Next Article in Journal
Politics, Power, and Influence: Defense Industries in the Post-Cold War
Next Article in Special Issue
“It’s Not Just about Work and Living Conditions”: The Underestimation of the COVID-19 Pandemic for Black Canadian Women
Previous Article in Journal / Special Issue
Entrepreneurial Behavior of SMEs and Characteristics of the Managers of Northwest Mexico
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:

Do Work–Life Measures Really Matter? The Impact of Flexible Working Hours and Home-Based Teleworking in Preventing Voluntary Employee Exits

Faculty of Sociology, Bielefeld University, Universitätsstraße 25, D-33615 Bielefeld, Germany
Author to whom correspondence should be addressed.
Soc. Sci. 2021, 10(1), 9;
Received: 10 December 2020 / Revised: 31 December 2020 / Accepted: 31 December 2020 / Published: 5 January 2021
(This article belongs to the Special Issue Gender, Work-Family Interface and Organizational Action)


Numerous studies have demonstrated the importance of work–life measures, which are designed to contribute to job quality and help reconcile employees’ work and personal lives. In our study, we asked whether such measures can also work as inducements to prevent employees from voluntarily leaving a firm. We considered flexible working hours and home-based teleworking as flexibility measures that are potentially attractive to all employees. To address the possible bias caused by sketchy implementation and their actual selective use, we chose to examine employees’ perceptions of the offer of these measures. We investigated the moderation of the effect by organizational culture and supervisor and coworker support. We controlled for several indicators of job quality, such as job satisfaction and perceived fairness, to isolate specific ways in which work–life measures contributed to voluntary employee exit, and checked for a selective attractiveness of work–life measures to parents and women as the main caregivers. Using a three-wave panel employer–employee survey, we estimated multilevel mixed-effects logistic regression models for 5452 employees at 127 large German establishments. Our results confirmed that both types of flexibility measures were associated with a lower probability of voluntarily exit. This applied more to men than to women, and the probability was reduced by a demanding organizational culture. Both measures seemed not to be specifically designed to accommodate main caregivers but were attractive to the whole workforce.

1. Introduction

With the growing diversity of the workforce over recent decades, the job elements that employees appreciate and expect from their employers have become more heterogeneous (Coyle-Shapiro and Conway 2004; Coyle-Shapiro and Shore 2007; Eurofond 2017; Guest 2016). A good job is no longer defined by income or career opportunities alone but also by flexibility options and the opportunity to reconcile work and personal life (Rhee et al. 2020). Moreover, the obligation to create these conditions has increasingly shifted from being an individual concern to an institutional and organizational responsibility (Kossek and Distelberg 2009; Moen 2015). Consequently, measures that provide opportunities to better reconcile employees’ work and personal lives—so-called work–life measures—have been increasingly important and have become an integral part of both public and academic discussions (Kossek and Distelberg 2009; Moen 2015). Indeed, previous research has shown convincingly that such work–life measures are positively related to job satisfaction (Baltes et al. 1999; Baruch 2000; Kelliher and Anderson 2008), work–life balance (Kelly et al. 2011), organizational commitment (Grover and Crooker 1995; Kelliher and Anderson 2010) and exit intentions (Batt and Valcour 2003; Moen et al. 2017). However, the effects of work–life measures are not unequivocally positive. They may be regarded as a form of compensating differentials to higher pay (Abendroth and Diewald 2019; Van der Lippe et al. 2019), may blur the boundaries between work and personal life in a way that overtaxes employees’ self-organization (Lott 2015) or may lead to conflicts with coworkers and supervisors (Gajendran and Harrison 2007). Despite these ambivalent findings, the implementation of work–life measures is in line with the increased need for flexibility among the workforce and sends a strong signal that a company is willing to take the need to reconcile their employees’ work and personal lives seriously (Grover and Crooker 1995).
Still, little is known about whether these measures might not only improve job satisfaction and commitment but also might help dissuade employees from leaving a firm. Voluntary employee exit is when an employee decides to leave a firm without being forced to do so as a result of a dismissal. An employee may decide to exit voluntarily in reaction to a severe disappointment with the present employer or due to better prospects elsewhere in the labor market, or both. It is therefore far from evident that simply offering work–life measures will have a distinct influence on an employee’s decision to exit; rather, these measures are but one possible strategy for strengthening an employee’s organizational commitment and perceived support, which are crucial for preventing an unwanted exit (Allen et al. 2003; Maertz et al. 2007; McFarlane Shore et al. 1990). Even when the lack of work–life measures may trigger an intention to exit, not all such intentions lead to action1, although they are a strong predictor of actual exits (Vandenberg and Barnes Nelson 1999). Because exiting behaviors are ridden with uncertainty and can have far-reaching consequences, they tend to occur only when strong push or pull factors are present (Cho and Lewis 2012; Lee and Mitchell 1994).
In this paper, we consider the association of work–life measures and voluntary employee exits. To the best of our knowledge, ours is the first study to investigate the impact of these measures on actual exits instead of mere exit intentions in a heterogeneous sample of employees from different occupational and socio-demographic groups and in different work environments in Germany.
At the side of work–life measures, we focus on two measures which enhance flexibility: the already widespread flexibilization of working hours and the comparatively less widespread (especially in Germany) home-based teleworking (see Chung and van der Lippe 2020). Such measures have the potential to improve the reconciliation of work and personal life for all groups within the workforce. Although these two measures are not specifically designed to address dependent-care responsibilities (e.g., childcare or eldercare) and can facilitate dual-career lifestyles or accommodate employees’ leisure activities and recreation, they may also help to alleviate such demands (Winslow 2005). However, since the duties required in dependent care are the least negotiable, employees with such responsibilities may be in greater need of flexibility measures than are those without such responsibilities. In general, this consideration applies to women more than to men (e.g., Hobler et al. 2020; Trappe et al. 2015) and to parents with children at home more than to childless employees (Winslow 2005).
For flexibility measures to be effective as a perceived benefit, an employer must convincingly communicate their availability to the employee. Research has shown that when a company has a family-hostile organizational culture, insists on a rigid presence culture and fails to provide support, the employee’s willingness to accept such measures is undermined (Den Dulk 2001; Lott and Abendroth 2020). In our analysis, we addressed these concerns in two ways: first, we did not rely on statements provided by human resource managers that the establishment offers such measures; instead, we relied on the individual employees’ perceptions that such measures exist, which proves to be a more restrictive indicator than information obtained from the HRM (= Human Resource Management) . The employees would (a) indicate the existence of these measures mostly if they perceived them as being available to themselves; (b) most likely not confirm the existence of these measures if their implementation would be negatively stereotyped (see Cook et al. 2020); and (c) not recognize it as an actual offer if the employer communicated the availability of these measures tentatively and reservedly. By focusing on the employees’ perceptions, we were able to avoid bias, in that these measures might be implemented solely to pay lip service to the public and for political reasons and are only communicated to the workforce as a benefit with some reservations (Brunsson 2003; De Menezes and Kelliher 2017; Den Dulk 2001; Den Dulk and De Ruijter 2008). The second way in which we addressed the concerns stated above was to determine whether the association between voluntary employee exits and the employees’ perceived availability of flexibility measures was moderated by organizational culture and by supervisor and coworker support.
For our survey, we used the LEEP-B3 data (Diewald et al. 2014), a linked employer–employee study based on a representative sample of large German establishments (i.e., establishments with more than 500 employees who paid social security taxes) and a random sample of the employees nested within these firms. The survey included 5452 employees from 127 large establishments in Germany. To date, the study has consisted of three waves. The panel design enabled us to calculate whether the availability of flexibility measures in the previous wave was associated with voluntary employee exits until the subsequent wave. The design also allowed us to carry out multilevel analyses, with mixed effects clustered by establishments, thus more successfully controlling for unobserved confounders than if we had used data from unrelated employees.

2. Theory and Hypotheses

2.1. Work–Life Measures as an Inducement to Prevent Voluntary Employee Exits

Previous research has consistently shown that dissatisfaction with working conditions and unfulfilled expectations are associated with exit intentions (Batt and Valcour 2003; Mobley 1977; Moen et al. 2017; Rousseau 1995). Actual exits, however, are linked to more prerequisites than simply the perception of job satisfaction or commitment: the employee’s personal situation overall or specific work conditions and the availability of alternative job opportunities are also important predictors of voluntary employee exit (Mobley 1977). Accordingly, whereas it can be expected that work–life measures are likely to have a strong impact on job satisfaction and similar aspects of job quality, as well as on exit intentions, the relevance of these effects to actual exits is less obvious.
There are three reasons why flexibility-focused work–life measures may nevertheless serve as inducements to prevent voluntary employee exits: first, these measures serve as a signal to the employees that their employer cares about their reconciliation needs and work–life balance (Butts et al. 2013; Grover and Crooker 1995). Second, they serve as a resource for employees by providing a basis for negotiations regarding an increase of flexibility within the employment relationship, since the bargaining power of employees is strengthened and a signal against negative stereotyping is sent to those who are influential in the bargaining process (mainly supervisors or human resource managers) (Grover and Crooker 1995; Van der Lippe et al. 2019). Moreover, flexibility measures in particular imply greater autonomy with respect to the employee’s work time and workplace (Lott 2015; Moen et al. 2011, 2017). Third, although the extent to which work–life measures have been implemented has grown over the recent past, major differences still exist depending on the size of the company. As can be seen in our sample, though there exists remarkable heterogeneity in the offer of work–life measures, they are widespread across large organizations, whereas according to other research small and medium-sized firms are still lagging behind this trend (Frodermann et al. 2018). Therefore, these measures may still be a potential pull factor for employees.
However, following the theory of neoinstitutionalism, employers may implement HRM practices to meet an increased external demand and thus gain legitimacy from other actors, such as politicians, the media or customers, without actually implementing these measures internally (Brunsson 2003; Den Dulk 2001; Den Dulk and De Ruijter 2008). One reason for this lack of follow-through is that some firms still adhere to the classical concepts of an “ideal worker norm” (Acker 1990; Lott and Abendroth 2020). In this sense, an “implementation gap” can be observed (Abendroth and Diewald 2019), meaning that, despite the formal implementation of such measures, the workforce remains unaware of the measures because the employer fails to communicate their availability. When an employer does not seriously implement these measures, their signaling effect can be lessened or even denied (De Menezes and Kelliher 2017; Den Dulk 2001; Philippe and Koehler 2005; Scheidler et al. 2019).
Instead of examining the promise of work–life measures, many scholars have focused on their actual use (e.g., Abendroth and Reimann 2018; Kelliher and Anderson 2008; Moen et al. 2017). However, actual use at a given point in time may be dependent on specific personal living conditions that change over an employee’s individual life course or on specific time-dependent working conditions, such as seasonal requests or demands from coworkers or supervisors (Lambert et al. 2008). We prefer to focus on whether employees know about the opportunity to use work–life measures in their particular establishment. A formal offer of work–life measures is by no means identical to a respective entitlement for all groups within a workforce (Kossek and Lautsch 2018; Lott and Abendroth 2020), nor are employees made aware of or even encouraged to make use of them (Den Dulk 2001). The perception of availability is shaped by at least two additional factors (e.g., Cook et al. 2020). First, a sense of entitlement depends on whether the measures being offered are in line with the employee’s own conception of what is normal, feasible and appropriate work–family support (Lewis and Smithson 2001). Second, perceived availability may be dependent on the organizational work–family culture and the extent to which an organization does not stigmatize flexible workplace arrangements but unanimously supports and values the integration of employees’ work and family lives (Lott and Abendroth 2020; Kossek et al. 2010). Only when the employer seriously implements and communicates this type of support, as reflected in the employee’s perception that such work–life measures are available, may the offer of work–life measures serve as an inducement to remain in an establishment.
Thus, for our first hypothesis, we assume the following:
Hypothesis 1 (H1).
If employees perceive work–life measures to be available in their establishment, they are less likely to exit.

2.2. The Influence of a Supportive Work Environment

Besides the seriousness of implementing work–life measures, the success of signaling these strategies depends on their embeddedness within the organizational norms and values, as well as on support provided by gatekeepers and those in key positions of responsibility (Galinsky and Stein 1990). Current research shows that there are different organizational logical approaches to the implementation of work–life measures that are strongly influenced by an organization’s goals and norms (Abendroth and Diewald 2019). Here, a rough distinction can be made between two modes: the implementation can be undertaken either to support the flexibility and reconciliation needs of employees or, counteractively, to serve the flexibility needs of the employer and strengthen the norm of the ideal worker (Abendroth and Diewald 2019; Den Dulk 2001).
In the first mode, the implementation would be accompanied by “cultural work–life support” (Kossek et al. 2010, p. 4), which includes support provided by supervisors and coworkers and the family-friendliness of the organization. An overall family-friendly organizational culture and support provided by direct supervisors and coworkers not only have a positive impact on the reconciliation of work and personal demands, but they also influence whether and how formal offers of work–life measures are understood by employees as being seriously meant (Van der Voet et al. 2015). However, aside from the successful communication of the availability of these measures, a supportive work environment may also increase the negative impact of work–life measures on voluntary employee exits.
Research has shown not only that perceived supervisor support reduces exit intentions (Kalidass and Bahron 2015; Maertz et al. 2007), but also that supervisors are a critical intersection between an organization’s practices and the employee (Thompson et al. 1999; Van der Voet et al. 2015). Thus, supervisors have an influence on whether the measures are implemented supportively, are denied or are even counteracted (Allen 2001; Den Dulk and De Ruijter 2008; Hammer et al. 2009). A denial or counteraction of the possibility to reconcile work and personal life can turn the measures into meaningless offers and even reinforce the negative stereotyping of employees who would benefit from them (Kossek et al. 2010). Supervisor support in the association between flexibility measures and work–life balance has a significant positive mediating effect (Abendroth and Reimann 2018) and should therefore play a major role in the signaling effect of the measures and how they influence exits.
Moreover, support from coworkers (e.g., by taking over another employee’s work tasks) is often seen as a resource for employees who have to balance work and personal-life demands. Research has shown that such support is positively related to perceived organizational involvement and work–life balance (e.g., Hayton et al. 2012; Pisarski et al. 2008). Thus, we assume the following hypothesis:
Hypothesis 2 (H2).
The negative association of flexibility measures with voluntary employee exits is stronger for employees who have highly supportive supervisors and coworkers than for those who receive little or even no support.
Furthermore, family-friendliness can be represented by the organizational culture, which carries values and norms (Denison 1996) and thus determines the implementation and acceptance of work–life measures (e.g., Galinsky and Stein 1990; Starrels 1992; Thompson et al. 1999). Within establishments in which reconciliation-supporting norms prevail, the offer of work–life measures should be perceived as less prejudicial (Almer et al. 2004; Konrad and Yang 2012), and the measures should also be implemented and executed in pursuit of the accomplishment of their original goal. Within a reconciliation-supporting “family-friendly” culture, the offer of work–life measures should project a clearer and more reliable signal to employees and have a stronger impact on their working lives. Thus, we assume the following:
Hypothesis 3 (H3).
The negative association of flexibility measures with voluntary employee exits is stronger when the organizational culture is perceived as family-friendly compared with establishments that have a less family-friendly culture.
In contrast, flexibility measures can be implemented to primarily serve the employer’s needs for flexibility (e.g., in the deployment of employees), thus even harming employees’ reconciliation of work and personal-life demands (Abendroth and Diewald 2019; Chung and van der Horst 2020). This applies especially to establishments in which the need for reconciling work and personal-life demands hurts organizational norms—in particular, the still predominant “ideal-worker norm”, which favors workers with few personal-life demands (Acker 1990; Lott and Klenner 2018). In this case, the flexibility measures are used not to support the reconciliation of work and personal life but to serve the need for the permanent availability of a firm’s employees, and such measures even tend to exacerbate work–family conflicts (Abendroth and Reimann 2018).
Thus, the signaling effect of the offer of flexibility measures may be lessened or may even take on a negative connotation, leading to the following assumption:
Hypothesis 4 (H4).
The negative association of flexibility measures with voluntary employee exits is reduced when the organizational culture is perceived as highly demanding compared with establishments that have a less demanding culture.

2.3. Group Differences in the Addressing by Work-Life Measures

Other than providing care-centered and parenthood-related support, flexibility measures aim to support a better balance between work and personal life for all employees (Kossek and Ollier-Malaterre 2013). Nevertheless, flexibility options do not have the same relevance for all living arrangements and lifestyles, so the needs of all employee groups are not addressed by work–life measures in the same way (Chung and van der Horst 2020; Hammermann et al. 2019; Kossek and Ollier-Malaterre 2013); in addition, firms may not consider work–life measures to be a desirable resource and an inducement to retain their employees to the same degree. In Germany, a gendered distribution of work prevails. Since women—even those without children—continue to be faced with more personal-life and family demands than men (e.g., in terms of the amount of household work they perform), we assume that women have a greater need for flexibility than men and would thus benefit more from flexibility measures (Hobler et al. 2020; Trappe et al. 2015). Moreover, parents are entrusted with more care duties because they have childcare responsibilities (Winslow 2005). The combination of gender and parenthood should therefore strengthen the effects of flexibility measures, especially for mothers. Despite indications that the perception of the demands of fatherhood is changing (Abendroth and Pausch 2018; Pollmann-Schult 2008), it is mothers especially who are still predominantly entrusted with care and maintenance duties and therefore have greater demands that would benefit from reconciliation (Dechant and Blossfeld 2015; Lott 2019). Flexibility measures should play a particularly important role for these employees with considerable and ongoing reconciliation needs. For this reason, we hypothesized the following:
Hypothesis 5 (H5).
The negative association of flexibility measures with voluntary employee exits is stronger for employees with greater reconciliation needs (women and parents, especially mothers) than for other employees.

3. Data and Methods

We used the three waves from the LEEP-B3 data (Diewald et al. 2014): wave 1 (w1) = 2012/13, wave 2 (w2) = 2014/15, and wave 3 (w3) = 2018/19. The linked employer–employee data consist of a representative sample of large German establishments2 (defined as having at least 500 workers who pay social security taxes), a random sample of employees within these establishments and linked administrative data. These data are representative of employees in large establishments, although marginally employed workers (those who work fewer than 10 h a week), employees whose nationality is not German and employees without vocational training or whose educational degrees are not known are slightly underrepresented.
To analyze actual voluntary employee exits, we linked work–life measures in the earlier wave to whether the employee exited the firm before the following wave3. This resulted in using the independent variables from wave 1 and the dependent variable from wave 2, as well as the independent variables from wave 2 and the dependent variable from wave 3.4
After approximately 30% of the original sample was dropped5, the working sample contained 5452 employees at 127 establishments, including 369 employees (w1 to w2 = 182; w2 to w3 = 187) who reportedly left their establishments during the course of the survey, and 5083 employees who were employed at the same establishment at two time points at least.6
Dependent variable: Employees who exited their establishments between two time points were defined as leavers and were coded by a dummy variable (0 = stay; 1 = exit). Moreover, cases in which respondents left the establishments because of business shutdowns, operational reorganization or retirement and those who were fired by the employer with no chance of re-employment at that firm were excluded from the sample, since our goal was to focus on intentional exits.
Work–life measures: Employees were asked whether their establishments offered work–life measures with regard to two different measures: flexible working hours and -home-based teleworking (0 = no; 1 = yes).
Employee groups: To consider differences in the impact of work–life measures on exits between specific employee groups, we considered gender (0 = men; 1 = women), parenthood (0 = no child; 1 = at least one child) and a combination of the two (0 = father; 1 = mother).
Predictors on the organizational level: Support by the employee’s direct supervisor was measured with the item “In general, my supervisor seeks to support the employees concerning the compatibility of family life and work”. For coworkers’ support, we used the item “When possible, my coworkers help me to do my work, when I have to leave earlier or when I am late for work for personal reasons”. In both cases, the response was measured initially on a five-point Likert scale, ranging from 1 (“applies completely”) to 5 (“does not apply at all”). We then reversed the scale and transformed it into a dichotomous measurement (1 to 3 = no/low support; 4 and 5 = relatively high/high support).
The family-friendliness of the firm was measured using the question “To what extent is it true that employees who make use of family supportive measures are viewed as less committed in your establishment?”. Responses ranged from 1 (“it is not true”) to 5 (“it is true”) and were aggregated to the firm level.
Lastly, we assessed the level to which a demanding organizational culture prevailed with the item “How often do you have to answer emails or phone-calls from your boss/coworkers/clients outside your official working time?”. The possible responses on a scale of 1 to 5, respectively, were “daily”, “weekly”, “monthly”, “rarely” or “never” (categories reversed) and were also aggregated to the firm level.
Controls: Since an employee’s exit is influenced by many prerequisites regarding their work conditions and job satisfaction, we controlled for several individual and structural characteristics that have proved to be relevant in previous research. At the individual level, we considered general factors that influence the probability of voluntary employee exits based on existing research (Caillier 2013; Igbaria and Siegel 1992). For job characteristics, we included tenure in years, hourly wages (log.), supervisory responsibility (0 = no; 1 = yes), contractually agreed monthly working hours (centered) and amount of overwork (i.e., actual monthly working hours/contractually agreed monthly working hours). Moreover, we included qualifications (0 = no vocational training; 1 = low-track secondary school (Hauptschule)/intermediate-track secondary school (Realschule) with vocational training; 2 = school-leaving certificate for German university entrance (Abitur) with vocational training; 3 = university degree and self-assessed chances in the labor market (difficulty in obtaining a similar or better job)) using a five-point Likert scale from 1 (“very easy”) to 5 (“very difficult”). To control for care demands and a possible “tied stayer” effect (Büchel 2000), we used partnership status (0 = no partner; 1 = in a partnership), parenthood (0 = no children; 1 = children) and gender (0 = men; 1 = women) in the overall analyses.
Since job satisfaction and fairness of exchange in the employment relationship are influenced by work–life measures (Baruch and Nicholson 1997; Kelliher and Anderson 2010; Scandura and Lankau 1998), and since both these factors are also relevant to voluntary employee exits (Mobley 1977; Rousseau 1995), we controlled for job satisfaction (“How satisfied are you currently with your job?” (range: 0 = totally unsatisfied to 10 = totally satisfied)) and the balance of expectations and gratifications (“All in all, is there a balance between what you achieve/perform at your workplace and what you usually receive for it?” (range: 1 = totally balanced to 5 = totally unbalanced; categories reversed)).
At the level of establishments, we controlled for establishment characteristics which were found to predict differences in the firms’ family-friendliness and offer of flexibility opportunities (Bächmann et al. 2020; Gerlach et al. 2012); these were divided into the sector (0 = private sector; 1 = public sector), the branch (1 = production/energy/water/construction; 2 = retail/transport/hospitality; 3 = information/economic services; 4 = administration/education/health) and firm size (0 = 500 to 699 employees; 1 = 700 to 999 employees; 2 = 1000 to 1499 employees; 3 = 1500 or more). We controlled for the place in which the firm was located (0 = Eastern Germany; 1 = Western Germany). These regions still differ insofar as full-time female employment and fewer interruptions of employment due to motherhood are more prevalent in Eastern Germany (Pfau-Effinger and Smidt 2011). Research also demonstrates that these establishments are less family-conscious than their Western counterparts (Gerlach et al. 2012). Furthermore, even after nearly 30 years of reunification, there is significantly more public full-day-childcare available in Eastern Germany, which may lead to differences in the importance and need for establishment-sided flexibility options of employees between these regions (Mätzke 2019; Pfau-Effinger and Smidt 2011; Schober and Stahl 2014).
The first two waves of the study were collected during an initial funding period, whereas the third wave was collected in the course of a following project. This explains the time span differences of two years between waves 1 and 2 versus the four years between waves 2 and 3. We are aware of the methodological problem that there might be more unobserved changes within the longer time span from wave 2 to wave 3 as well as a higher panel attrition compared to the time span between waves 1 and 2 (for differences in the panel attrition of the waves, see Marx et al. 2020), on the one hand; on the other hand, the longer time span between the last two waves allows for more observations of quite rare life events, as is the case for employee exits. We take account of this difference by adding a dummy variable to distinguish between the two observation periods (w1 to w2 = 0; w2 to w3 = 1).
To analyze our data, we used multilevel logistic regression models with mixed effects and robust standard errors to consider the clustering of employees in the establishments and to account for variability within and across the establishments, as well as for unobserved heterogeneity. To test our moderation hypotheses, we included interaction effects and further estimated the margins of these effects to consider the current methodological debate on interaction terms in nonlinear regression models. In doing so, we followed the recommendation of Mize (2019), who stressed that the interactional effect with another variable may lead to a variation in the curve (nonlinear) of the predicted variable, which cannot be displayed in one single coefficient.

4. Results

4.1. Descriptive Statistics

Table 1 displays the means and standard deviations of the variables included in our analyses. Of the employees in the sample, 6.8% left the establishment between w1 and w2 or between w2 and w3. The gender distribution of our sample was fairly balanced, with 45.7% of the employees being women. More than two thirds of the employees had children, and about the same number (2/3) of respondents were in a partnership.
The average tenure was about 9.2 years, and hourly wages were €26.20 (gross income); 75.9% of the employees worked full-time, and more than one-third were in supervisory positions. Overall, a fairly high number of employees felt that they were highly supported by their direct supervisors with respect to the reconciliation of work and family (74.0%) and by their coworkers (78.1%). The level of job satisfaction was moderately high (7.36 on a scale of 1 to 10) and the employment relationship balanced (3.60 on a scale of 1 to 5).
About 60% of the employees completed vocational training, and almost 38.6% of the respondents had a tertiary degree. With an average of 3.55 (on a scale of 1 to 5), the respondents perceived their chances of obtaining a similar or better job in the labor market as neither promising nor gloomy.
In total, 78.7% of the employees knew about the availability of flexible working times in their establishment, and 44.9% knew about home-based teleworking. The average perceived level of family-friendliness was rather modest (3.87 on a scale of 1 to 5). Blurred boundaries between personal life and working hours affected a considerable portion of the workforce (3.77 on a scale of 1 to 5). The highest number of employees worked in administration/education/health sectors, followed by nearly one-third in the production/energy/water/construction branches; 36.2% of the employees were working in establishments in the public sector and 83.3% in establishments located in Western Germany. Finally, more than two-thirds of the sample worked in establishments with fewer than 1000 employees.

4.2. Results of Multivariate Analyses

Table 2 shows the results of the logistic regression models, with organizational mixed effects and clustered standard errors. Both flexible working times and home-based teleworking were significantly negatively related to voluntary employee exit, confirming our first hypothesis (H1) that employees were less likely to exit the firm if work–life measures were perceived to be available. This is a remarkable result, since this effect remained statistically significant even when job satisfaction and the balance of expectations and gratifications in the employment relationship were controlled for. Since we controlled for tenure, working time and parenthood, this result cannot even be ascribed to these circumstances. The significant effect of the wave may emerge from a longer risk time between w2 and w3, since the labor market chances within these years should not differ tremendously.
Table 2 also shows negative associations between voluntary employee exit and being a parent as well as being a woman. These associations replicate the lower mobility of parents (e.g., Geist and McManus 2012; Nivalainen 2004). The gender effect is contrary to results from Weisberg and Kirschenbaum (1993), who found fewer firm shifts for women. Moreover, they found that the antecedents of exits differed for men and women. By controlling for various job and personal characteristics, we took these differences into account. The reversed effects may also be due to differences in times. As Cho and Lewis (2012) mention, with the growing participation of women in the workplace, there may be a shift in gender mobility within the labor market, as women are more attached to the labor market and are less often perceived to be the only caregiver. Working women may also no longer be as tied to their partner’s mobility (as “tied movers”) (Geist and McManus 2012; Nivalainen 2004).
To affirm the appropriateness of measuring perceived work–life measures, we re-ran these models basing the offer of measures on evaluations by human resource managers (results not included here but can be obtained upon request). For this alternative operationalization, we found no significant effects; in other words, the purely formal and possibly selective implementation of these measures does not work as a strategy for keeping employees in the organization.
We also tested the actual use of both flexibility measures (results not included here but can be obtained upon request). Again, other than for the perceived availability, we could not find any significant association between the use of these measures and voluntary employee exit. This may seem surprising, as the actual use of these measures may represent an even more comprehensible benefit for employees. However, these results are in line with research from Butts et al. (2013), who showed that the relationship between the perceived availability of a work–family support policy and work attitudes (such as job satisfaction or the intention to look for a different employer) is stronger than the relationship between actual policy use and these attitudes. They argue that the perception that family supportiveness is available to all employees without possible backlashes for those employees who make use of these policies has a symbolic effect.

4.2.1. The Organizational Environment

Table 3 presents the results of our analyses of interaction effects for coworker and supervisor support and for organizational culture (H2–H4). We expected that a high level of support for the reconciliation of work and family demands by an employee’s direct supervisor and coworkers or a family-friendly organizational culture would negatively moderate the association between the perceived availability of work–life measures and voluntary employee exits. Interestingly, we found that high support by supervisors or coworkers had no effect; thus, H2 must be rejected. Nevertheless, we did find a negative main effect of a family-friendly culture on exits, but no moderation of the association between work–life measures and exits. Thus, our hypothesis supposing a negative moderating effect of a reconciliation-supporting environment (H3) must also be rejected. However, we found a moderating effect for a demanding organizational environment (H4): the frequency with which employees were contacted outside normal working hours had a negative effect on whether the perceived availability of home-based teleworking prevented voluntary employee exits (see Figure 1 for the predictive marginal effects). This means that when the gain in flexibility through home-based teleworking is counteracted by blurred boundaries, it is no longer seen as a benefit in terms of preventing between-firm mobility. Furthermore, in establishments with a less demanding culture, the perceived availability of home-based teleworking was significantly associated with fewer exits, whereas in establishments with higher demands of permanent availability, it did not make any difference whether or not home-based teleworking was perceived to be available.

4.2.2. Group Differences

Finally, we investigated whether the association between the perceived availability of work–life measures and voluntary employee exit varied between groups within the workforce with an assumed higher or lower need for flexibility because of care duties, especially regarding gender and parenthood (H5).
The results displayed in Table 4 showed no group differences in the association between flexible working times and exits. For home-based teleworking, however, there were statistically (though only slightly) significant interaction effects with gender and the gender of parents (p < 0.10).
In analyzing the margins of the interaction between gender and work–life measures (see Figure 2), we found that, contrary to our hypothesis, this effect is driven by differences in men’s—not women’s—probability of exiting the firm. Men who perceived the availability of home-based teleworking were less likely to exit the firm than were men who did not. For women, this was not the case. The same pattern also applied to fathers versus mothers (see Figure 3); obviously, it is not their greater responsibility for care duties that makes home-based teleworking attractive—at least not predominantly. For both non-parents and parents, we found no significant interaction with work–life measures, with parents having a lower probability of exiting the firm in any case.

5. Discussion and Conclusions

With the increase in workforce diversity and more individualized life courses, balancing work and personal life has become a fundamental issue for achieving overall job quality (Eurofond 2017; Kelliher and Anderson 2008). Work–life measures provided by firms have been investigated and discussed intensively as an opportunity for employers to support employees’ reconciliation needs (Kossek and Distelberg 2009; Moen 2015). Previous research has indicated that the offer of such work–life measures is positively related to job satisfaction and commitment (Baltes et al. 1999; Baruch 2000; Grover and Crooker 1995; Kelliher and Anderson 2008, 2010), to fewer work–family conflicts (Kelly et al. 2011) and to a decrease in exit intentions (Batt and Valcour 2003; Moen et al. 2017). Much less is known about whether the availability of flexibility measures can also help to prevent actual voluntary employee exits, since such exits are preconditioned by a number of push and pull factors. Therefore, despite the impact of work–life measures on job quality and organizational commitment, an isolated effect of such measures is far from evident.
In this paper, we focused on two flexibility measures that address the whole workforce, rather than only those employees with care duties within the family context. Based on three waves of linked employer–employee data representative of large establishments in Germany, the results of our study add to previous research on the impact of work–life measures by showing that the availability of flexibility measures can indeed contribute to lowering the risk of employee exits, even when controlling for several important job-quality characteristics.
This conclusion is not trivial. Existing research has demonstrated extensively that a lack of career or wage advancements (Igbaria and Siegel 1992), a low level of job satisfaction and perceived unfairness in the employment relationship (Mobley 1977; Rousseau 1995) are important predictors of exit intentions as well as of actual exits. These determinants are much more comprehensive indicators of overall job quality than is the possible effect of specific work–life measures. Moreover, leaving an employer can also be explained by many other factors, such as personal life events or a continuous boom (Hom et al. 2017; Mobley 1977). With this in mind, it is all the more striking that we still found significant effects of the perceived availability of work–life measures when job satisfaction, overall balance between contributions and gratifications and other relevant characteristics were controlled for.
By focusing on the perceived availability of work–life measures, we avoided the bias inherent in an establishment’s merely formal, hypocritical implementation of these measures (Brunsson 2003). If employees are not aware of the measures offered, or if these measures are implemented in such a way that they are not perceived as supportive offers, it is highly unlikely that such offers will have a positive impact on employee outcomes, especially in terms of far-reaching decisions such as exiting a firm. The formal offer of work–life measures as reported at the employer level (by HRM) had no effect on exits at all, strengthening the assumption that work–life measures are only perceived as inducements to stay when they are communicated to the employees and executed in a way that is understood by the whole workforce.
In line with current research about the role of the organizational environment for work-related outcomes (e.g., Tomaskovic-Devey and Avent-Holt 2019), the relevance of how work–life measures are implemented and mediated in practice is also emphasized by our results with regard to the moderating relevance of organizational culture. If a highly demanding workplace culture representing the ideal-worker norm prevails in an establishment, the probability of exit due to the availability of home-based teleworking is lower when compared with a work environment that is not highly demanding. Even though the availability of this work–life measure is generally negatively associated with exits, a demanding culture still limits the potential use of this flexibility option because it may be connected to other disadvantages, such as less appreciation from supervisors or coworkers or fewer career opportunities, as compared with those employees who do not use home-based teleworking (Lott and Abendroth 2020; Lott and Klenner 2018). The lack of an effect of a family-friendly culture and work–life support by supervisors and coworkers is remarkable but less counterintuitive on further thought: first, a part of the positive effect of a family-friendly environment and supervisor support as gatekeepers for the internal execution of HRM practices may already be represented by the operationalization of our dependent variable. The perceived availability already comprises a successful communication of the offer, and it can be assumed that in the majority of cases it is perceived as actually available to the respondents, which may include the idea that supervisors would not discourage the use of the measures (Van der Voet et al. 2015; Den Dulk 2001). Second, coworkers’ support can certainly alleviate everyday struggles in managing competing demands in work and family demands (Hayton et al. 2012; Pisarski et al. 2008); however, it cannot replace the flexibilities enabled by managerial and supervisory consent and support. Its impact has repeatedly shown to be rather limited (Abendroth and Reimann 2018; Blair-Loy and Wharton 2002; Hämmig 2017).
Against the background of the ongoing discussion on the need for flexibility, on the part of mothers in particular (Dechant and Blossfeld 2015; Lott 2019), our results demonstrate instead that the challenge to balancing work and personal life is an issue for all employees, not only for women and parents. Organizational measures that support individual endeavors in meeting reconciliation needs are not only a predictor of job satisfaction and work–family conflicts; these measures have also become important in such a way that they can even help to prevent voluntary employee exits. Men in particular seem to value the availability of home-based teleworking. However, we are not able to control for motives other than the assumed needs for reconciliation and specific care duties of parents. Variations in the reasons for making use of flexibility measures might explain the differences in their relevance for women and men (Chung and van der Horst 2020; Lott 2019). For women, these measures serve to meet family demands, whereas men often use these measures to expand their work hours and availability. These motives would also explain the lack of an effect among women if the offer of home-based teleworking reflects a persistent flexibility stigma—one which may be perceived by women more than by men (Chung 2020; Lott and Abendroth 2020). Nevertheless, since we did not look at the actual use of work–life measures but rather their perceived availability, the effect of variations in motives should (to some extent at least) be subsumed by the focus on work–life measures as a signaling effect of an employer’s support of flexibility needs in general, which might be relevant for all employees independent of their individual specific motives.
The practical implementations of these results are almost self-evident. Given a truly supportive offer of such measures, they represent a benefit that retains employees in the entire workforce and not only those who have care duties. Since we focused on voluntary employee exits rather than exits in general, these results are even more valuable for establishments, providing implications to retain employees that they most probably do not want to let go. Moreover, it is notable that the benefit of these measures goes beyond actual use, since, under these preconditions, it sends a powerful signal to the whole workforce that an organization cares about their flexibility needs. Clearly, a firm’s sketchy, hypocritical implementation alone does not have such an impact. In times of the growing importance of flexibility options, our results highlight that establishments should aim for a sufficient execution rather than only the formal implementation of flexibility policies and break-up the traditional “ideal-worker” norm if they really want to make these measures work as an inducement for the whole workforce.

6. Limitations and Further Research

Although our contribution to this field profited from the rich information we drew from a large linked-employer–employee panel dataset, some limitations to this study have to be considered. First, according to the study design, employees who left the establishment in one wave were no longer part of the sample in the following waves. Thus, we did not have extensive information on an employee’s new job or the availability of work–life measures there, and we were not able to conduct a panel analysis. By using information on work–life measures from one wave and the information on exits from the following wave, we approximated a quasi-longitudinal design. Nevertheless, causal inferences must be drawn carefully.
Second, to gain reliable case numbers, we pooled exits from the second and third waves. This procedure included possible biases, since employees who were employed in the same establishment in all three waves were systematically more similar than other employees (e.g., in having higher earnings as a result of seniority effects), and there might have been a considerable change over time in the professional and personal situation of employees who participated in all three waves (e.g., having children, within-firm career advancements or changes in working times) that affected their needs; there might also be a bias due to the different time spans between the waves. To some degree, we considered this bias by controlling for the wave and various individual characteristics in all the regression models. We do not claim the associations to be causal. In addition, we ran separate models for single waves as well as for different employee groups, which led to conclusions that were fairly the same as in the comprehensive models. However, we found fewer significant effects in the combination of the first and the second waves compared with the second and third waves. This might be explained by the longer time span between w2 and w3 (i.e., more time to leave the establishment) and by panel selection.
Third, although we covered the entire industrial structure in our sample, the data included only large establishments with more than 500 employees. The relevance of work–life measures on voluntary employee exit might be different in small and medium-sized establishments (Frodermann et al. 2018). Further empirical research will be needed to include all establishment sizes in order to arrive at any conclusions about the generalizability of the results.
In addition to addressing these statistical issues, future research should investigate alternatives to voluntary employee exits, such as internal changes within a division (De Lange et al. 2008). Additionally, in our study, we focused on flexibility measures that are potential inducements for all employees to remain in their jobs. Taking a closer look at other work–life measures, such as those for parental leave or employer-provided childcare, would be fruitful for clarifying whether measures that specifically address parents might be more helpful in supporting this group of employees.
Data collection took place before the Covid-19 pandemic, which led to a massive increase of home-based teleworking within a very short time period. Urged by infection control efforts, employers were forced to make working from home possible wherever it was imaginable. Experiences during this time may have removed many blockades against home-based teleworking; e.g., that it is less productive and not manageable by employees, coworkers and supervisors. Moreover, employers are increasingly under political pressure to offer home-based teleworking. Although not all employees benefitted from this enforced implementation and struggled with less contact and blurred boundaries between work and private life, many employees said that they hope to retain the increased flexibility also after the pandemic (Alipour et al. 2020; Grunau et al. 2020). It is very likely that employers will have to respond more to needs for home-based teleworking and other flexible workplace arrangements than before. We doubt that the experiences with home-based teleworking during these extraordinary times can simply be inferred to more normalized circumstances. It still requires research to explore what implementations will actually work in heterogeneous workplaces. In future, the offer of more flexibility in terms of time and place will surely play an increased role in workplace commitment and staff retention compared to the pre-pandemic period. Possible shifts in the importance of these measures for employees and the attraction of employers offering these measures, differences between specific groups of employees and the significance of the organizational environment in the execution should therefore be analyzed not only during the ongoing pandemic situation, but also in the context of its long-term consequences and the practicability of solutions in the long term.

Author Contributions

Conceptualization, C.K.M., M.R. and M.D.; Methodology, C.K.M., M.R. and M.D.; Software, C.K.M.; Validation, C.K.M., M.R. and M.D.; Formal Analysis, C.K.M.; Writing—Original Draft Preparation, C.K.M., M.R. and M.D; Writing—Review & Editing, C.K.M., M.R. and M.D.; Visualization, C.K.M.; Supervision, M.D.; Project administration, M.D.; Funding Acquisition, M.D. All authors have read and agreed to the published version of the manuscript.


This research was funded by the German Research Foundation (DFG) [grant no.: 373090005]. We acknowledge support for the publication costs by the Open Access Publication Fund of Bielefeld University.

Informed Consent Statement

Informed consent was obtained from all subjects involved in the study.

Data Availability Statement

The data presented in this study are available on request from the corresponding author. The data are not publicly available due to legal data protection regulations.

Conflicts of Interest

The authors declare no conflict of interest.


  1. Abendroth, Anja-Kristin, and Martin Diewald. 2019. Auswirkungen von Teleheimarbeit auf geschlechtsspezifische Einkommensungleichheiten in Arbeitsorganisationen. Die Bedeutung unterschiedlicher Umsetzungsformen. Kölner Zeitschrift für Soziologie und Sozialpsychologie 71: 81–109. [Google Scholar] [CrossRef]
  2. Abendroth, Anja-Kristin, and Stephanie Pausch. 2018. German fathers and their preferences for shorter working hours for family reasons. Community, Work & Family 21: 463–81. [Google Scholar] [CrossRef]
  3. Abendroth, Anja-Kristin, and Mareike Reimann. 2018. Telework and work-family conflict across workplaces. Investigating the implications of work-family-supportive and high-demand workplace cultures. In Contemporary Perspectives in Family Research. The Work-Family Interface: Spillover, Complications, and Challenges. Edited by Sampson Lee Blair and Josip Obradovic. Bingley: Emerald Published, pp. 323–48. ISBN 978-1-78769-112-4. [Google Scholar]
  4. Acker, Joan. 1990. Hierarchies, jobs, bodies: A theory of gendered organizations. Gender & Society 4: 139–58. [Google Scholar] [CrossRef][Green Version]
  5. Alipour, Jean-Victor, Oliver Falck, and Simone Schüller. 2020. Homeoffice während der Pandemie und die Implikationen für eine Zeit nach der Krise. IFO Schnelldienst 73: 30–36. Available online: (accessed on 30 December 2020).
  6. Allen, Tammy D. 2001. Family-Supportive Work Environments: The Role of Organizational Perceptions. Journal of Vocational Behavior 58: 414–35. [Google Scholar] [CrossRef]
  7. Allen, David G., Lynn M. Shore, and Rodger W. Griffeth. 2003. The Role of Perceived Organizational Support and Supportive Human Resource Practices in the Turnover Process. Journal of Management 29: 99–118. [Google Scholar] [CrossRef]
  8. Almer, Elizabeth D., Jeffrey R. Cohen, and Louise E. Single. 2004. Is it the Kids or the Schedule? The Incremental Effect of Families and Flexible Scheduling on Perceived Career Success. Journal of Business Ethics 54: 51–65. [Google Scholar] [CrossRef]
  9. Bächmann, Ann-Christin, Corinna Frodermann, Daniela Grunow, Marina Hagen, and Dana Müller. 2020. Family-Friendly Organizational Arrangements—Anything but “a Fuss” (Over Nothing)! Nürnberg: IAB-Forum 20th of February 2020. Available online: (accessed on 26 December 2020).
  10. Baltes, Boris B., Thomas E. Briggs, J. W. Huff, Julie A. Wright, and George A. Neuman. 1999. Flexible and Compressed Workweek Schedules: A Meta-Analysis of Their Effects on Work-Related Criteria. Journal of Applied Psychology 84: 496–513. [Google Scholar] [CrossRef]
  11. Baruch, Yehuda. 2000. Teleworking: benefits and pitfalls as perceived by professionals and managers. New Technology Work and Employment 15: 34–49. [Google Scholar] [CrossRef]
  12. Baruch, Yehuda, and Nigel Nicholson. 1997. Home, Sweet Work: Requirements for Effective Home Working. Journal of General Management 23: 15–30. [Google Scholar] [CrossRef]
  13. Batt, Rosemary, and P. Monique Valcour. 2003. Human resource practices as predictors of work-family outcomes and employee turnover. Industrial Relations 42: 189–220. [Google Scholar] [CrossRef]
  14. Blair-Loy, Mary, and Amy S. Wharton. 2002. Employees’ Use of Work-Family Policies and the Workplace Social Context. Social Forces 80: 813–45. [Google Scholar] [CrossRef]
  15. Brunsson, Nils. 2003. The Organization of Hypocrisy: Talk, Decisions and Actions in Organizations, 2nd ed. Copenhagen: Copenhagen Business School Press, ISBN 978-8763001069. [Google Scholar]
  16. Büchel, Felix. 2000. Tied movers, tied stayers: The higher risk of overeducation among married women in West Germany. In Gender and the Labour Market. Edited by Siv Gustafsson and Daniele Meulders. London: Macmillan Publishers Ltd., pp. 133–46. ISBN 978-0-333-80442-1. [Google Scholar]
  17. Butts, Marcus M., Wendy J. Casper, and Tae Seok Yang. 2013. How important are work-family support policies? A meta-analytic investigation of their effects on employee outcomes. Journal of Applied Psychology 98: 1–25. [Google Scholar] [CrossRef]
  18. Caillier, James Gerard. 2013. Are Teleworkers Less Likely to Report Leave Intentions in the United States Federal Government Than Non-teleworkers Are? American Review of Public Administration 43: 72–88. [Google Scholar] [CrossRef]
  19. Cho, Yoon Jik, and Gregory B. Lewis. 2012. Turnover Intention and Turnover Behavior: Implications for Retaining Federal Employees. Review of Public Personnel Administration 32: 4–23. [Google Scholar] [CrossRef]
  20. Chung, Heejung. 2020. Gender, Flexibility Stigma and the Perceived Negative Consequences of Flexible Working in the UK. Social Indicators Research 151: 521–45. [Google Scholar] [CrossRef][Green Version]
  21. Chung, Heejung, and Mariska van der Horst. 2020. Flexible Working and Unpaid Overtime in the UK: The Role of Gender, Parental and Occupational Status. Social Indicators Research 151: 495–520. [Google Scholar] [CrossRef][Green Version]
  22. Chung, Heejung, and Tanja van der Lippe. 2020. Flexible Working, Work-Life Balance, and Gender Equality: Introduction. Social Indicators Research 151: 365–81. [Google Scholar] [CrossRef][Green Version]
  23. Cook, Rose, Margaret O’Brien, Sara Connolly, Matthew Aldrich, and Svetlana Speight. 2020. Fathers’ Perceptions of the Availability of Flexible Working Arrangements: Evidence from the UK. Work, Employment and Society. [Google Scholar] [CrossRef]
  24. Coyle-Shapiro, Jacqueline A.-M., and Neil Conway. 2004. The employment relationship through the lens of social exchange. In The Employment Relationship: Examining Psychological and Contextual Perspectives. Edited by Jacqueline A.-M. Coyle-Shapiro, Lynn M. Shore, Susan M. Taylor and Lois Tetrick. Oxford: Oxford University Press, pp. 5–28. ISBN 978-0-19926-913-6. [Google Scholar]
  25. Coyle-Shapiro, Jacqueline A.-M., and Lynn M. Shore. 2007. The Employee–Organization Relationship: Where Do We Go from Here? Human Resource Management Review 17: 166–79. [Google Scholar] [CrossRef][Green Version]
  26. De Lange, Annet H., Hans De Witte, and Guy Notelaers. 2008. Should I stay or should I go? Examining longitudinal relations among job resources and work engagement for stayers versus movers. Work & Stress 22: 201–23. [Google Scholar] [CrossRef]
  27. De Menezes, Lilian M., and Clare Kelliher. 2017. Flexible Working, Individual Performance, and Employee Attitudes: Comparing Formal and Informal Arrangements. Human Resource Management 56: 1051–70. [Google Scholar] [CrossRef][Green Version]
  28. Dechant, Anna, and Hans-Peter Blossfeld. 2015. Changes in the division of labor within highly educated German couples when the first child is born. Zeitschrift Für Familienforschung 27: 373–96. [Google Scholar] [CrossRef][Green Version]
  29. Den Dulk, Laura. 2001. Work-Family Arrangements in Organisations: A Cross-National Study in the Netherlands, Italy, the United Kingdom and Sweden. Amsterdam: Rozenberg, ISBN 90-51705-42-5. [Google Scholar]
  30. Den Dulk, Laura, and Judith De Ruijter. 2008. Managing work-life policies: Disruption versus dependency arguments. Explaining managerial attitudes towards employee utilization of work-life policies. The International Journal of Human Resource Management 19: 1222–36. [Google Scholar] [CrossRef][Green Version]
  31. Denison, Daniel R. 1996. What is the Difference between Organizational Culture and Organizational Climate? A Native’s Point of View on a Decade of Paradigm Wars. The Academy of Management Review 21: 619–54. [Google Scholar] [CrossRef]
  32. Diewald, Martin, Reinhard Schunck, Anja Abendroth, Silvia Maja Melzer, Stephanie Pausch, Mareike Reimann, Björn Adernach, and Peter Jacobebbinghaus. 2014. The SFB-B3 Linked Employer-Employee Panel Survey (LEEP-B3). Schmollers Jahrbuch 134: 379–89. [Google Scholar] [CrossRef]
  33. Eurofond. 2017. Sixth European Working Conditions Survey—Overview Report (2017 Update). Luxembourg: Publications Office of the European Union, Available online: (accessed on 7 December 2020).
  34. Frodermann, Corinna, Ann-Christin Bächmann, Marina Hagen, Daniela Grunow, and Dana Müller. 2018. Betriebliche Angebote zur Vereinbarkeit von Familie und Beruf: Mütter kehren schneller zu familienfreundlichen Arbeitgebern zurück. In IAB-Kurzbericht 18. Nürnberg: Institut für Arbeitsmarkt- und Berufsforschung (IAB), Available online: (accessed on 11 December 2020).
  35. Gajendran, Ravi S., and David A. Harrison. 2007. The Good, the Bad, and the Unknown About Telecommuting: Meta-Analysis of Psychological Mediators and Individual Consequences. Journal of Applied Psychology 92: 1524–41. [Google Scholar] [CrossRef][Green Version]
  36. Galinsky, Ellen, and Peter J. Stein. 1990. The impact of human resource policies on employees: Balancing work/family life. Journal of Family Issues 11: 368–83. [Google Scholar] [CrossRef]
  37. Geist, Claudia, and Patricia A. McManus. 2012. Different Reasons, Different Results: Implications of Migration by Gender and Family Status. Demography 49: 197–217. [Google Scholar] [CrossRef]
  38. Gerlach, Irene, Helmut Schneider, Helen Wieners, David Juncke, and Jana Heinze. 2012. Status quo des betrieblichen Familienbewusstseins deutscher Unternehmen—Ergebnisse einer empirischen Erhebung. In Betriebliche Familienpolitik. Edited by Gerlach Irene and Helmut Schneider. Wiesbaden: Springer, pp. 139–52. [Google Scholar] [CrossRef]
  39. Grover, Steven L., and Karen J. Crooker. 1995. Who appreciates family-responsive human resource policies: The impact of family-friendly policies on the organizational attachment of parents and non-parents. Personnel Psychology 48: 271–88. [Google Scholar] [CrossRef]
  40. Grunau, Philipp, Susanne Steffes, and Stefanie Wolter. 2020. Homeoffice in Zeiten Von Corona: In vielen Berufen Gibt es Bislang Ungenutzte Potenziale. IAB-Forum 25. März 2020. Available online: (accessed on 30 December 2020).
  41. Guest, David E. 2016. Trust and the Role of the Psychological Contract in Contemporary Employment Relations. In Building Trust and Constructive Conflict Management in Organizations. Edited by Patricia Elgoibar, Martin Euwema and Lourdes Munduate. Industrial Relations & Conflict Management Series; Cham: Springer International Publishing, pp. 137–49. ISBN 978-3-319-31475-4. [Google Scholar]
  42. Hammer, Leslie B., Ellen Ernst Kossek, Nanette L. Yragui, Todd E. Bodner, and Ginger C. Hanson. 2009. Development and Validation of a Multidimensional Measure of Family Supportive Supervisor Behaviors (FSSB). Journal of Management 35: 837–56. [Google Scholar] [CrossRef] [PubMed][Green Version]
  43. Hammermann, Andrea, Schmidt Jörg, and Oliver Stettes. 2019. Unternehmensmonitor Familienfreundlichkeit 2019. Berlin: BMFSJF, Available online: (accessed on 7 December 2020).
  44. Hämmig, Oliver. 2017. Health and well-being at work: The key role of supervisor support. SSM—Population Health 3: 393–402. [Google Scholar] [CrossRef] [PubMed]
  45. Hayton, James C., Carnabuci Gianluca, and Robert Eisenberger. 2012. With a little help from my colleagues: A social embeddedness approach to perceived organizational support. Journal of Organizational Behavior 33: 235–49. [Google Scholar] [CrossRef]
  46. Hobler, Dietmar, Lott Yvonne, Pfahl Svenja, and Karin Schulze Buschoff. 2020. Stand der Gleichstellung von Frauen und Männern in Deutschland. WSI Report 56. Available online: (accessed on 7 December 2020).
  47. Hom, Peter W., Thomas Lee, Jason D. Shaw, and John P. Hausknecht. 2017. One Hundred Years of Employee Turnover Theory and Research. Journal of Applied Psychology 102: 530–45. [Google Scholar] [CrossRef] [PubMed]
  48. Igbaria, Magid, and Sidney R. Siegel. 1992. The reasons for turnover of information systems personnel. Information & Management 23: 321–30. [Google Scholar] [CrossRef]
  49. Kalidass, Anneswary, and Arsiah Bahron. 2015. The Relationship between Perceived Supervisor Support, Perceived Organizational Support, Organizational Commitment and Employee Turnover Intention. International Journal of Business Administration 6: 82–89. [Google Scholar] [CrossRef][Green Version]
  50. Kelliher, Clare, and Deirdre Anderson. 2008. For better or for worse? An analysis of how flexible working practices influence employees’ perceptions of job quality. The International Journal of Human Resource Management 19: 419–31. [Google Scholar] [CrossRef][Green Version]
  51. Kelliher, Clare, and Deirdre Anderson. 2010. Doing more with less? Flexible working practices and the intensification of work. Human Relations 63: 83–106. [Google Scholar] [CrossRef]
  52. Kelly, Erin L., Moen Phyllis, and Eric Tranby. 2011. Changing Workplaces to Reduce Work-Family Conflict: Schedule Control in a White-Collar Organization. American Sociological Review 76: 265–90. [Google Scholar] [CrossRef]
  53. Konrad, Alison M., and Yang Yang. 2012. Is using work-life interface benefits a career-limiting move? Journal of Organization Behavior 31: 367–94. [Google Scholar] [CrossRef]
  54. Kossek, Ellen Ernst, and Brian Distelberg. 2009. Work and Family Employment Policy for a Transformed Work Force: Trends and Themes. In Work-Life Policies. Edited by Ann C. Crouter and Alan Booth. Washington, DC: Urban Institute Press, pp. 1–51. ISBN 978-0877667483. [Google Scholar]
  55. Kossek, Ellen Ernst, and Brenda A. Lautsch. 2018. Work-Life Flexibility for Whom? Occupational Status and Work-Life Inequality in Upper, Middle and Lower Level Jobs. Academy of Management Annals 12: 5–36. [Google Scholar] [CrossRef]
  56. Kossek, Ellen Ernst, and Ariane Ollier-Malaterre. 2013. Work-life policies: Linking national contexts, organizational practice and people for multi-level change. In Expanding the Boundaries of Work-Family Research. Edited by Steven Poelmans, Jeffrey Greenhaus and Mireia Las Heras Maestro. Hampshire: Palgrave Macmillan, pp. 3–31. ISBN 978-1-137-00600-4. [Google Scholar]
  57. Kossek, Ellen Ernst, Suzan Lewis, and Leslie B. Hammer. 2010. Work-life initiatives and organizational change: Overcoming mixed messages to move from the margin to the mainstream. Human Relations 63: 3–19. [Google Scholar] [CrossRef][Green Version]
  58. Lambert, Alysa D., Janet H. Marler, and Hal G. Gueutal. 2008. Individual differences: Factors affecting employee utilization of flexible work arrangements. Journal of Vocational Behavior 73: 107–17. [Google Scholar] [CrossRef]
  59. Lee, Thomas W., and Terence R. Mitchell. 1994. An Alternative Approach: The Unfolding Model of Voluntary Employee Turnover. Academy of Management Review 19: 51–89. [Google Scholar] [CrossRef]
  60. Lewis, Suzan, and Janet Smithson. 2001. Sense of Entitlement to Support for the Reconciliation of Employment and Family Life. Human Relations 54: 1455–81. [Google Scholar] [CrossRef][Green Version]
  61. Lott, Yvonne. 2015. Working-time flexibility and autonomy: A European perspective on time adequacy. European Journal of Industrial Relations 21: 259–74. [Google Scholar] [CrossRef]
  62. Lott, Yvonne. 2019. Weniger Arbeit, mehr Freizeit? Wofür Mütter und Väter Flexible Arbeitsarrangements Nutzen. WSI Report 47. Available online: (accessed on 7 December 2020).
  63. Lott, Yvonne, and Anja-Kristin Abendroth. 2020. The non-use of telework in an ideal worker culture: Why women perceive more cultural barriers. Community, Work & Family. [Google Scholar] [CrossRef]
  64. Lott, Yvonne, and Christina Klenner. 2018. Are the ideal worker and ideal parent norms about to change? The acceptance of part-time and parental leave at German workplaces. Community, Work & Family 21: 564–80. [Google Scholar] [CrossRef]
  65. Maertz, Carl P., Jr., Rodger W. Griffeth, Nathanael S. Campbell, and David G. Allen. 2007. The effects of perceived organizational support and perceived supervisor support on employee turnover. Journal of Organizational Behavior 28: 1059–75. [Google Scholar] [CrossRef]
  66. Marx, Charlotte, Anja-Kristin Abendroth, Ann-Christin Bächmann, Diewald Martin, Silvia Maja LauraLükemann, Eileen Peters Melzer, and Mareike Reimann. 2020. Employee and Partner Surveys Wave 3 of the Linked-Employer-Employee-Panel (LEEP-B3) Project (DFG-373090005): Organizational Inequalities and Interdependencies between Capabilities in Work and Personal Life: A Study of Employees in Different Work Organizations. Technical Report. Bielefeld: Universität, Fakultät für Soziologie. Available online: (accessed on 7 December 2020).
  67. Mätzke, Margitta. 2019. Comparative Perspectives on Childcare Expansion in Germany: Explaining the Persistent East-West Divide. Journal of Comparative Policy Analysis: Research and Practice 21: 47–64. [Google Scholar] [CrossRef]
  68. McFarlane Shore, Lynn, Lucy A. Newton, and George C. Thornton II. 1990. Job and organizational attitudes in relation to employee behavioral intentions. Journal of Organizational Behavior 11: 57–67. [Google Scholar] [CrossRef]
  69. Mize, Trenton D. 2019. Best practices for Estimating, Interpreting, and Presenting Nonlinear Interaction Effects. Sociological Science 6: 81–117. [Google Scholar] [CrossRef][Green Version]
  70. Mobley, William H. 1977. Intermediate linkages in the relationship between job satisfaction and employee turnover. Journal of Applied Psychology 62: 237–40. [Google Scholar] [CrossRef]
  71. Moen, Phyllis. 2015. An Institutional/Organizational Turn: Getting to Work-Life Quality and Gender Equality. Work & Occupations 42: 174–82. [Google Scholar] [CrossRef]
  72. Moen, Phyllis, Erin L. Kelly, and Rachelle Hill. 2011. Does Enhancing Work-Time Control and Flexibility Reduce Turnover? A Naturally Occurring Experiment. Social Problems 58: 69–98. [Google Scholar] [CrossRef] [PubMed][Green Version]
  73. Moen, Phyllis, Erin L. Kelly, Shi Rong Lee, J. Michael Oakes, Wen Fan, Jeremy Bray, David Almeida, Leslie Hammer, David Hurtado, and Orfeu Buxton. 2017. Can a flexibility/support initiative reduce turnover intentions and exits? Results from the Work, Family, and Health Network. Social Problems 64: 53–85. [Google Scholar] [CrossRef][Green Version]
  74. Nivalainen, Satu. 2004. Determinants of Family Migration: Short Moves vs. Long Moves. Journal of Population Economics, 17: 157–75. [Google Scholar] [CrossRef]
  75. Pfau-Effinger, Birgit, and Maike Smidt. 2011. Differences in women’s employment patterns and family policies: eastern and western Germany. Community, Work & Family 14: 217–32. [Google Scholar] [CrossRef]
  76. Philippe, Thomas W., and Jerry W. Koehler. 2005. A Factor Analytical Study of Perceived Organizational Hypocrisy. Advanced Management Journal 70: 13–20. [Google Scholar]
  77. Pisarski, Anne, Sandra A. Lawrence, Philip Bohle, and Christine Brook. 2008. Organizational influences on the work -life conflict and health of shiftworkers. Applied Ergonomics 39: 580–88. [Google Scholar] [CrossRef]
  78. Pollmann-Schult, Matthias. 2008. Familiengründung und gewünschter Erwerbsumfang von Männern—Eine Längsschnittanalyse für die alten Bundesländer. Zeitschrift Für Soziologie 37: 498–515. [Google Scholar] [CrossRef]
  79. Rhee, Min-Kyoung, Soo Kyung Park, and Chung-Kwon Lee. 2020. Pathways from workplace flexibility to turnover intention: Role of work-family conflict, family-work conflict, and job satisfaction. International Journal of Social Welfare 29: 51–61. [Google Scholar] [CrossRef]
  80. Rousseau, Denise M. 1995. Psychological Contracts in Organizations: Understanding Written and Unwritten Agreements. California: SAGE Publications, Inc., ISBN 978-0803971059. [Google Scholar]
  81. Scandura, Terri A., and Melenie J. Lankau. 1998. Relationships of gender, family responsibility and flexible work hours to organizational commitment and job satisfaction. Journal of Organizational Behavior 18: 377–91. [Google Scholar] [CrossRef]
  82. Scheidler, Sabrina, Laura Marie Edinger-Schons, Jelena Spanjol, and Jan Wieseke. 2019. Scrooge posing as Mother Theresa: How hypocritical social responsibility strategies hurt employees and firms. Journal of Business Ethics 157: 339–58. [Google Scholar] [CrossRef]
  83. Schober, Pia S., and Juliane F. Stahl. 2014. Childcare Trends in Germany—Increasing Socio-Economic Disparities in East and West. DIW Economic Bulletin 11. Berlin: German Institute for Economic Research (DIW), Available online: (accessed on 26 December 2020).
  84. Starrels, Marjorie E. 1992. The evolution of workplace family policy research. Journal of Family Issues 13: 259–78. [Google Scholar] [CrossRef]
  85. Thompson, Cynthia Α., Laura L. Beauvais, and Karen S. Lyness. 1999. When Work-Family Benefits Are Not Enough: The Influence of Work-Family Culture on Benefit Utilization, Organizational Attachment, and Work-Family Conflict. Journal of Vocational Behavior 54: 392–415. [Google Scholar] [CrossRef]
  86. Tomaskovic-Devey, Donald, and Dustin Avent-Holt. 2019. Relational Inequalities: An Organizational Approach. Oxford: Oxford University Press. [Google Scholar] [CrossRef]
  87. Trappe, Heike, Matthias Pollmann-Schult, and Christian Schmitt. 2015. The Rise and Decline of the Male Breadwinner Model: Institutional Underpinnings and Future Expectations. European Sociological Review 31: 230–42. [Google Scholar] [CrossRef]
  88. Van der Lippe, Tanja, Leonie van Breeschoten, and Margriet van Hek. 2019. Organizational Work-Life Policies and the Gender Wage Gap in European Workplaces. Work and Occupations 46: 111–48. [Google Scholar] [CrossRef]
  89. Van der Voet, Joris, Ben S. Kuipers, and Sandra Groeneveld. 2015. Implementing Change in Public Organizations: The relationship between leadership and affective commitment to change in a public sector context. Public Management Review 18: 842–65. [Google Scholar] [CrossRef]
  90. Vandenberg, Robert J., and Jodi Barnes Nelson. 1999. Disaggregating the motives underlying turnover intentions: When do intentions predict turnover behavior? Human Relations 52: 1313–36. [Google Scholar] [CrossRef]
  91. Weisberg, Jacob, and Alan Kirschenbaum. 1993. Gender and Turnover: A Re-examination of the Impact of Sex on Intent and Actual Job Changes. Human Relations 46: 987–1006. [Google Scholar] [CrossRef]
  92. Winslow, Sarah. 2005. Work-family conflict, gender, and parenthood, 1977–1997. Journal of Family Issues 26: 727–55. [Google Scholar] [CrossRef]
This gap between the intention to leave and actually leaving was also shown in our sample: 17% of the employees expressed such an intention, but only 6% to 7% actually left the firm voluntarily.
In Germany, all establishments are given an establishment number (Betriebsnummer) in the context of the social security registration process. Establishment numbers cover individual firms but also single units or subsidiaries of companies.
Thus, we were able to capture changes in the personal, family and job-related situation for employees between each wave. Independent variables were always taken from the wave directly preceding the wave in which the dependent variable was measured; i.e. the predictors for the dependent variable of a voluntary exit in wave 3 were “updated” with information from wave 2 instead of wave 1. Furthermore, there were no substantial differences regarding the distribution of personal and job characteristics in the samples of each wave, which excludes the possibility that results may be influenced by such discrepancies due to a pooling of the samples.
Calculating panel analyses was not possible, since we did not have information on the organizational environment in the next wave for those who had previously left the establishment.
These cases were dropped because linkage to administrative data from the IAB was not permitted or information concerning the observed individual variables was missing. Additionally, we dropped companies with missing information regarding the availability of work–life offers (n = 2) or in cases in which only male or female employees were surveyed (n = 3).
We are aware of the possible drawbacks of this study design: first, those employees who were employed in the same establishment in all three waves are systematically more similar than other employees (e.g., in having higher earnings as a result of seniority effects). However, owing to the unequal distribution of the dependent variable, with few cases for leavers, we decided to process this approach. Second, to further analyze possible biases due to different time spans between waves, we included the wave as a control variable, but we also calculated the models separately for each wave combination. The conclusions from these models were fairly similar for the overall effect of flexibility measures (the group differences were not that robust which may emerge from really small case numbers when separating the waves); however, the coefficients of the measures were not significant when we considered the transition from wave 1 to wave 2. We attribute this result mostly to the time-span difference between the waves (w1 to w2 = 2 years; w2 to w3 = 4 years), with a higher relative percentage of exits from wave 2 to wave 3 (w1 to w2 = 5.9%; w2 to w3 = 9.3%). By controlling for the wave and various individual characteristics, we took these biases into account within the analyses.
Figure 1. Predictive margins of the association between the perceived availability of home-based teleworking and voluntary employee exit moderated by a highly demanding culture. Note. Predictive margins of the known availability of home-based teleworking and a highly demanding culture with 95% confidence intervals (Cis) using stata are based on the mixed-effects regressions in Table 3.
Figure 1. Predictive margins of the association between the perceived availability of home-based teleworking and voluntary employee exit moderated by a highly demanding culture. Note. Predictive margins of the known availability of home-based teleworking and a highly demanding culture with 95% confidence intervals (Cis) using stata are based on the mixed-effects regressions in Table 3.
Socsci 10 00009 g001
Figure 2. Predictive margins of gender differences in the association between the perceived availability of home-based teleworking and voluntary employee exit. Note. Predictive margins of gender and home-based teleworking with 95% CIs using stata are based on the mixed-effects regressions in Table 4.
Figure 2. Predictive margins of gender differences in the association between the perceived availability of home-based teleworking and voluntary employee exit. Note. Predictive margins of gender and home-based teleworking with 95% CIs using stata are based on the mixed-effects regressions in Table 4.
Socsci 10 00009 g002
Figure 3. Predictive margins of gender differences for parents in the association between the perceived availability of home-based teleworking and voluntary employee exit. Note. Predictive margins of gender and home-based teleworking with 95% CIs using stata are based on the mixed-effects regressions in Table 4.
Figure 3. Predictive margins of gender differences for parents in the association between the perceived availability of home-based teleworking and voluntary employee exit. Note. Predictive margins of gender and home-based teleworking with 95% CIs using stata are based on the mixed-effects regressions in Table 4.
Socsci 10 00009 g003
Table 1. Means and standard deviations of study variables.
Table 1. Means and standard deviations of study variables.
Voluntary employee exit0.070.25
Individual characteristics
Job characteristics
Tenure (years)9.217.78
Hourly wages (log.)3.160.43
Hourly wages26.1719.60
Supervisory responsibility0.370.48
Contractual agreed monthly working hours35.157.16
Fulltime job0.760.43
Amount of overwork1.130.17
Support by direct supervisor0.740.44
Support by coworkers0.780.41
Job satisfaction7.361.82
Balance of expectations & gratifications3.600.88
Qualification and self-assessment
Not known/no vocational degree0.030.16
Low-track/intermediate-track secondary school with vocational training0.400.49
School-leaving certificate for German university entrance with vocational training0.190.39
University degree0.390.49
Self-assessed chances at the labor market3.551.25
Establishments characteristics
Work–life measures
Flexible working times0.790.41
Home-based teleworking0.450.50
Family-friendly culture3.870.24
Demanding culture3.770.46
Public sector0.360.48
Western Germany0.830.37
Economic sector
Information/economic services0.210.41
Firm size
500–699 employees0.390.49
700–999 employees0.320.47
1000–1499 employees0.180.39
1500 and more employees0.110.32
Table 2. The effect of flexibility measures on voluntary employee exit (logistic regression with organizational mixed effects and clustered standard errors; odds ratios).
Table 2. The effect of flexibility measures on voluntary employee exit (logistic regression with organizational mixed effects and clustered standard errors; odds ratios).
Flexible working times0.700*(0.109)
Home-based teleworking 0.621**(0.090)
Individual characteristics
Sex (Ref = men)0.666**(0.088)0.668**(0.088)
Parenthood (Ref = no children)0.741*(0.110)0.732*(0.110)
Partnership (Ref = no partner)0.945 (0.146)0.970 (0.151)
Job characteristics
Tenure (in years)0.896***(0.014)0.894***(0.015)
Hourly wages (log.)1.084 (0.198)1.117 (0.198)
Supervisory responsibility (Ref = no supervisees)0.982 (0.146)0.969 (0.145)
Contractually agreed monthly working hours0.971**(0.010)0.972**(0.010)
Amount of overwork1.615 (0.588)1.658 (0.613)
Job satisfaction0.788***(0.027)0.786***(0.027)
Balance of expectations and gratifications0.958 (0.066)0.964 (0.066)
Qualification & self-assessment
Self-assessed chances at the labor market0.796***(0.043)0.795***(0.043)
(Ref = university degree)
No degree1.189 (0.457)1.147 (0.443)
Low-track/intermediate-track secondary school with vocational training0.668**(0.099)0.651**(0.097)
School-leaving certificate for German university entrance with vocational training0.985 (0.155)1.000 (0.154)
Establishments characteristics
Public sector (Ref = private sector)0.978 (0.238)0.959 (0.224)
Western Germany (Ref = Eastern Germany)0.974 (0.177)1.004 (0.179)
Economic sector
(Ref = production/energy/water/construction)
Retail/transport/hospitality1.111 (0.314)1.131 (0.326)
Information/economic services1.679*(0.440)1.778*(0.468)
Administration/education/health1.430 (0.432)1.487 (0.435)
Firm size
(Ref = 500–699 employees)
700–999 employees0.661+(0.157)0.670+(0.158)
1000–1499 employees0.854 (0.205)0.849 (0.200)
1500 and more employees1.184 (0.292)1.170 (0.287)
Wave (Ref = wave 1 to 2)2.209***(0.347)2.211***(0.350)
Constant0.472 (0.140)2.192 (1.885)
Note. Clustered robust standard errors in parentheses; + p < 0.10, * p < 0.05, ** p < 0.01; *** p < 0.001; Ref = reference category.
Table 3. The moderating effect of the organizational environment on the association between flexibility measures and voluntary employee exit (logistic regression with organizational mixed effects and clustered standard errors; odds ratios).
Table 3. The moderating effect of the organizational environment on the association between flexibility measures and voluntary employee exit (logistic regression with organizational mixed effects and clustered standard errors; odds ratios).
Flexible Working TimesHome-based Teleworking
Work-life measures (Ref = availability not perceived)0.792 (0.198)0.610*(0.153)
High support by the supervisor (Ref = low support)1.156 (0.286)1.009 (0.172)
Work-life measure x support by the supervisor0.829 (0.242)1.025 (0.309)
Constant2.626 (2.494)2.174 (1.962)
Work-life measures (Ref = availability not perceived)0.897 (0.244)0.709 (0.181)
High support by coworkers (Ref = low support)1.292 (0.322)1.104 (0.194)
Work-life measure x support by coworkers0.726 (0.214)0.838 (0.206)
Constant2.495 (2.262)2.052 (1.800)
Work-life measures (Ref = availability not perceived)0.608 (1.379)0.919 (2.423)
Family-friendly culture0.324*(0.165)0.354*(0.159)
Work-life measure x family-friendly culture1.047 (0.609)0.910 (0.616)
Work-life measures (Ref = availability not perceived)1.129 (1.660)0.039**(0.041)
Highly demanding culture1.958*(0.589)1.319 (0.300)
Work-life measure x high demanding culture0.895 (0.332)2.056**(0.561)
Constant0.321 (0.450)1.120 (1.234)
Note. Clustered robust standard errors in parentheses; * p < 0.05, ** p < 0.01; Controlled for gender, parenthood, partnership, tenure, hourly wages, supervisory responsibility, self-assessed chances at the labor market, contractually agreed monthly working hours (centered), amount of overwork, job satisfaction, qualification, public sector, place of establishment’s residence, economic sector, firm size and wave; tables including control variables available on request; Ref = reference category.
Table 4. Group differences in the association between voluntary employee exit and work-life measures (logistic regression with organizational mixed effects and clustered standard errors).
Table 4. Group differences in the association between voluntary employee exit and work-life measures (logistic regression with organizational mixed effects and clustered standard errors).
Flexible Working TimesHome-based Teleworking
Work-life measures (Ref = availability not perceived)0.624*(0.122)0.518***(0.087)
Gender (Ref = men)0.575**(0.119)0.566**(0.095)
Work-life measure x gender1.237 (0.300)1.553+(0.366)
Constant3.247 (2.870)0.464 (0.132)
Work-life measures (Ref = availability not perceived)0.606*(0.128)0.697+(0.137)
Parenthood (Ref = no children)0.623+(0.161)0.792 (0.134)
Work-life measure x parenthood1.288 (0.348)0.816 (0.184)
Constant3.381 (3.066)2.016 (1.772)
N 5452
Work-life measures (Ref = availability not perceived)0.737 (0.223)0.416***(0.104)
Parents (Ref = fathers)0.684 (0.255)0.600+(0.160)
Work-life measure x parents1.140 (0.396)1.796+(0.595)
Constant1.834 (2.359)1.410 (1.712)
Note. Clustered robust standard errors in parentheses; + p < 0.10, * p < 0.05, ** p < 0.01; *** p < 0.001; Controlled for gender, parenthood, partnership, tenure, hourly wages, supervisory responsibility, self-assessed chances at the labor market, contractually agreed monthly working hours (centered), amount of overwork, job satisfaction, qualification, public sector, place of establishment’s residence, economic sector, firm size and wave; tables including control variables available on request; Ref = reference category.
Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Share and Cite

MDPI and ACS Style

Marx, C.K.; Reimann, M.; Diewald, M. Do Work–Life Measures Really Matter? The Impact of Flexible Working Hours and Home-Based Teleworking in Preventing Voluntary Employee Exits. Soc. Sci. 2021, 10, 9.

AMA Style

Marx CK, Reimann M, Diewald M. Do Work–Life Measures Really Matter? The Impact of Flexible Working Hours and Home-Based Teleworking in Preventing Voluntary Employee Exits. Social Sciences. 2021; 10(1):9.

Chicago/Turabian Style

Marx, Charlotte K., Mareike Reimann, and Martin Diewald. 2021. "Do Work–Life Measures Really Matter? The Impact of Flexible Working Hours and Home-Based Teleworking in Preventing Voluntary Employee Exits" Social Sciences 10, no. 1: 9.

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Back to TopTop