Transition to fatherhood and adjustments in working hours: The importance of organizational policy feedback

Objective: This study investigates whether the normalization of the use of the family-friendly workplace policy fle xiplace in the organization affects men’s adjustments in working hours following their transition to fatherhood. Background: Men’s stable full -time employment after childbirth remains to be a barrier to the equal distribution of care and paid work. Recent research suggests that state family policies promoting dual-earner/dual-carer family models can involve new norm setting of active fatherhood, albeit so far with only modest consequences for fathers’ working hours. Unclear is, however, whether famil y-friendly workplace policies, such as flexiplace, and involved organizational policy feedback are of complementary importance. Method: We estimated fixed-effects regression analyses on men’s adjustments in actual and contracted hours after a transition to fatherhood. Analyses are based on linked employer-employee panel data (2012/13; 2014/15; 2018/19) from large German work organizations, considering a random sample of 1,687 men in 131 work organizations. Results: Findings revealed that the normalization of using flexiplace in the work organization was associated with a reduction in men’s overall working hours as well as marginal adjustments in their contracted hours after transitioning to fatherhood. Conclusion: Although a normalization of flexiplace is more likely in demanding workplace contexts, men experience at least some leeway in adjusting extensive temporal investments to cater to private demands.


Introduction
Adapting working hours to an employee's family situation is one strategy for better aligning work and family responsibilities.Whereas women are more likely to work fewer hours when they have children at home (e.g., Rosenfeld & Birkelund, 1995;Van der Lippe, 2001), men, once they become fathers, rarely work fewer hours.Instead, they are more likely to work even longer hours.More recently, the ideal of active fatherhood has become more important in several European countries (e.g., Bünning and Pollmann-Schult, 2016;Bünning, 2015;Daly, 2011;Hobson & Fahlén, 2009;Pfau-Effinger, 2012;Schober, 2014), with fathers increasingly seeking a reduction in their working hours to devote more time to their families (Abendroth & Pausch, 2018;Hobson & Fahlén, 2009;Pollmann-Schult & Reynolds, 2017).Policy feedback theory (Campbell, 2012;Gangl & Ziefle, 2015;Grunow et al., 2018) suggests that this is a result of state policies that increasingly support a dual earner/dual carer family model, e.g., with periods of parental leave exclusively reserved for fathers.However, existing research indicates that the share of fathers who have actually reduced their working hours remains relatively low (Adams & Golsch 2022;Hobson & Fahlén, 2009;Pollmann-Schult & Reynolds, 2017).In the family domain, financial considerations have been found to be of importance (Pollmann-Schult & Reynolds, 2017) which aligns with previous research on the importance of the family domain for work-family conflicts (Schulz & Reimann, 2022).
The aim of this research is to turn the focus towards the work domain by investigating the importance of organizational family-friendly policies as a necessary precondition for adjustments in working hours after a transition to fatherhood.Workplaces are the sites where working hours are negotiated (Den Dulk, 2001;Tomaskovic-Devey & Avent-Holt, 2019).In these workplaces, the norm of an ideal worker with a traditional male life course has a long tradition, which is likely to function as an organizational barrier to fathers' reduction in working hours (Acker, 1990;Kelly et al., 2010;Williams et al., 2013).Indeed, part-time work often means lower hourly pay (Aaronson & French, 2004;Hirsch, 2005;Paul, 2016) and hampered career progress (Durbin & Tomlinson, 2010;van Osch & Schaveling, 2020), whereas overtime work seems to be rewarded by a wage premium (Cha & Weeden, 2014).
However, some organizations have responded to the growing number of dual earner families and the involved struggles in combining work and family demands by implementing family-friendly policies (Den Dulk, 2001;Frodermann et al., 2018).In this context, flexiplace 1 is one of the common and widely studied examples in terms of access and use (Homberg et al, in press;Arntz et al., 2022), its impact on work-life conflicts (Abendroth, 2022;Van der Lippe & Lippényi, 2020), and working hours (Arntz et al., 2022;Chung & Van der Horst, 2018).
Applying the mechanism of norm-setting and cultural diffusion in policy feedback theory, the provision and normalization of the use of flexiplace as a family-friendly policy can also contribute to a norm-setting of family-friendliness which contrasts the ideal worker norm: it signals to fathers that a reduction in working hours is an available opportunity and to supervisors that it is a legitimate request.In addition, the mechanism of role exposure suggests changes in fathers' work-family preferences due to exposure to new family roles while working from home.Previous research, however, also provides a more skeptical view on a norm-setting of family-friendliness as flexible working was found to be associated with expectations to be responsive to high work demands in return (Lott & Chung, 2016;Gschwind & Vargas, 2019).Therefore, we ask: Does the normalization of using flexiplace in the organization increase the likelihood that men reduce their contractual or total working hours after a transition to parenthood?
In addressing this research question, our study contributes to existing research on adjustments in men's working hours after transitions to fatherhood in several ways.Previous research has contributed a family perspective highlighting the importance of the gendered division of labor within couples as a central barrier to fathers' reductions in working hours (Bünning and Pollmann-Schult, 2016;Bünning, 2015;Pollmann-Schult & Reynolds, 2017) and the importance of family and partner support to reduce their workto-family conflict (for review see Reimann et al., 2022).On the importance of the work domain, crosssectional evidence shows that the working hours of fathers (Alemann et al., 2017;Bernhardt & Bünning, 2018;Haas & Hwang, 2016;Liebig et al., 2017) and their work-family conflicts (Abendroth, 2022;Schulz & Reimann, 2022) not only depend on the family but also the workplace context.For example, fathers were found to work fewer hours in work organizations where regulated measures for work-life reconciliation are made available to all employees (Bernhardt & Bünning, 2018).However, an organizational policy feedback perspective has not yet been applied.
Second, we contribute to existing research by using unique German linked employer-employee panel data (LEEP-B3).It combines information on employees' adjustments in working hours (total and contracted) between 2012/13 and 2018/19 (in three survey waves) with information on organizational policies and practices in large German work organizations where flexiplace is especially common (Den Dulk, 2001;Frodermann et al., 2018).This allows us to investigate adjustments in men's working hours rather than general differences in working hours between men and fathers who work in different familyfriendly organizations.As political initiatives to increase active fatherhood have gained importance over the time-period studied, the case of Germany is an especially interesting one to investigate adjustments in men's working hours after a transition to parenthood (e.g., Zerle-Elsäßer & Li, 2017).

Societal change toward active fatherhood in Germany
Germany has long adhered to the traditional male-breadwinner family model that was politically supported by long parental leave regulations and comprehensive financial benefits for child-rearing for periods of up to three years (Schober, 2014), as well as joint income taxation on married couples (Hipp et al., 2015).These regulations incentivized the gendered division of labor among couples and fostered gendered expectations regarding mothers' and fathers' responsibilities for household duties and paid work, respectively.
However, over the past two decades legal regulations and family policies have been introduced that increasingly reflect the dual earner/dual carer family model in which both parents are employed and involved in care work.This shift is meant to promote equal opportunities for mothers and fathers to participate in both care and paid employment, thus setting incentives for both partners to return to work earlier after birth and increasing the possibilities for parents to reconcile their work and family lives (Leitner et al., 2004;Samtleben et al., 2019).Early evidence, however, shows that political strides toward a dual earner/dual carer family model were only modestly accompanied by fathers' greater involvement in childcare or a reduction in their working hours (Bünning, 2015;Schober, 2014;Pollmann-Schult & Reynolds, 2017).

Organizational barriers towards active fatherhood
Work organizations are the locations where fathers' working hour adjustments are negotiated and perceived to be legitimate or not (for a discussion of relational inequality theory, see Tomaskovic-Devey & Avent-Holt, 2019).Following Van Breeschoten et al. (2019), organizations function as "gatekeepers to the statutory work-family policies, supporting or hindering fathers' actual uptake of formal possibilities for reconciling work and private life" (p.78).The norm of the ideal worker has a long tradition in work organizations.In this context, work organizations expect and reward working long hours and presence at the regular work site.This imposes barriers for fathers to request or even realize a reduction in working hours (Acker 1990;Kelly et al., 2010;Williams et al., 2013).Evidence of organizational barriers to reductions in fathers' working hours are provided by studies which show that fathers do not feel entitled to reduce their working hours and therefore often do not do so (Alemann et al., 2017;Bernhardt & Bünning, 2018;Clarkberg & Moen, 2001;Hobson & Fahlén, 2009;Munsch et al., 2014;Thornthwaite, 2004).For Sweden, Haas & Hwang (2016) have also shown that fathers' adoption of working reduced hours was hampered by weak organizational support.Research has further conveyed that employees in general fear repercussions for their career when they deviate from the norm of the ideal worker (Beauregard, 2011;Chung, 2020;Den Dulk & De Ruijter, 2008).

Family-friendly policies and policy feedback
Some organizations, have, however, responded to increased normative and/or economic pressures to be more family-friendly by providing policies meant to support the integration of work and family demands (Den Dulk, 2001, Den Dulk et al., 2012).Flexiplace, in this regard, is a workplace policy aimed at supporting employees' work-family reconciliation.In 2018, flexiplace was available to about 26% of workers (Grunau et al., 2019).It is a practice which deviates from the standard of physical presence as is common in the ideal worker norm.
Viewed through the lens of policy feedback theory (Campbell, 2012;Gangl & Ziefle, 2015;Grunow et al., 2018), the provision of flexiplace as a common organizational family-friendly policy might contribute to setting a more general norm of family-friendliness in the organization additionally disrupting the norm of high work commitment (see also Abendroth, 2022).Norm-setting in policy feedback theory describes that policies set a broader normative framework to which individuals adapt their behaviors, attitudes, and beliefs (Gangl & Ziefle, 2015;Grunow et al., 2018).Cultural diffusion further describes that those who use policies meant to better align work and family encourage others of their use and to show interest in work family reconciliation.This dynamic is supported in the study by Bygren & Duvander (2006) on the normalization of fathers taking leave in work organizations.This suggests that a general norm-setting of familyfriendliness by flexiplace is a longer process which requires a normalization of its use.Previous research has shown that in some organizations flexiplace is not widely used because employees are afraid of a flexibility stigma (Chung, 2022;Munsch, 2016;Williams et al., 2013).A normalization of its use, in contrast, indicates the dissolution of these concerns and a more general family-friendly work culture.Indeed, initial evidence shows that the normalization of working from home goes hand in hand with reduced work-family conflicts among employees regardless of the individual use, thus pointing to more family-friendly work cultures (Abendroth, 2022).Moreover, family-friendly organizations have not only been identified by the availability of family-friendly policies but also by the lack of career penalties involved (Thompson et al., 1999).In line with this, Moen (2011) noted that "the language of 'family-friendliness' and 'best corporations for working mothers' means that family obligations became a salient human resource issue" (p.13).To conclude, an increased use of working from home might gradually normalize deviations from the ideal worker norm and institutionalize family-friendly structures and practices at the workplace in the long run.
We argue that this general norm-setting and cultural diffusion of family-friendliness shapes the likelihood of men's adjustments in working hours.First, the normalization of the use of flexiplace is likely to signal to fathers that it is legitimate to choose to better integrate work and family demands by reducing working hours after a transition to fatherhood.Second, the normalization is likely to signal that the use of this policy to better align work with care does not involve career repercussions or the stigma of being less committed to work.Thus, fathers in general feel entitled to reduce (overlong) working hours in order to have more time for their children.Third, we argue that this general norm-setting and cultural diffusion of family-friendliness increases the likelihood that supervisors perceive reductions in working hours to be legitimate as well.Increasing experience with the use of working from home might additionally undermine the stigma of being less committed to work when using work arrangements which help to better align work and family, that is, the belief on the part of colleagues, supervisors, and/or management that the use of flexible working arrangements is associated with lower commitment, productivity, and work performance (Chung, 2020;Leslie et al., 2012;Munsch, 2016;Carlson et al., 2021;Williams et al., 2013).This is in line with previous research which describes claims-making as a two-step process: an employee poses a claim and a powerful actor decides on the legitimacy of the claim, which determines the access to organizational resources (Tomaskovic-Devey & Avent-Holt, 2019).
Additional policy feedback can be implied for fathers who use this policy.Following the process of norm-setting, employees who use flexiplace could experience it as an alternative work arrangement that provides them with autonomy to organize their work in alignment with their family demands (Chung & Van der Horst, 2018).Hence, workers might feel greater entitlement to adjust their working hours while using flexiplace arrangements, as the norm of physical presence is non-existent.Additionally, in line with the mechanism of role exposure, working from home during transitions to fatherhood can increase fathers' interest in reducing working hours as they are more exposed to caring obligations at home due to the shared location of paid work and care work (Carlson et al., 2021).Prior research related fathers' reductions in working hours after taking parental leave to their increased exposure to child-rearing (e.g., Bünning, 2015).Similarly, using flexiplace after childbirth increases fathers' exposure, at least to some extent, because family and work tasks are done from the same location (Carlson et al., 2021;Ge Gao & Ruan, 2022;Tomei, 2021).Thus, men become more aware of the needs and pleasures of a larger involvement in childrearing and, hence, readjust their time spent in unpaid and paid work by reducing their working hours.
Therefore, we hypothesize: H1: The normalization of the use of flexiplace in an organization increases the likelihood that men reduce their working hours after their transition to fatherhood.
However, previous research which describes the use of family-friendly policies as a gift-exchange dynamic suggests that this form of policy feedback is rather weak.Gift-exchange dynamics describe that employers perceive the provision of family-friendly policies and practices as a gift and expect higher work commitment in return (Lott & Chung, 2016).Indeed, existing research findings indicate that flexiplace can involve work intensification and increased work-family conflicts especially for men (Glass & Noonan, 2016;Gschwind & Vargas, 2019;Kelliherr & Anderson, 2010; Van der Lippe & Lippenyi, 2020)2 .Thus, policy feedback might be hampered by supervisors' concerns about work commitment and responsiveness to high work demands.

Data
Analyses are based on three waves (2011/12, 2014/15, 2018/19) of the German Linked Employer-Employee Panel Survey (LEEP-B3; Marx et al., 2020), which was conducted in cooperation with the Institute for Employment Research (IAB) in Nuremberg (Abendroth et al., 2014).The LEEP-B3 includes information on organizations and the employees working in them.The sample of work organizations was randomly drawn from administrative records.Within these organizations, employees were randomly selected to participate in the survey.In 2014/15 and 2018/19, additional refreshment samples were taken. 3The final data set is representative of workers who are required to make social security contributions and who are employed in large German work organizations that have at least 500 employees (Diewald et al., 2014).The hierarchical structure of the data allows for combining information from three different levels, with respondents clustered in work organizations and clustered in years.
For the analyses, which focus on men's transition to parenthood, data were restricted to employed and partnered men between the ages of 17 and 55, who participated in at least 2 waves of the survey (4,749 observations and 2,013 respondents).Listwise deletion of missing information for the variables of interest resulted in a final sample of 1,683 men and 3,867 observations.About 18% of men experienced the birth of their first child within the observation period4 .On average, respondents participated in 2.3 waves.

Working hours
Respondents provided information on their a) contracted weekly working hours and their b) actual working hours, including paid and unpaid overtime.Contracted working hours were top coded at 40 hours and actual working hours were top coded at 70 hours per week.

Transition to parenthood
We defined a dummy variable coded 0 when respondents had no children and coded 1 when respondents had at least 1 biological, adopted, foster, or stepchild.Transitions are marked by a change of the variable from 0 to 1.

Flexiplace
To examine the role of organizational policies for family-friendliness, we used the individual-level information of respondents' uptake of flexiplace, based on employees' responses on whether they worked from home.The individual use variable differentiates between individual use (coded 1) or no use (coded 0).The normalization of the use of flexiplace in the organization is measured with the help of the average flexiplace usage in the organization, indicating the general acceptance of relying on flexible work arrangements within a workplace.For this measure, we used all available information5 on individual use of flexiplace and aggregated it on the organizational level.We compare two measurements.First, a timeinvariant indicator that was taken from the organization's first observation to identify the importance of policy feedback from the base-level of normalization of flexiplace for men's working-time adjustments in the subsequent observations.And, second, a time-varying indicator identifying policy feedback by increases in the normalization of flexiplace over time.We refrained from a lagged version due to the limited number of waves in the sample and the subsequent loss of a substantial amount of cases.Organizational flexiplaceuse also seems to be an indicator of general acceptance of using family-friendly work arrangements, as shown by positive correlations between flexiplace-use, flextime-use and lower experiences of flexibilitystigma (see Appendix, Table A1).

Work experience
Considering that working hours vary based on career stage, a yearly measure of work experience was added.Additionally, a squared term was added to account for the u-shaped function of working hours: workers first increase their working hours and reduce them later in their careers (see also Costa & Sartori, 2007).

Partnership arrangement
Previous research has shown that fathers' working hours after childbirth are dependent on household level characteristics (Reynolds & Pollmann-Schult, 2017).To account for household-bargaining, we additionally built a combined measure containing information on respondents' marital status and the employment status of their partner or spouse.The categorical variable differentiates between 1) unmarried men, partner not employed, 2) married men, partner not employed, 3) married men, partner employed, and 4) unmarried men, partner employed.
Sample descriptives further reveal that childless men and fathers show only little variation in the use of flexiplace or the organizational average (Table 1).Thus, it seems that neither is the uptake of flexiplace related to parental status, nor do fathers sort themselves into organizations where better possibilities for combining work and family are provided (e.g., by offering flexiplace).However, fathers have, on average, greater work experience and are more likely to be married than non-fathers.Comparing the descriptive of the final sample with the sample including missing information reveals only marginal differences (see Appendix, Table A2).Only observations of fathers with a partner who is not working are somewhat more common in the deleted sample with missing information.Nevertheless, prior publications on coupled, male employees in Germany, indicate that 80% have a partner in paid employment (Reynolds & Pollmann-Schult, 2017).This number is similar to our sample (see Table 1 with descriptive information).

Method
The empirical analyses were based on individual fixed effects models with clustered standard errors on the individual level (Allison, 2009).Using fixed-effects regression models is advantageous for studying transitions to parenthood, as effects are estimated based on only intra-individual variation over time; that means instead of considering the variation between childless men and fathers, only the change of parental status (from being childless to becoming a first-time father) is used to estimate the effect of parental status on working hours.The models are built stepwise: the first model estimates the effect of the transition to parenthood on men's working hours, net of general time-trends, work experience, and the partnership arrangements.Next, flexiplace, the organizational family-friendly policy of interest, is considered: first, interactions between transitions to fatherhood and the time-invariant indicator for normalization of the use of flexiplace and, second, the time-variant indicator are included into the models to investigate the hypothesis on policy feedback.
Following the main analyses, we additionally ran sensitivity analyses (provided in Appendix).To examine how men's working hours are generally distributed between organizations with a different availability of flexiplace, we first estimated random effects models on men's working hours (see Table A3).FE-modelling does not allow us to estimate the general association between organizational use of flexiplace and men's working hours because this measure did not vary over time.Therefore, RE modelling is helpful as it considers the differences between organizations and workers as well.Moreover, we estimated our set of FE-models for different subsamples: first, we split the sample according to organizational use of flexiplace, differentiating between organizations in which up to 25% of workers used flexiplace and those where the usage exceeded 25% (see Table A4).Secondly, we split the sample according to men's gender beliefs (using three different items) to see whether reductions in working time mainly rely on personal beliefs instead of organizational work-family policies (flexiplace), and how gender beliefs potentially shape their use of family-friendly workplace policies (see Table A5-A7).

Results
Table 2 shows the results of the fixed-effects regression models on men's adjustments in their actual and contracted working hours when transitioning to fatherhood.

Transition to fatherhood
Model 1 in Table 2 reveals that men do not reduce their actual working hours when becoming fathers.
Although the coefficient shows a small and negative tendency in working time when transitioning to parenthood, the effects do not reach statistical significance (-0.446, p>0.1).Interestingly, changes in men's family arrangements seem to be relevant to men's working hours instead.Men's working hours increase by about 1 to 2 hours when they transition to marriage (1.836, p<0.01 when partner is not employed, 1.446, p<0.05, when partner is employed).This finding points towards the relevance of the economic situation in the household and bargaining between partners.Looking at contractual working hours (Table 2, M4 to M6) reveals that these remain unaffected by childbirth as well.

Normalization of flexiplace arrangements at the workplace level
Models 2 and 3 investigate the importance of the normalization of the use of flexiplace in the organization.
Both variables (time-constant/time-varying measure) range from 0 (no one works from home) to 100% (all of workers in the organization work from home).
Model 2 shows the interaction-term between transitioning to parenthood and the time-invariant measure of organizational normalization of flexiplace-use based on each organizations' first observation in the data. 6The interaction term shows that in organizations where the use of flexiplace has already normalized, men indeed worked fewer hours following childbirth (-0.090, p<0.05).Figure 1 depicts the marginal effects-plot based on Model 2, containing the estimated difference in working hours for men after childbirth due to the normalization of the use of flexiplace in the organization.The figure shows this relationship in more detail: in organizations in which flexiplace-usage surpassed a threshold of 40% of the workforce at the beginning of the observation period, first-time fathers work significantly fewer hours compared to before childbirth.The figure suggests a reduction of 5 hours in organizations where already 70% of workers used flexiplace at the time the organization entered the survey.This finding is supportive of hypothesis 1, stating that the normalization of the use of flexiplace in an organization increases the likelihood that men reduce their working hours after their transition to fatherhood.
In model 3, the time-varying measure of flexiplace-use which is based on yearly average flexiplace-use within organizations is integrated into the models.Incorparating this alternative measure additionally reveals that an increase in organizational use of flexiplace-arrangements over time does not affect men's working hours per se (0.029; p<0.1).However, when use in flexiplace-arrangements increases men who become fathers do reduce their working time (M3; -0.039; p<0.1).The results are graphically displayed in Figure 2 and show a similar development compared to figure 1.The more widely used flexiplacearrangements become, the more substantial are first-time fathers working time-reductions.However, this result is only evident with a significance level of p<.1.Thus, this result only lends some support for the assumption that the normalization of flexiplace-use are accompanied by greater reductions in working hours when men transition to fatherhood.
Followingly, we estimated the same set of models on fathers contractual working hours (Model 4-Model 6).Model 4 in Table 2 shows that adjustments in contracted working hours remain marginal.However, results on policy feedback point to a similar direction as findings for actual working hours (M5: -0.025; p<0.1, M6 -0.017; p<0.05) further strengthening the support of hypothesis 1 on organizational policy feedback although for a modest degree of reductions in working hours after transitions to parenthood.Lastly, we ran additional sensitivity analyses: we first split the analyses by organizations ranking (below) average (about 25%) in flexiplace-use or above average.Secondly, we estimated models differentiating beliefs on the gendered division of labor to consider different preferences for active fatherhood and whether fathers with more egalitarian beliefs are more likely to work in organizations with more family-friendly workplace policies and practices.Applying these different robustness checks, we found that men only reduced their working hours in organizations in which flexiplace-use exceeded average-levels in the beginning of the observation period; yet, childless men worked longer hours prior to birth (see Table A4).Hence, analyses suggest that in each type of organization, fathers in the end work similar hours.Moreover, comparing different measures for traditional/ or egalitarian gender beliefs among fathers, findings conformingly reveal that fathers with more egalitarian gender beliefs only reduce their working hours in organizations where flexiplace usage has been normalized (see Appendix, Tables A5-A7).However, the results also indicate that more egalitarian gender beliefs are an additional precondition for reductions in fathers' overly long working hours.
6 Because this model relies on the time-invariant measure of WP normalization of flexiplace-use, no direct effect can be estimated using FE modelling.Hence, this type of model (despite its advantages) cannot tell us how men's working hours generally vary between organizations with different degrees of normalization of flexiplace usage.We additionally estimated random-effects models considering variations within and between individuals and organizations, showing that men generally work longer hours in organizations in which flexiplace-use was already widespread in the beginning of the survey (Appendix, Table A3, Model 2; 0.054; p<0.01).

Discussion
Work organizations have a long tradition of the norm of the ideal worker following traditional male life courses (Acker, 1990;Kelly et al., 2010;Williams et al., 2013).It has been exacerbated in times of globalization with high expectations regarding availability for work even outside regular working hours and work sites (Cha & Weeden, 2014;Wajcman, 2016).Because the norm of the ideal worker also carries expectations about traditional male careers, fathers' reductions in working hours for work-family reconciliation are hindered, even in spite of growing preferences for doing so in line with the emerging ideal of involved fatherhood.More recent strides of organizations to become more family-friendly have been explained by neo-institutionalist arguments (Den Dulk, 2001;DiMaggio & Powell, 1983;Frodermann et al., 2018;Meyer & Rowan, 1977), rational choice and business case reasoning (Coleman 1990;Den Dulk, 2001;Den Dulk et al., 2012) pointing to increasing normative and economic pressures on work organizations to become more family-friendly.It is unclear, however, if the implementation of family-friendly policies involves policy feedback contributing to a norm-setting of family-friendliness which contrasts the ideal worker norm and signals to fathers that reducing working hours is an available option and to supervisors that it is a legitimate request and/or practice.
Against this backdrop, this study investigated the importance of normalizing family-friendly work arrangements, specifically the spread of flexiplace usage in organizations for men's adjustments in total and contractual working hours following their transition to fatherhood.Applying arguments of policy feedback theory (i.e., norm-setting and cultural diffusion as well as role exposure) to the organizational sphere, we expected that transitions to fatherhood lead to more pronounced reductions in working hours for men working in organizations where the use of flexiplace arrangements had already been normalized.
For the observation period 2012/13-18/19, we firstly conclude that men working in large German workplaces have, on average, not adjusted their working hours when becoming fathers.Instead, partnership arrangements seemed to be a relevant driver of increases of men's working hours.Entering more traditional partnerships arrangements (e.g., being married with a spouse not in employment) lead to an increase in working hours for men.This is in line with the findings by Pollmann-Schult & Reynolds (2017), suggesting that financial considerations in the family domain restrict fathers' opportunities to work reduced hours (Pollmann-Schult & Reynolds, 2017).Moreover, it is in line with the study by Adams & Golsch (2022) showing that scaling back is not a common strategy to deal with work-family conflict.
Turning to the work organization, we secondly conclude that the organizational context is of additional importance for fathers' adjustments in working hours after a transition to fatherhood.The results show that men were more likely to decrease their overall and contractual hours after transitions to fatherhood when working in organizations where the use of flexiplace had already been normalized.However, the size of the reduction remains modest.This finding provides some support for the argument that the normalization of family-friendly policies can contribute to a new norm-setting, signaling to fathers and supervisors that a reduction in working hours is an available and legitimate request and/or practice.This finding supports existing research emphasizing the role of the workplace context for fathers' working hours (Bernhardt & Bünning, 2020), and work-family conflicts (Abendroth, 2022;Lippenyi & Van der Horst, 2021).However, our findings also suggest that policy feedback of family-friendly policies has its limits as it does not seem to disrupt the norm of the ideal worker.Normalization of using flexiplace arrangements was more likely in organizations where average temporal investments in work were generally higher, compared to organizations where the use of flexiplace among workers was less widespread.Thus, men experience only some leeway to adjust their extensive temporal investments to cater to private demands.In the end, they work similar working hours in comparison to fathers working in less demanding organizations.To conclude, active fatherhood remains to be attenuated not only by financial considerations in the household but also by the persistence of the norm of the ideal worker in organizations.One explanation for the limited policy feedback is offered by previous research suggesting that the provision and use of family-friendly policies in organizations follows gift-exchange dynamics (Lott & Chung, 2016) where employers provide such policies but expect higher work commitment in return.In line with this, flexiplace is provided as a family-friendly policy for better reconciliation of work and family life but is also used to realize flexibility interests of employers in return (Lott & Chung, 2016).A reduction in working hours during transitions to fatherhood in highly demanding work organizations might be provided to sustain fathers' work commitment in spite of modest caring obligations in the household and/or as a gift with the expectation to be responsive to high work demand when the children grow older.
Our study has some limitations.The results allow conclusions only with regard to large workplaces, which are more visible in the organizational environment and are therefore more likely to be responsive to normative pressures.Further research will be needed to investigate whether similar patterns can be identified for small-to medium-sized work organizations.Similarly, the results allow conclusions to be made only for coordinated market economies, such as Germany.This indicates a need for additional research on liberal market economies to investigate whether, in these contexts, the pressures of market volatility and global competition outweigh the pressures imposed on work organizations to be more familyfriendly.Moreover, younger men's working time reductions might be more pronounced (Pollmann-Schult & Reynolds, 2017), which we potentially underestimated, but sample size did not allow for a comparison between different birth cohorts.Yet, controlling for egalitarian gender beliefs might reflect these differences to some extent, as gender beliefs are highly correlated with age.Prior research indicated the relevance of formal versus informal flexiplace arrangements for fathers' work-family reconciliation (Troup & Rose, 2012).In this study, we could not differentiate between the different types of these work arrangements, which might also be related to fathers' working hours.
Our results also raise new questions for further research.Cha and Weeden (2014) have shown that the gender wage gap is driven mainly by men's overlong working hours.Changes in the predominance of men's overlong working hours would raise the question of whether these would also weaken gender pay inequalities in the labor market.Moreover, this research was conducted before the COVID-19 pandemic which has increased the use of flexiplace in organizations not as a family-friendly policy but as a practice to reduce the spread of the virus.Finally, future research is required which investigates whether skilled labor shortage is more important for fathers' adjustments in working hours as it shapes their negotiation power.Finally, our research points to the importance of both the family and workplace context pointing to the need for future research to research their intersection.

Overall conclusion
Overall, we conclude, that the ideal worker norm in organizations continues to attenuate active fatherhood.Even though family-friendly policies seem to involve some policy feedback establishing the norm of familyfriendliness, this majorly concerns the integration of basic family responsibilities with high work demands.In line with current debates stressing the relevance of family-dynamics for work-family conflict (Schulz & Reimann, 2022), our findings further highlight that financial considerations and negotiation processes within couples and families remain an important constrain to fathers' reductions in working hours.This implies that future research should jointly consider how family and organizational dynamics may limit or support fathers' attempts to reconcile work and family life.More specifically, policy feedback theory and its application to work-family reconciliation strategies seem to require paying additional attention to the question how policies at the organizational (or national) level alter the exchange relations not only between employers and employees but also between partners.Lastly, this also pertains to the implications for preference formation regarding work-family reconciliation.
The findings of this study show that the increased acceptance of family-friendly organizational policies has at least some impact on fathers to consider working-time adjustments as a strategy for better reconciling work and family.This also stresses the importance of finding influential advocates of such policies in the organization, for example on the managerial level, to support work-family integration.Moreover, the findings on the household-level further emphasize the role of national level policies to support gender equality (i.e., abolition of joint income taxation in marriage) in order to allow a more egalitarian division of labor within households.

Figure 1 :Figure 2 :
Figure 1: Average reduction of actual working hours by WP normalization of flexiplace (based on fixed effects models including the time-invariant indicator of flexiplace-use)

Table 2 :
Fixed-effects regression models on men's working time adjustments when transitioning to fatherhood