.3 Previous Unemployment as a Moderator

Hypothesis 6 stated that previous unemployment experience moderates the relationships between short-time work and job insecurity over time, such that employees with previous unemployment experience would respond to short-time work with higher levels (H6a) and more pronounced changes in perceived job insecurity (H6). In a first step, we compared the model fit of different multiple group LGCMs, subsequently freeing the growth parameters for the two groups with (n = 481) and without (n = 730) previous unemployment. For job satisfaction, a model with freely varying intercepts, slopes and quadratic terms across groups had the best fit (∆ χ2 (2) = 7.81, p = 0.020 compared to a model with quadratic terms constrained to be equal). For life satisfaction (∆ χ2 (2) = 13.86, p = 0.001) and affective well-being (∆ χ2 (2) = 11.42, p = 0.003), freely varying intercepts improved the fit; further freeing slopes or quadratic terms did not. Across the three models, respondents with previous unemployment experience showed higher initial levels of job insecurity (MIntercept = 1.87, SE = 0.05, p < 0.001) than respondents without previous unemployment (MIntercept = 1.65, SE = 0.04, p < 0.001; WT(1) = 13.88, p < 0.001). In the second step, we tested for group differences in the path coefficients from short-time work to job insecurity using Wald tests. The two groups differed neither in the effects of short-time work on the job insecurity intercept (no previous unemployment: B = 0.30–0.31; β = 0.53–0.55, p < 0.001; previous unemployment: β = 0.35–0.38, p = 0.001–0.003; WT(1) = 1.14–1.73, p = 0.188–0.286), nor on the slope (no previous unemployment: β = 0.33–0.48, p = 0.185–0.384; previous unemployment: β = 0.80–0.94, p = 0.015–0.051; WT(1) = 0.33–1.15, p = 0.284–0.566), or the quadratic growth term (no previous unemployment: β = -1.05– -1.94, p = 0.000–0.035; previous unemployment: β = − 1.81– − 2.00, p < 0.001; WT(1) = 0.21–1.01; p = 0.316–0.646). Respondents with previous unemployment thus showed higher levels of job insecurity, but did not react to short-time work more or less strongly. Hypothesis 6 was not supported.

4 Discussion

The aim of this study was to investigate the development of job insecurity and employee well-being in response to short-time work as a disruption of employment, taking into account previous unemployment as a potential moderator. The findings contribute to a growing body of studies on the long-term temporal mechanisms between job insecurity, its antecedents and outcomes (e.g., Kinnunen et al., 2014; Klug et al., 2019; Watson & Osberg, 2017), as well as the impact of past labor market experiences on later job insecurity and well-being (Barrech et al., 2011; De Witte, 2016; Ellonen & Nätti, 2015; Erlinghagen, 2008). Investigating the long-term repercussions for employee well-being, our findings also complement existing studies on immediate employee reactions to short-time work (Rauvola et al., 2022; Wels et al., 2022), and support job insecurity as an important explanatory mechanism underlying lingering effects on well-being.

4.1 Trajectories of Job Insecurity and Well-being Following Short-time Work

Our findings show that despite serving to prevent job loss, short-time work can have the paradoxical effect of triggering job insecurity, with negative repercussions for employee well-being: Short-time work predicted higher levels (H1), as well as an immediate increase in perceived job insecurity, followed by a decrease over time (H2). With regard to employee well-being, both levels (H3) and changes (H4) in job insecurity perceptions were negatively related to levels and changes in well-being, with the exception that changes in job insecurity were not related to changes in life satisfaction. Overall, short-time work was related to less favorable trajectories of well-being over time via job insecurity (H5). These findings are in line with previous research on organizational change and career disruptions as antecedents of job insecurity (Hofer et al., 2021; Keim et al., 2014), adding short-time work as a specific new determinant of insecurity perceptions. The findings are also in line with stress theories (Hobfoll, 1989; Lazarus & Folkman, 1984) and previous studies on the consequences of job insecurity (Jiang & Lavaysse, 2018).
What our study adds is a detailed analysis of the nonlinear temporal dynamics between antecedents, insecurity, and outcomes: Compared to the control group, employees affected by short-time work not only showed an immediate increase in job insecurity and a decrease in well-being, but these negative effects persisted 2 to 3 years after short-time work before they receded. This pattern suggests that short-time work may have long-lasting effects on employees’ perceptions of security, comparable to effects in the aftermath of actual layoffs (Dlouhy & Casper, 2021; Maertz et al., 2010). Looking at longer time frames and considering job insecurity as an explanatory mechanism may also explain why previous research has found that restorative effects of short-time work on well-being were rather short-lived (Rauvola et al., 2022).
Contrary to our expectations (H6), previous unemployment experiences did not seem to heighten employees’ susceptibility to job insecurity in response to objective antecedents (i.e., short-time work). In line with previous research (Ellonen & Nätti, 2015; Erlinghagen, 2008), past unemployment predicted present insecurity perceptions, but it does not seem to affect employees’ appraisals of the same situation as more threatening, at least not in the case of short-time work. Although these results require replication, they suggest that employee reactions to short-time work may be irrespective of previous labor market experiences.

4.2 Theoretical and Practical Implications

This study has several implications for theory and future research regarding the dynamics between job insecurity, its antecedents and consequences for employee well-being. First, regarding our research question how employee reactions to short-time work develop over time, our findings indeed suggest a nonlinear response in which employees’ job insecurity perceptions increased, then decreased, but remained elevated even after short-time work ended, only receding 3 years later (2 years when we took repeated short-time work into account). We also found different temporal dynamics between job insecurity and its antecedents compared to between job insecurity and well-being: As described above, short-time work had lingering effects on job insecurity perceptions. As insecurity perceptions decreased eventually, our results resemble a prolonged stress reaction model (Frese & Zapf, 1988), in which stress appraisals decrease only slowly over the years after the objective source of insecurity subsides: Employees showed elevated threat appraisals even after the objective stressor had ceased, but in contrast to an accumulation model, these reactions decreased eventually and did not persist throughout the whole study period. Regarding the consequences for well-being, the results resemble a more immediate stressor–strain reaction model (Frese & Zapf, 1988). That is, well-being responds immediately with synchronous decreases to increasing insecurity, but also synchronous improvements once perceived job insecurity is reduced (see also Bernhard‐Oettel et al., 2020; Kinnunen et al., 2014).
Second, our findings are generally in line with both appraisal theory (Lazarus & Folkman, 1984) and COR theory (Hobfoll, 1989), such that short-time work signaled a threat that triggered perceptions of job insecurity, which, in turn, were consistently linked to employee well-being over time. Considering the prolonged effect of short-time work on job insecurity, our findings are by comparison more supportive of COR theory, which states that resource loss tends to affect people more strongly than resource (re-)gain (i.e., the primacy of loss principle; Hobfoll et al., 2018).
Third, it is an unresolved question in the literature why past unemployment may affect future job insecurity perceptions. Our findings did not support the explanation that previously unemployed people react to the same objective cue (i.e., short-time work) with more subjective insecurity than their colleagues, suggesting that selection into less secure jobs upon re-employment may be more likely (see Shoss, 2017). The lack of an interaction between previous unemployment and short-time work also challenges the assumption in both COR and appraisal theory that previous experiences of loss would affect people’s susceptibility to future loss or threat. However, according to appraisal theory, the effect of unemployment may additionally depend on how successfully or unsuccessfully people have coped with it (for example, if they quickly found a new, stable full-time job, the threat of job loss might be less severe in the future). More research is needed for a more differentiated analyses of these mechanisms, for example considering the length, frequency and recency of unemployment spells (see Erlinghagen, 2008). To reach a conclusion, more research is needed to systematically test these mechanisms in a comparative approach and for different antecedents of job insecurity (e.g., restructuring, temporary employment). Additionally, the scope could be expanded to predictors and moderators of job insecurity perceptions beyond unemployment, such as employability (De Cuyper et al., 2012). It may also play a role how a previous job loss was experienced or attributed (e.g., to external circumstances or personal failure, see Winefield et al., 1992).
In terms of practical implications, our study underlines the importance of reducing employment uncertainty as an important determinant of employee well-being, as many others have already noted (Cheng & Chan, 2008; Jiang & Lavaysse, 2018; Sverke et al., 2002). Our analysis from the previous recession (2008–2009) can be informative regarding potential long-term effects of job retention schemes in the current crisis, which have been popular measures to buffer the economic repercussions of the pandemic (Drahokoupil & Müller, 2021).
We want to stress that our findings should not be read as advice to refrain from implementing short-time work. Obviously, it is a useful policy instrument during crises and preferable to mass layoffs, for employees, organizations, and societies alike (OECD, 2020b). Analyses of short-time work (furlough) in the UK during the pandemic indeed suggest that employees in job retention schemes reported worse mental health compared to secure employees, but were still better off than those who lost their jobs completely (Wels et al., 2022). Nevertheless, employers should be aware of the side effects for employee well-being and strive to alleviate experiences of job insecurity, for example by improving organizational communication to reduce rumors, re-organizing work efficiently to avoid ambiguity, and developing such measures in a participatory approach with employees (Abildgaard et al., 2018; Keim et al., 2014).

4.3 Strengths and Limitations

The study has some notable strengths: The large-scale longitudinal data allowed us to delve into within-person processes of job insecurity as it unfolds over time, and investigate long-term, nonlinear trajectories, addressing questions regarding the onset and reversibility of effects (see Klug et al., 2020, 2024a; Watson & Osberg, 2017). Keeping in mind that no single study can establish perfect causality, our study provides comparably strong evidence, because a) short-time work as our independent variable reflects an event with an identifiable onset, b) the repeated measures design allowed tracking the process unfolding upon the onset of short-time work, and c) the matched control group design accounted for selection effects into short-time work based on sociodemographic characteristics. As for effects of short-time work on employee perceptions, reverse causality (i.e., previous insecurity or well-being predicting short-time work) or self-selection based on other psychological dispositions and personality characteristics is unlikely, because short-time work is implemented for whole organizations or departments, not for individual employees. As for job insecurity and well-being, we analyzed synchronous relationships where reverse and reciprocal causation are possible (De Cuyper et al., 2012). However, the parallel process models allowed separating baseline levels from within-person changes in the variables over time. All in all, we can therefore conclude that short-time work likely triggered the rise in job insecurity and, in turn, a decline in subjective well-being.
Some limitations should be kept in mind as well when interpreting the results. The first limitation concerns the operationalization of constructs: All variables were measured as self-reports such that we cannot rule out common method bias (Podsakoff et al., 2003). However, our measurement models suggested that the variables differentiated sufficiently between one another, and it was unlikely that a single method factor explained all of the shared variance. Moreover, concepts such as perceived job insecurity, satisfaction and affective well-being require an individual’s introspection and are difficult to measure otherwise.
A related second limitation was due to the secondary analysis of existing data instead of a primary data collection specifically tailored to our study. This strategy was necessary to observe a large, representative sample over a long time frame, but limited the measurement of some constructs: Job insecurity and job satisfaction were each measured with single items which may have reduced their reliability (Gnambs & Buntins, 2017). However, regarding job satisfaction and job insecurity, single items have been supported as valid and reliable measures (Fisher et al., 2016; Matthews et al., 2022), and our results give no reason to suspect that these measures performed worse than the multi-item scales for life satisfaction and affective well-being. Regarding job insecurity, the item that we used did not allow differentiating affective from cognitive insecurity (see Shoss, 2017). However, specific mechanisms of cognitive versus affective job insecurity were not central to our research questions—we were interested in the overall experience of job insecurity, which includes both cognition and affect (Klug et al., 2024a, 2024b). This seemed appropriate in light of evidence showing that cognitive job insecurity tends to show substantial correlations with affective job insecurity (Chirumbolo et al., 2020; Jiang & Lavaysse, 2018). The correlations with short-time work as an objective indicator of insecurity also support this item’s validity in the study context. Additionally, single items of job insecurity have shown meaningful relationships with employee well-being, but with lower effect sizes than multi-item scales (Sverke et al., 2002). Therefore, the effects in this study may be underestimated, rendering Type I errors less likely (see also Gnambs & Buntins, 2017). Nevertheless, future research could replicate our findings with multidimensional job insecurity measures and physiological well-being indicators. Such an approach could also include qualitative job insecurity measures to explore whether short-time work may relate to concerns about deteriorating job quality, in addition to threats of losing the job as such.
Third, in our analysis of trajectories of job insecurity and well-being following short-time work, we were not able to account for time-varying predictors to rule out alternative explanations for the observed changes in the variables of interest (such as changes in task structures and workload, or turnover). We also did not differentiate effects depending on the duration of short-time work or the varying extent of reductions in working hours and earnings (Drahokoupil & Müller, 2021; Tušl et al., 2021). Accounting for the duration would have complicated comparisons with the control group, whereas information about the intensity was not available.
Fourth, the yearly time lag between observations may have been too long to detect more complex dynamics between short-time work and its consequences. On the other hand, a strength of our design was that with short-time work, the onset of the stressor was clearly identified—in such cases, longer time lags are preferable, because exposure time is comparable for all respondents (Dormann & Van de Ven, 2016). Still, additional studies could provide a more nuanced analysis across shorter intervals and consider the duration and extent of short-time work. Future studies could also consider additional, potentially opposing mediators, such as recovery from job demands through reduced hours, which might explain why the total effects in the mediation models across time points were often non-significant (Möhring et al., 2021; Rauvola et al., 2022; see also Rucker et al., 2011).
Regarding the generalizability of our findings, it should be noted that our study was situated in the specific context of Germany in the aftermath of the 2008 recession and focused on individual well-being. Due to the matching procedure, the sample was no longer random or representative of the whole active labor force in Germany, but rather of the subpopulation likely to encounter short-time work in their career. Typically, short-time work is more prevalent among labor market insiders (i.e., prime age, qualified, in permanent full-time employment), and in male-dominated sectors such as manufacturing and construction (Brenke et al., 2013; Tarullo & Desiere, 2023). At large, this is what we also observe in our data, with the subpopulation in short-time work being more likely to be male, manual workers with vocational degrees whereas, for example, age was comparable to the general population. Nevertheless, the findings are in line with meta-analyses across a range of countries and employees showing positive effects of organizational change on perceived job insecurity (Keim et al., 2014) and negative effects of job insecurity on employee well-being (Cheng & Chan, 2008; Jiang & Lavaysse, 2018; Llosa et al., 2018; Sverke et al., 2002). Since effect sizes between job insecurity, its antecedents and outcomes have been found to vary depending on the social safety net and economic climate (Fullerton et al., 2020; Probst & Jiang, 2017; Sverke et al., 2019), future studies should also test and replicate our findings in different countries and time periods. Comparative studies could also explicitly consider benefits and drawbacks of job retention schemes on labor market flexibility and the resulting effects for well-being on the population level.
Generalizing across different subgroups in the workforce also warrants some reflections: Our use of propensity score matching aimed at holding constant sociodemographic and job-related differences between employees with and without short-time work, to find the average pattern of effects irrespective of such influences. This approach and the necessary specifications for subgroup analyses in propensity score matching (Green & Stuart, 2014) limited our ability to test more moderators than previous unemployment. But some of the matching variables, such as age or contract type, could affect employee reactions to short-time work or job insecurity as additional moderators (see Cheng & Chan, 2008; De Cuyper & De Witte, 2007; Lübke & Erlinghagen, 2014). An employees’ family and financial situation may also influence the degree to which they feel dependent on their current job and thus influence the magnitude of effects (Richter et al., 2014). Whereas we tested moderating effects of previous unemployment experience, future studies could take different moderators into account to investigate the generalizability across sociodemographic groups, job- and household contexts.
Finally, this paper focused on the population of employees and their job insecurity perceptions, because only people who have a job can report such perceptions. Therefore, the study may have been influenced by a healthy worker effect (Dahl, 1993). That is, individuals with very strong strain reactions may have dropped out of employment, as well as job-insecure individuals who may have actually become unemployed (see also Garst et al., 2000). This may partly explain why some effects in the mediation analyses were rather small. However, the magnitude of relationships between objective antecedents such as short-time work, subjective stressors such as job insecurity, and outcomes of mental well-being is unlikely to be large for other reasons as well: First, especially in correlational studies, the complexities regarding measurement errors and potential unobserved influences render large effects unlikely. Second, the work environment is only one of many social, psychological and biological factors influencing an individual’s well-being trajectory, so that one particular stressor cannot reasonably be expected to explain a large share of the variance (Frese & Zapf, 1988).

5 Conclusion

This study is the first to establish long-term effects of short-time work on employee well-being, proposing job insecurity as an explanatory mechanism. Short-time work elicited a rise in job insecurity, accompanied by declining well-being over time, suggesting a slow stress-reaction model: Supporting the primacy of loss principle in COR theory, employees showed immediate negative reactions to short-time work in terms of job insecurity, but did not recover immediately after their resources were restored. The study provides novel insights, and comparably strong causal evidence, about the psychological side effects of job retention schemes, as well as temporal dynamics between job insecurity, its antecedents and consequences for well-being. Although short-time work is a useful measure to avoid layoffs, employers and policy makers should provide additional support to reduce accompanying experiences of job insecurity to protect employee well-being in economic crises.

Acknowledgements

The data used in this publication were made available to the authors by the German Socio-Economic Panel Study (SOEP) at the German Institute for Economic Research (DIW), Berlin. The second and third author participated within the NOWSTARS research program, with financial support from the Swedish Research Council for Health, Working Life and Welfare (FORTE Grant No. 2019-01311). Correspondence regarding this article should be addressed to Katharina Klug, University of Bremen, Faculty of Business Studies and Economics, Enrique-Schmidt-Strasse 1, 28359 Bremen.

Declarations

Conflict of interest

The authors declare no conflict of interest.

Ethical Approval

This study was based on secondary analysis of anonymized data, and therefore an ethics approval was not required. Detailed information on ethical clearance related to the German Socio-Economic Panel Study (SOEP) can be found on the website of the German Institute for Economic Research (DIW), Berlin (https://​www.​diw.​de/​soep).
Participants gave their informed consent prior to data collection. Detailed information on informed consent given by the participants related to the German Socio-Economic Panel Study (SOEP) can be found on the website of the German Institute for Economic Research (DIW), Berlin (https://​www.​diw.​de/​soep).
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://​creativecommons.​org/​licenses/​by/​4.​0/​.

Publisher's Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Anhänge

Supplementary Information

Below is the link to the electronic supplementary material.
Fußnoten
1
Throughout this paper, we use the term “short-time work” in keeping with the European terminology (European Commission, 2020), also to distinguish from the specific furlough in public service during government shut-downs in the US context (Baranik et al., 2019; S. Lee & Sanders, 2013).
 
2
Note that we did not calculate inferential test statistics to compare the study sample (i.e., the short-time workers and the control group together) with the sample of all employees, because these represent overlapping populations. It should also be noted that persons with a migration background are oversampled in the SOEP (Jacobsen et al., 2021). The actual share of employees with a migration background in the German workforce has been estimated around 27% (www.​destatis.​de).
 
Literatur
Zurück zum Zitat Barrech, A., Baumert, J., Emeny, R. T., Gündel, H., & Ladwig, K. H. (2011). Mid-life job insecurity associated with subjective well-being in old age: Results from the population-based MONICA/KORA study. Scandinavian Journal of Work, Environment and Health, Supplement, 37(2), 170–174. https://doi.org/10.5271/sjweh.3546CrossRef Barrech, A., Baumert, J., Emeny, R. T., Gündel, H., & Ladwig, K. H. (2011). Mid-life job insecurity associated with subjective well-being in old age: Results from the population-based MONICA/KORA study. Scandinavian Journal of Work, Environment and Health, Supplement, 37(2), 170–174. https://​doi.​org/​10.​5271/​sjweh.​3546CrossRef
Zurück zum Zitat Bernhard-Oettel, C., Eib, C., Griep, Y., & Leineweber, C. (2020). How do job insecurity and organizational justice relate to depressive symptoms and sleep difficulties: A multilevel study on immediate and prolonged effects in Swedish workers. Applied Psychology, 69(4), 1271–1300. https://doi.org/10.1111/apps.12222CrossRef Bernhard-Oettel, C., Eib, C., Griep, Y., & Leineweber, C. (2020). How do job insecurity and organizational justice relate to depressive symptoms and sleep difficulties: A multilevel study on immediate and prolonged effects in Swedish workers. Applied Psychology, 69(4), 1271–1300. https://​doi.​org/​10.​1111/​apps.​12222CrossRef
Zurück zum Zitat Blackmore, C., & Kuntz, J. R. C. (2011). Antecedents of job insecurity in restructuring organisations: An empirical investigation. New Zealand Journal of Psychology, 40(3), 7–18. Blackmore, C., & Kuntz, J. R. C. (2011). Antecedents of job insecurity in restructuring organisations: An empirical investigation. New Zealand Journal of Psychology, 40(3), 7–18.
Zurück zum Zitat Bollen, K. A., & Curran, P. J. (2006). Latent Curve Models. Wiley-Interscience. Bollen, K. A., & Curran, P. J. (2006). Latent Curve Models. Wiley-Interscience.
Zurück zum Zitat Brenke, K., Rinne, U., & Zimmermann, K. F. (2013). Short-time work: The German answer to the Great Recession. International Labour Review, 152(2), 287–305.CrossRef Brenke, K., Rinne, U., & Zimmermann, K. F. (2013). Short-time work: The German answer to the Great Recession. International Labour Review, 152(2), 287–305.CrossRef
Zurück zum Zitat De Witte, H. (1999). Job insecurity and psychological well-being: Review of the literature and exploration of some unresolved issues. European Journal of Work and Organizational Psychology, 8(2), 155–177.CrossRef De Witte, H. (1999). Job insecurity and psychological well-being: Review of the literature and exploration of some unresolved issues. European Journal of Work and Organizational Psychology, 8(2), 155–177.CrossRef
Zurück zum Zitat De Cuyper, N., Mäkikangas, A., Kinnunen, U., Mauno, S., & De Witte, H. (2012). Cross-lagged associations between perceived external employability, job insecurity, and exhaustion: Testing gain and loss spirals according to the Conservation of Resources Theory. Journal of Organizational Behavior, 33(6), 770–788. https://doi.org/10.1002/job.1800CrossRef De Cuyper, N., Mäkikangas, A., Kinnunen, U., Mauno, S., & De Witte, H. (2012). Cross-lagged associations between perceived external employability, job insecurity, and exhaustion: Testing gain and loss spirals according to the Conservation of Resources Theory. Journal of Organizational Behavior, 33(6), 770–788. https://​doi.​org/​10.​1002/​job.​1800CrossRef
Zurück zum Zitat Dormann, C., & Van Ven, B. (2016). Timing in methods for studying psychsosocial factors at work. In M. F. Dollard, A. Shimazu, R. B. Nordin, P. Brough, & M. R. Tuckey (Eds.), Psychosocial Factors at Work in the Asia Pacific: From Theory to Practice (pp. 89–116). Springer. https://doi.org/10.1007/978-3-319-44400-0CrossRef Dormann, C., & Van Ven, B. (2016). Timing in methods for studying psychsosocial factors at work. In M. F. Dollard, A. Shimazu, R. B. Nordin, P. Brough, & M. R. Tuckey (Eds.), Psychosocial Factors at Work in the Asia Pacific: From Theory to Practice (pp. 89–116). Springer. https://​doi.​org/​10.​1007/​978-3-319-44400-0CrossRef
Zurück zum Zitat Esping-Andersen, G. (1990). The Three Worlds of Welfare Capitalism. Polity Press. Esping-Andersen, G. (1990). The Three Worlds of Welfare Capitalism. Polity Press.
Zurück zum Zitat European Commission. (2020). Coucil regulation on the establishment of a European instrument for temporary support to mitigate unemployment risks in an emergency (SURE) following the COVID-19 outbreak. European Commission. (2020). Coucil regulation on the establishment of a European instrument for temporary support to mitigate unemployment risks in an emergency (SURE) following the COVID-19 outbreak.
Zurück zum Zitat Ferrie, J. E., Shipley, M. J., Stansfeld, S. A., & Marmot, M. G. (2002). Effects of chronic job insecurity and change in job security on self reported health, minor psychiatric morbidity, physiological measures, and health related behaviours in British civil servants: The Whitehall II study. Journal of Epidemiology & Community Health, 56(6), 450–454. https://doi.org/10.1136/jech.56.6.450CrossRef Ferrie, J. E., Shipley, M. J., Stansfeld, S. A., & Marmot, M. G. (2002). Effects of chronic job insecurity and change in job security on self reported health, minor psychiatric morbidity, physiological measures, and health related behaviours in British civil servants: The Whitehall II study. Journal of Epidemiology & Community Health, 56(6), 450–454. https://​doi.​org/​10.​1136/​jech.​56.​6.​450CrossRef
Zurück zum Zitat Frese, M., & Zapf, D. (1988). Methodological issues in the study of work stress: Objective vs subjective measurement of work stress and the question of longitudinal studies. In C. L. Cooper & R. Payne (Eds.), Causes, coping and consequences of stress at work (pp. 375–411). Wiley. Frese, M., & Zapf, D. (1988). Methodological issues in the study of work stress: Objective vs subjective measurement of work stress and the question of longitudinal studies. In C. L. Cooper & R. Payne (Eds.), Causes, coping and consequences of stress at work (pp. 375–411). Wiley.
Zurück zum Zitat Garst, H., Frese, M., & Molenaar, P. C. M. (2000). The temporal factor of change in stressor–strain relationships: A growth curve model on a longitudinal study in East Germany. Journal of Applied Psychology, 85(3), 417–438.CrossRef Garst, H., Frese, M., & Molenaar, P. C. M. (2000). The temporal factor of change in stressor–strain relationships: A growth curve model on a longitudinal study in East Germany. Journal of Applied Psychology, 85(3), 417–438.CrossRef
Zurück zum Zitat Hu, L., & Bentler, P. M. (1999). Cutoff criteria for fit indexes in covariance structure analysis: Conventional criteria versus new alternatives. Structural Equation Modeling, 6(1), 1–55.CrossRef Hu, L., & Bentler, P. M. (1999). Cutoff criteria for fit indexes in covariance structure analysis: Conventional criteria versus new alternatives. Structural Equation Modeling, 6(1), 1–55.CrossRef
Zurück zum Zitat Infurna, F. J., Gerstorf, D., Ram, N., & Heckhausen, J. (2016). Analytic strategies for the study of adaptation to major life events: Making the most of large-scale longitudinal surveys. In H. P. Blossfeld, J. von Maurice, M. Bayer, & J. Skopek (Eds.), Methodological issues of longitudinal surveys: The example of the national educational panel study (pp. 19–35). Springer. https://doi.org/10.1007/978-3-658-11994-2CrossRef Infurna, F. J., Gerstorf, D., Ram, N., & Heckhausen, J. (2016). Analytic strategies for the study of adaptation to major life events: Making the most of large-scale longitudinal surveys. In H. P. Blossfeld, J. von Maurice, M. Bayer, & J. Skopek (Eds.), Methodological issues of longitudinal surveys: The example of the national educational panel study (pp. 19–35). Springer. https://​doi.​org/​10.​1007/​978-3-658-11994-2CrossRef
Zurück zum Zitat Kinnunen, U., Mäkikangas, A., Mauno, S., De Cuyper, N., & De Witte, H. (2014). Development of perceived job insecurity across two years: Associations with antecedents and employee outcomes. Journal of Occupational Health Psychology, 19(2), 243–258.CrossRef Kinnunen, U., Mäkikangas, A., Mauno, S., De Cuyper, N., & De Witte, H. (2014). Development of perceived job insecurity across two years: Associations with antecedents and employee outcomes. Journal of Occupational Health Psychology, 19(2), 243–258.CrossRef
Zurück zum Zitat Klug, K., Selenko, E., Van Hootegem, A., Sverke, M., & De Witte, H. (2024a). A lead article to go deeper and broader in job insecurity research: Understanding an individual perception in its social and political context. Applied Psychology: An International Review, 1–34. Klug, K., Selenko, E., Van Hootegem, A., Sverke, M., & De Witte, H. (2024a). A lead article to go deeper and broader in job insecurity research: Understanding an individual perception in its social and political context. Applied Psychology: An International Review, 1–34.
Zurück zum Zitat Klug, K., Selenko, E., Van Hootegem, A., Sverke, M., & De Witte, H. (2024b). Seeing the forest for the trees: A response to commentaries on job insecurity conceptualizations, processes and social context. Applied Psychology, 1–11. https://doi.org/10.1111/apps.12537 Klug, K., Selenko, E., Van Hootegem, A., Sverke, M., & De Witte, H. (2024b). Seeing the forest for the trees: A response to commentaries on job insecurity conceptualizations, processes and social context. Applied Psychology, 1–11. https://​doi.​org/​10.​1111/​apps.​12537
Zurück zum Zitat Lazarus, R. S., & Folkman, S. (1984). Stress, appraisal and coping. Springer. Lazarus, R. S., & Folkman, S. (1984). Stress, appraisal and coping. Springer.
Zurück zum Zitat Liu, L. C., Flay, B. R., Amuwo, S. A., Bell, C. C., Berbaum, M. L., Campbell, R. T., Julia Cowell, R. N., Cooksey, J., Dancy, B. L., Graumlich, S., Hedeker, D., Jagers, R. J., Levy, S. R., Paikoff, R. L., Punwani, I., Segawa, E., & Weissberg, R. P. (2009). Evaluating mediation in longitudinal multivariate data: Mediation effects for the Aban Aya Youth Project drug prevention program. Prevention Science, 10(3), 197–207. https://doi.org/10.1007/s11121-009-0125-1CrossRef Liu, L. C., Flay, B. R., Amuwo, S. A., Bell, C. C., Berbaum, M. L., Campbell, R. T., Julia Cowell, R. N., Cooksey, J., Dancy, B. L., Graumlich, S., Hedeker, D., Jagers, R. J., Levy, S. R., Paikoff, R. L., Punwani, I., Segawa, E., & Weissberg, R. P. (2009). Evaluating mediation in longitudinal multivariate data: Mediation effects for the Aban Aya Youth Project drug prevention program. Prevention Science, 10(3), 197–207. https://​doi.​org/​10.​1007/​s11121-009-0125-1CrossRef
Zurück zum Zitat Möhring, K., Naumann, E., Reifenscheid, M., Wenz, A., Rettig, T., Krieger, U., Friedel, S., Finkel, M., Cornesse, C., & Blom, A. G. (2021). The COVID-19 pandemic and subjective well-being: Longitudinal evidence on satisfaction with work and family. European Societies, 23(S1), S601–S617. https://doi.org/10.1080/14616696.2020.1833066CrossRef Möhring, K., Naumann, E., Reifenscheid, M., Wenz, A., Rettig, T., Krieger, U., Friedel, S., Finkel, M., Cornesse, C., & Blom, A. G. (2021). The COVID-19 pandemic and subjective well-being: Longitudinal evidence on satisfaction with work and family. European Societies, 23(S1), S601–S617. https://​doi.​org/​10.​1080/​14616696.​2020.​1833066CrossRef
Zurück zum Zitat Müller, T., & Schulten, T. (2020). Ensuring fair short-time work - a European overview. In ETUI Policy Brief: European Economic, Employment and Social Policy (Issue 7). Müller, T., & Schulten, T. (2020). Ensuring fair short-time work - a European overview. In ETUI Policy Brief: European Economic, Employment and Social Policy (Issue 7).
Zurück zum Zitat Muthén, L. K., & Muthén, B. O. (1998–2017). Mplus User’s Guide (8th ed.). Muthén & Muthén. Muthén, L. K., & Muthén, B. O. (1998–2017). Mplus User’s Guide (8th ed.). Muthén & Muthén.
Zurück zum Zitat OECD. (2020b). Job retention schemes during the COVID-19 lockdown and beyond (Issue August). OECD. (2020b). Job retention schemes during the COVID-19 lockdown and beyond (Issue August).
Zurück zum Zitat Rauvola, R. S., Rudolph, C. W., & Zacher, H. (2022). Short-term effects of short-term work: Dynamics in fatigue across two national lockdowns. Journal of Occupational and Environmental Medicine. Rauvola, R. S., Rudolph, C. W., & Zacher, H. (2022). Short-term effects of short-term work: Dynamics in fatigue across two national lockdowns. Journal of Occupational and Environmental Medicine.
Zurück zum Zitat Rubin, D. B. (1973). Matching to remove bias in observational studies. Biometrics, 29(1), 159–183.CrossRef Rubin, D. B. (1973). Matching to remove bias in observational studies. Biometrics, 29(1), 159–183.CrossRef
Zurück zum Zitat Schulten, T., & Müller, T. (2020). Kurzarbeitergeld in der Corona-Krise. Aktuelle Regelungen in Deutschland und Europa (Policy Brief, Issue 38). Schulten, T., & Müller, T. (2020). Kurzarbeitergeld in der Corona-Krise. Aktuelle Regelungen in Deutschland und Europa (Policy Brief, Issue 38).
Zurück zum Zitat Sverke, M., & Hellgren, J. (2002). The nature of job insecurity: Understanding employment uncertainty on the brink of a new millennium. Applied Psychology: An International Review, 51(1), 23–42.CrossRef Sverke, M., & Hellgren, J. (2002). The nature of job insecurity: Understanding employment uncertainty on the brink of a new millennium. Applied Psychology: An International Review, 51(1), 23–42.CrossRef
Zurück zum Zitat Sverke, M., Hellgren, J., & Näswall, K. (2002). No security: A meta-analysis and review of job insecurity and its consequences. Journal of Occupational Health Psychology, 7(3), 242–264.CrossRef Sverke, M., Hellgren, J., & Näswall, K. (2002). No security: A meta-analysis and review of job insecurity and its consequences. Journal of Occupational Health Psychology, 7(3), 242–264.CrossRef
Zurück zum Zitat Sverke, M., Låstad, L., Hellgren, J., Richter, A., & Näswall, K. (2019). A meta-analysis of job insecurity and employee performance: Testing temporal aspects, rating source, welfare regime, and union density as moderators. International Journal of Environmental Research and Public Health, 16(14), 2536. https://doi.org/10.3390/ijerph16142536CrossRef Sverke, M., Låstad, L., Hellgren, J., Richter, A., & Näswall, K. (2019). A meta-analysis of job insecurity and employee performance: Testing temporal aspects, rating source, welfare regime, and union density as moderators. International Journal of Environmental Research and Public Health, 16(14), 2536. https://​doi.​org/​10.​3390/​ijerph16142536CrossRef
Zurück zum Zitat Tarullo, G., & Desiere, S. (2023). Short-Time Work or Unemployment ? On the role of individual traits and job-related features during COVID-19. 8th European User Conference for EU-Microdata. Microdata from Eurostat SILC, LFS, AES, SES, CIS, CSIS, EHIS, HBS and TUS. Tarullo, G., & Desiere, S. (2023). Short-Time Work or Unemployment ? On the role of individual traits and job-related features during COVID-19. 8th European User Conference for EU-Microdata. Microdata from Eurostat SILC, LFS, AES, SES, CIS, CSIS, EHIS, HBS and TUS.
Zurück zum Zitat Tušl, M., Brauchli, R., Kerksieck, P., & Bauer, G. F. (2021). Perceived impact of the COVID-19 crisis on work and private life and its association with mental well-being and self-rated health in German and Swiss employees: A cross-sectional study. BMC Public Health, 21(741), 1–21. Tušl, M., Brauchli, R., Kerksieck, P., & Bauer, G. F. (2021). Perceived impact of the COVID-19 crisis on work and private life and its association with mental well-being and self-rated health in German and Swiss employees: A cross-sectional study. BMC Public Health, 21(741), 1–21.
Zurück zum Zitat Vander Elst, T., Van den Broeck, A., De Cuyper, N., & De Witte, H. (2014). On the reciprocal relationship between job insecurity and employee well-being: Mediation by perceived control? Journal of Occupational and Organizational Psychology, 87(4), 671–693. https://doi.org/10.1111/joop.12068CrossRef Vander Elst, T., Van den Broeck, A., De Cuyper, N., & De Witte, H. (2014). On the reciprocal relationship between job insecurity and employee well-being: Mediation by perceived control? Journal of Occupational and Organizational Psychology, 87(4), 671–693. https://​doi.​org/​10.​1111/​joop.​12068CrossRef
Zurück zum Zitat Warr, P. (1994). A conceptual framework for the study of work and mental health. Work & Stress, 8(2), 84–97.CrossRef Warr, P. (1994). A conceptual framework for the study of work and mental health. Work & Stress, 8(2), 84–97.CrossRef
Zurück zum Zitat Wels, J., Booth, C., Wielgoszewska, B., Green, M. J., Di Gessa, G., Huggins, C. F., Griffith, G. J., Kwong, A. S. F., Bowyer, R. C. E., Maddock, J., Patalay, P., Silverwood, R. J., Fitzsimons, E., Shaw, R., Thompson, E. J., Steptoe, A., Hughes, A., Chaturvedi, N., Steves, C. J., & Ploubidis, G. B. (2022). Mental and social wellbeing and the UK coronavirus job retention scheme: Evidence from nine longitudinal studies. Social Science and Medicine., 308, 115226. https://doi.org/10.1016/j.socscimed.2022.115226CrossRef Wels, J., Booth, C., Wielgoszewska, B., Green, M. J., Di Gessa, G., Huggins, C. F., Griffith, G. J., Kwong, A. S. F., Bowyer, R. C. E., Maddock, J., Patalay, P., Silverwood, R. J., Fitzsimons, E., Shaw, R., Thompson, E. J., Steptoe, A., Hughes, A., Chaturvedi, N., Steves, C. J., & Ploubidis, G. B. (2022). Mental and social wellbeing and the UK coronavirus job retention scheme: Evidence from nine longitudinal studies. Social Science and Medicine., 308, 115226. https://​doi.​org/​10.​1016/​j.​socscimed.​2022.​115226CrossRef
Metadaten
Titel
The Paradox of Job Retention Schemes: A Latent Growth Curve Modeling Approach to Immediate and Prolonged Effects of Short-Time Work on Job Insecurity and Employee Well-Being
verfasst von
Katharina Klug
Claudia Bernhard-Oettel
Magnus Sverke
Publikationsdatum
01.08.2024
Verlag
Springer Netherlands
Erschienen in
Journal of Happiness Studies / Ausgabe 6/2024
Print ISSN: 1389-4978
Elektronische ISSN: 1573-7780
DOI
https://doi.org/10.1007/s10902-024-00787-y