Transitions to parenthood, flexible working and time-based work-to-family conflicts: A gendered life course and organisational change perspective

Objective: This study investigates how flexitime and flexiplace moderate the consequences of transitions to parenthood for time-based work-to-family conflicts for women and men, and whether the normalisation of their use in organisations additionally contributes to reducing work-to-family conflicts. Background: Although flexible working has been described as a resource for better aligning demands in the domains of work and family, the findings of previous – mainly cross-sectional – research on its consequences for work – family conflict are inconsistent. Method: Individual fixed effects analyses were conducted using linked employer-employee panel data for 1,973 partnered men and 1575 partnered women in 132 large work organisations in Germany. Results: Time-based work-to-family conflicts after transition to parenthood increased for men but decreased for women. This can be explained by women reducing their working hours. However, work-to-family conflicts remained rather stable despite of the transition to parenthood among women who used flexitime. This can partly be explained by their weaker work-to-family conflicts already before the transition as well as to adjustments in work investments being less common among them. There is some evidence that the normalisation of flexitime and flexiplace in the organisation is associated with fewer work-to-family conflicts among women and men. Conclusion: Flexitime seems to be not an additional but an alternative resource to decrease the likelihood of more frequent time-based work-to-family conflicts after transition to parenthood among women. The normalization of flexible working depicts organizational change towards more family-friendliness.


Introduction
With the rise in dual-earner families, conflicts between demands in the work and family domains are increasing (Adams & Golsch, 2021).The more common types of work-to-family conflict (Frone et al., 1992;O'Driscoll et al., 2004) are known to impair health, well-being, and the quality of parent-child relationships (Allen et al., 2000;Vieira et al., 2016).
In this context, flexible working 1 that gives employees control over when and where they work has been described as a resource for better aligning demands in the domains of paid work and family which might help to sustain family well-being and a more egalitarian gendered division of labour after transitions to parenthood (Abendroth et al., 2012;Chung & van der Lippe, 2020;Den Dulk et al., 2012;Eurofound & the International Labour Office, 2017).If flexible working, for example, allows responding to unpredictable family needs experiences of parenting warmth and consistency in the parent-child relationship are likely to be sustained and not irritated by time-based work-to-family conflicts (Verweij, Helmerhorst & Keizer 2021;Cooklin et al. 2014;Cooklin et al. 2016).Similarly, new opportunities of responding to predictable family needs with the help of flexible working can sustain women's work capacities in periods of increased family demands facilitating a more egalitarian gendered division of labour (Chung & Van der Horst 2017).Flexible working has therefore also been labelled and promoted as a family-friendly workplace arrangement (Chung & van der Lippe, 2020;Den Dulk, 2001).Flexitime 2 refers to flexible starting and finishing times, as well as to the possibility of interrupting work for other responsibilities, thus giving employees some control over when they work during a regular work week (Chung & van der Lippe, 2020).Flexiplace arrangements allow workers to perform their work not only at their regular work premises but also, for example, from home (Chung & van der Lippe, 2020).Also known as teleworking, working from home, or remote working, flexiplace gives employees some control over where they work (Allen et al., 2015;Bianchi & Milkie, 2010;Felstead & Henseke, 2017).Demand perspectives instead suggest that flexible working involves an intensification of work increasing rather than reducing work-to-family conflict (Chung 2019).Findings of existing studies on the consequences of flexible working for work-to-family conflicts are inconsistent, sometimes revealing less and sometimes more work-to-family conflict when flexitime or flexiplace arrangements are used (see the reviews in Allen et al., 2015;Bianchi & Milkie, 2010;Byron, 2005;Chung & van der Lippe, 2020;Gajendran & Harrison, 2007).
Contributing a gendered life course perspective (Elder et al., 2003;Moen, 2011) to existing research, the present study asks, first, how the use of flexitime and flexiplace arrangements shapes the consequences of transition to parenthood for time-based work-to-family conflict and, second, whether this differs for men and women.The birth of a child is an important life course transition that not only brings new challenges to combine work and family, but also shapes identity, behaviour, events, and transitions in the work domain (Elder et al., 2003).Existing research refers here to gendered adaptations to parenthood.Whereas women are more likely to reduce their work investments in order to have enough time and energy for care work and to avoid work-to-family conflict, men are more likely to increase their work investments in order to secure the family income in line with the ideal of the male breadwinner (Bianchi & Milkie, 2010;Budig & England, 2001;Correll et al., 2007;Dechant & Blossfeld, 2015;Grunow & Evertsson, 2016).However, it is unclear whether women use flexible working as an additional or an alternative resource to reduce the likelihood of work-to-family conflicts after transition to parenthood.It is also unclear whether men use flexible working to reduce work-to-family conflict in line with the new ideal of active fatherhood (Bünning, 2015;Bünning & Pollmann-Schult, 2016;Hobson & Fahlén, 2009), or whether they use it to realise higher work demands, as indicated in the gendered transitions to parenthood. 1 The term "flexible working" is also used to refer to the flexibilisation of work contracts and employment relationships as a result of increases in temporary work, atypical forms of employment, and discontinuous work life courses in the post-Fordist era (Breen, 1997;Piore & Sabel, 1985).These aspects are not the subject of the present paper.2 More far-reaching freedom to control working hours and work schedules, which is also referred to as working-time autonomy, is not the subject of the present paper, because, in contrast to flexitime, it is provided mostly for performance-enhancement purposes rather than for the purpose of enhancing the family-friendliness of the organisation (Chung & van der Lippe, 2020).In addition, Chung and van der Lippe (2020) noted that flexible working can also entail workers having control over the number of hours they work, as in the case of part-time working.However, as such arrangements are less likely to provide control over working hours during a regular work week in order to respond not only to predictable but also to unpredictable family demands, they, too, are not within the scope of this paper.
In addition to adopting a gendered life course perspective, the present paper also contributes to existing research by taking an organisational change perspective, investigating the extent to which the normalisation of flexible working in organisations contributes to organisational change towards more family-friendliness, and thus to a general decrease in employees' work-to-family conflicts.Previous research points to increased pressures on organisations to offer flexible working as a family-friendly workplace policy against the backdrop of the increase in dual-earner families and expectations from policymakers to be more familyfriendly (Den Dulk, 2001).Viewed through the lens of policy feedback theory (Campbell, 2012;Gangl & Ziefle, 2015;Grunow et al. 2018), the provision of family-friendly work policies such as flexitime and flexiplace by employers might contribute to setting a more general norm of family-friendliness in organisations, and thus to more family-friendly structures and practices that also benefit those who cannot 3 or do not want to avail of more flexibility in the time and place of work.In line with this argument, existing research points to the emergence of family-friendly workplaces where colleagues, supervisors, and management show understanding and support for employees' family situations (Den Dulk, 2001;Thompson et al., 1999;Van der Lippe & Lippenyi, 2020).
To answer our research questions, we use data for employees in large organisations in Germany that are drawn from a unique dataset, the B3 Linked Employer-Employee Panel (LEEP-B3; Diewald et al., 2015).Individual fixed effects analyses are applied not only to depict transitions to parenthood and changes in the spread of the use of flexitime and flexiplace arrangements in organisations, but also to control for selection into different occupations and organisations that might contribute to spurious associations between the use of flexible working and work-to-family-conflict.Most research on the consequences of flexiplace or flexitime policies for work-to-family conflict has been cross-sectional, leaving little room for controlling for these selection dynamics (for reviews, see Allen et al., 2015;Bianchi & Milkie, 2010;Byron et al., 2005;Gajendran & Harrison, 2007).A recent review article on work-family conflict by Molina (2021) concluded that "the complete lack of longitudinal data makes it impossible to derive causal effects, with only correlations being possible" (p.4).
Studying the consequences of the use of flexiplace and flexitime arrangements and its normalisation in organisations for work-to-family conflict in Germany is especially interesting because, until early this century, Germany adhered to the traditional male breadwinner model of family life, and the use of flexible working was not widespread (Chung & van der Lippe, 2020;Schober, 2014).Moreover, part-time work is a common work arrangement among mothers (Schober, 2014).However, over the past two decades, statutory provisions and family policies have been introduced that increasingly reflect the dual-earner, dual-carer family model in which both parents are employed and both participate in care work.Moreover, the use of flexitime and flexiplace arrangements has increased, especially during the COVID-19 pandemic (Arntz et al., 2020).

Work-to-family conflict and transition to parenthood
The well-established concept of work-family conflict describes an inter-role conflict where demands in the domains of paid work and family interfere with each other (Frone et al., 1997;Greenhaus & Beutell, 1985;Kahn et al., 1964).Time-based work-to-family conflict, which is the focus of the present study, occurs when time pressure at work prevents workers from meeting their own expectations in the family domain.The job demands-resources model (Bakker & Demerouti, 2007;Bakker et al., 2004;Demerouti et al., 2001) suggests that high demands at work increase an employee's risk of experiencing work-to-family conflicts.However, time-based work-to-family conflicts are especially likely when workers with high work demands are confronted with high family demands.Existing research investigating work-family conflicts in different family life stages shows that employees with school-age and preschool-age children experience the largest work-family conflicts (Allen & Finkelstein, 2014;Erickson et al., 2010).This line of research suggests that work-to-family conflicts increase with transition to parenthood.Thus, the following hypothesis is proposed: From a gendered life course perspective (Elder et al., 2003;Moen, 2011), the transition to parenthood means that a gendered adaptation of the work life course also plays a role in whether and to what extent work-to-family conflicts are experienced after transition.More specifically, existing research points to a retraditionalisation of the gendered division of labour whereby women have the main responsibility for care work and try to avoid work-to-family conflicts by interrupting paid work, reducing their working hours, or switching to less demanding occupations (Abendroth et al., 2015;Bianchi & Milkie, 2010;Budig & England, 2001;Correll et al., 2007;Dechant & Blossfeld, 2015;Grunow & Evertsson, 2016;Rüling, 2007;Stone & Lovejoy, 2019).By contrast, once they become fathers, men rarely work fewer hours or switch to less demanding jobs, but rather are more likely to increase the amount of time and energy they invest in work (Fagan, 2004;Lewis et al., 2008).Research on gendered boundary management strategies additionally suggests that, in line with the gendered division of labour and with norms of intense mothering and male breadwinning, women establish stronger boundaries around family and men establish stronger boundaries around work (Pleck, 1979;Shockley et al., 2017).Following from this, it can be expected that men experience even more work-to-family conflicts than women after transition to parenthood.This is because (a) women reduce the amount of time and energy invested in work and erect strong mental barriers around family life in order to reduce work-to-family conflict, and (b) men invest more time and energy in work in order to secure the financial situation of the family.Thus, the following hypothesis is proposed: Hypothesis 1b: The expectation that the transition to parenthood increases time-based work-to-family conflicts is more likely to hold for men than for women.

The use of flexitime and flexiplace and the consequences of transition to parenthood for work-to-family conflicts
Resource perspectives suggest that flexible working arrangements such as flexitime and flexiplace can help employees to reorganise their work in order to make it more compatible with family obligations.Because such arrangements give employees control over when and where they work, they enable them to respond to family demands, be they everyday activities (e.g., driving children to sport training) or unpredictable occurrences (e.g., a child's illness; Abendroth & Den Dulk 2011;Bakker & Demerouti, 2007;Hill et al., 2010;Frone et al., 1997;Schieman & Glavin, 2008).This line of reasoning suggests that flexible working can be used as an alternative rather than an additional resource to decrease the likelihood of higher work-tofamily conflicts after transition to parenthood.In line with this assumption, women who use flexible working arrangements have been found to manage to stay in full-time employment and in human-capitalintensive jobs after childbirth (Chung & van der Horst, 2018;Fuller & Hirsh, 2018;Piasna & Plagnol, 2017) or to re-enter the labour market (Lott 2019).This suggests that women use flexiplace and flexitime arrangements to reduce work-to-family conflicts and sustain their work life course despite transition to parenthood.As a consequence, it can be expected that the use of flexitime and flexiplace decreases the likelihood that work-life conflicts will increase after transition to parenthood, even when work investments are not fundamentally adapted to the new family situation.The aforementioned resource perspectives further suggest that flexible working reduces work-to-family conflicts for men after transition to parenthood.The following hypotheses are thus proposed: Hypothesis 2a: The use of flexible working arrangements (flexitime and flexiplace) reduces the likelihood that transition to motherhood will increase time-based work-to-family conflicts.
Hypothesis 2b: The use of flexible working arrangements (flexitime and flexiplace) reduces time-based work-to-family conflicts after transition to fatherhood.
The demand perspective on flexible working, however, suggests that flexible working is not only a resource but also involves higher work demands which out weight the conflict reducing consequences described above reinforcing increases rather than decreases in work-to-family conflicts after transition to parenthood.Boundary management perspectives argue that the use of flexitime and flexiplace increases the likelihood that work will invade the family sphere because the timing and location of work and family life are not clearly established or separated (Allen et al. 2015;Clark 2000;Kossek et al., 2006).Social exchange theory has been used to describe the gift exchange dynamic through which an increase in work intensity may occur: "Workers work harder to reciprocate for the gift of control over their work their employers have given them; or because workers are better able to work harder and longer due to being able to work when they want; or because of employer enforcement of work intensity through the back door" (Chung, 2019, p. 25).From a life course perspective, flexitime and flexiplace arrangements might also be used to realise high work demands despite family obligations, to signal career ambition despite family responsibilities, and to secure pay and career mobility.This is supported by existing research showing that flexible working is also used realise high work demands (Abendroth & Reimann 2018;Lott, 2019;Lott & Chung, 2016).In line with all three theoretical perspectives, previous -albeit mainly cross-sectional -research has often revealed that flexible working is associated with increased work-to-family conflicts (for a review see Chung & van der Lippe, 2020).The demand perspective leads us to propose the following hypotheses: Hypothesis 3a: The use of flexible working arrangements (flexitime and flexiplace) increases the likelihood that transition to motherhood will increase time-based work-to-family conflicts.
Hypothesis 3b: The use of flexible working arrangements (flexitime and flexiplace) increases time-based work-to-family conflicts after transition to fatherhood.

Normalisation of the use of flexitime and flexiplace and its implications for work-to-family conflicts
The use of flexible working is not only a matter of individual choice.It is also a matter of feasibility and availability.Whether flexible working is feasible depends on the job in question; its availability depends on the organisational context, which determines whether the use of flexible working to manage transition to parenthood is perceived to be a legitimate interest of employees.Work organisations have a long tradition of the ideal worker norm whereby presence at the regular work premises and during regular working hours as well as high availability for work are expected and rewarded, thus imposing organisational constraints on the use of flexible working (Acker, 1990;Kelly et al., 2010;Williams et al., 2013).Indeed, studies show that employees express concerns that the use of flexible workplace arrangements would negatively impact their careers (Konrad & Yang, 2012) or report that their physical presence is expected by their supervisors (Lott & Abendroth 2020).Moreover, existing studies show that flexible working is rather used to realize flexibility interests of employers rather than employees as indicated in the demand perspective on flexible working (Chung 2019).
However, today's more diverse workforce, which includes employees with considerable family responsibilities, as well as state policies aimed at facilitating the integration of work and family life for both mothers and fathers have increased the normative and economic pressures on work organisations to offer flexible working as a family-friendly workplace arrangement (Davis & Kalleberg, 2006;Den Dulk, 2001;Frodermann et al., 2018;Wiß, 2017).In line with neo-institutional theory (DiMaggio & Powell, 1983;Meyer & Rowan, 1977), the expectations of, for example, policymakers, the public, and the media create normative pressures to offer family-friendly workplace arrangements to which work organisations may respond in order to gain or secure legitimacy in their organisational environment (e.g., Den Dulk, 2001;Frodermann et al., 2018;Wiß, 2017).In line with rational choice theory (e.g., Coleman, 1990) and business case arguments (Den Dulk, 2001), economic pressures on work organisations to be more family-friendly can come from a more diverse workforce with employees who increasingly need and want to align work and family demands.Here, employer investments in work-family supportive workplace arrangements are an investment both in the employability of their workforce and in their attractiveness as an employer (Den Dulk, 2001;Kanji & Samuel, 2017;Smithson & Stokoe, 2005;Iseke & Pull, 2019).In this context of increased normative and economic pressures, flexitime and flexiplace arrangements are increasingly provided -especially in large and diverse organisations -to enable employees to better align work and family demands (Chung & van der Lippe, 2020;Den Dulk, 2001;Frodermann et al., 2018).Increases in flexitime and flexiplace thus might indicate applicability and responsiveness to economic and normative pressures to be more family-friendly and in turn a decrease in work-to-family conflicts.
Applying policy feedback theory (Campbell, 2012;Gangl & Ziefle, 2015;Grunow et al. 2018) to the organisational context, the implementation of family-friendly policies such as flexitime and flexiplace in organisations should additionally contribute to gradual changes in the ideal worker norm.The norm-setting mechanism in policy feedback theory describes how state policies, for example family policies that encourage dual earner-dual carer families, have a broader norm-setting function (Gangl & Ziefle, 2015;Grunow et al., 2018).As family-friendly policies in organisations, flexitime and flexiplace arrangements can also contribute to setting a broader norm whereby the ideal worker is not necessarily always present at the regular work site, is not always available for work, and may also have family responsibilities.Increased use of flexitime and flexiplace arrangements might signal to colleagues and supervisors that the organisation supports the integration of work and family, thereby increasing both the feeling of being entitled to align work with family demands and the perceived legitimacy to use flexible working as a resource to better align work and family.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).Increasing experience with the use of flexiplace and flexitime might additionally undermine the flexibility stigma, 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, 2018;Leslie et al., 2012;Munsch, 2016;Williams et al., 2013).Moreover, the use of flexiplace and flexitime as family-friendly workplace arrangements by others in the workplace might serve workers whose jobs cannot be performed independently of time and place as a basis for claiming equal rights and entitlements to better align work and family (on the mechanism for claiming equal rights, see Tomaskovic-Devey & Avent-Holt, 2019; for results regarding career negotiations, see Bowles et al., 2019).Following this line of reasoning, increased use of flexiplace and flexitime might gradually normalise deviations from the ideal worker norm and institutionalise family-friendly work structures and practices in the long run.Organisational change towards family-friendliness also indicates to fathers that they do not need to use flexitime and flexiplace as a signal of high career ambition, but rather can use it in line with the emerging ideal of active fatherhood.Overall, this leads to the following hypothesis: Hypothesis 4: Increased use of flexible working arrangements (flexitime and flexiplace) in a work organisation decreases the likelihood of work-to-family conflict irrespective of whether the individual uses these arrangements.

Data
The data analyses are based on three waves (2012/13, 2014/15, 2018/19) of the B3 German Linked Employer-Employee Panel Survey (LEEP-B3: https://doi.org/10.4119/unibi/sfb882.2014.12).Data collection was conducted in cooperation with the Institute for Employment Research (IAB) in Nuremberg (Diewald et al. 2015) and includes refreshment samples in each additional survey year (Marx et al. 2020).LEEP-B3 includes information on organisations and their employees (including administrative data for both groups) as well as on employees' partners or spouses.Work organisations were randomly chosen from administrative records, and employees were then randomly selected from these organisations.If respondents gave their consent, their work histories were added to the survey data.The final dataset is representative of workers who are liable to social security contributions and who are employed in large work organisations in Germany with at least 500 employees (Diewald et al., 2015).The hierarchical structure of the data allows combining information from three different levels, with observations clustered in respondents and respondents clustered in work organisations.Considering the research aim, a sample of respondents with a partner, at least two observations, and no missing information on the measurements used was selected.The final sample used comprised 4,795 observations of 1,973 men and 3,793 observations of 1575 women in 132 work organisations.

Time-based work-to-family conflict
Time-based work-to-family conflict was measured using the Time-Based Work Interference with Family subscale of the Work-Family Conflict Scale developed by Carlson et al. (2000).The original Englishlanguage scale was translated into German for the purposes of the LEEP-B3 survey.The subscale measuring time-based work-to-family conflict comprises three items (e.g., "I have to miss family activities due to the amount of time I must spend on work responsibilities.")answered on a scale ranging from 0 (never) to 5 (very often).The mean score of the three items was calculated, with higher values indicating higher levels of conflict.The internal consistency of all three items was good; Cronbach's α exceeded the conventional level of acceptance (α = 0.84).

Transition to parenthood
A dummy variable was used in the analysis to indicate a transition to parenthood between the survey waves with 0 indicating that the respondent has no children and 1 indicating that the respondent has at least one biological, adopted, foster-, or stepchild.To create the dummy variable, information on having biological, adopted, foster-, or stepchildren and their birthday at the first observation and information on the birth of a child since the last observation in the subsequent waves was used 4 .

Use of flexiplace and flexitime
Employees responses on the use of flexible starting and finishing times and on working from home were used to measure the use of flexitime and flexiplace arrangements respectively 5 .

Normalisation of the use of flexiplace and flexitime in the organisation
At the organisational level, individual information on the use of flexitime and flexiplace was aggregated to depict the share of the workforce that used flexitime and flexiplace in the work organisation.The respective extent of the use of flexitime and flexiplace in the organisation was significantly and positively correlated (r = .427).

Work investments
The indicators total working hours, work experience, tenure and occupational status (ISEI values calculated based on four digit occupational classification ISCO-08) were used to investigate whether the association between transition to parenthood and time-based work-to-family conflict could be explained by adjustments to work investments.Table A1 and A2 in the Online Appendix provide descriptive information on all measurements once for women and once for men.Moreover, sensitivity analyses with hourly pay instead of occupational status are provided in in Table B in the Online Appendix.The major conclusions remain stable.Interestingly, changes in occupational status matter for the experiences of work-to-family conflicts but not the change in hourly pay.

Survey year
A categorical variable identifies the survey years; the first survey in 2012/13 is the reference category.Comparing Wave 1 (2012/13), Wave 2 (2014/15), and Wave 3 (2018/19), this variable allowed us to determine whether there was a general trend of decreasing time-based work-to-family conflicts over time.

Partner information
To keep the number of observations high, information on the employment status and working hours of the respondent's partner was added only in the additional sensitivity analysis (see Online Appendix, Table and  Figure C).These additional sensitivity tests did neither changed the results on the gendered implications of transitions to parenthood for work-to-family conflicts nor the results on the moderating role of flexible working.

Method
Individual fixed effects analyses with robust standard errors were used that allowed to control for selection into different jobs and organisations as well as for other unobserved individual heterogeneity.They also 4 The xttrans command in Stata reveals 20% of transitions to motherhood and 16% of transitions to fatherhood. 5 The xttrans command in Stata reveals that 9% of women and 14% of men started to use flexiplace in the observation period.It further shows that 20% of women and 37% of men started to use flexitime in the observation period.
allowed to examine transitions to parenthood rather than the general importance of different family life stages for time-based work-to-family-conflict.In a first step, individual fixed effects regression models were estimated separately for women and men to investigate the gender-specific consequences of transition to parenthood for the experience of time-based work-to-family conflict (see Table 1).Model 1 included only the variable transition to parenthood and the control variables age of the youngest child and survey year.It was used to estimate the overall consequences of transition to parenthood for time-based work-to-family conflict.Model 2 additionally incorporated information on work investments to identify whether changes in the frequency of time-based work-tofamily conflict after transition to parenthood could partly be explained by changes in work investments.
In a second step, individual fixed effects regression models were estimated separately for women and men to investigate how the use of flexible working shapes the consequences of transition to parenthood for time-based work-to-family conflict (see Table 2).Model 1 included the variables use of flexiplace and flexitime as well as transitions to parenthood, age of the youngest child, and survey year.Model 2 additionally incorporated interactions between transition to parenthood and the use of flexiplace and flexitime.In addition to the variables included in Model 2, Model 3 incorporated information on work investments in order to determine whether the moderation of time-based work-to-family conflicts by the use of flexitime and flexiplace was explained by differences in work investments among users and non-users of these forms of flexible working.
And finally in a third step, individual fixed effects regression models were estimated separately for women and men to investigate the consequences of the normalisation of the use of flexitime and flexiplace in the work organisation for the experience of time-based work-to-family conflict (Table 3).Model 1 included indicators for the extent of the use of flexitime and flexiplace in the work organisation and examined whether an increase in the use of these arrangements in the organisation decreased time-based work-to-family conflicts for all employees.Model 2 examined whether this effect existed irrespective of whether the individuals themselves used these arrangements or not.Model 3 additionally incorporated the survey year, family situation and work investments to investigate whether the effect of increases in the use of flexitime and flexiplace depicts a general trend of large organizations becoming more family-friendly rather than policy feedback mechanisms.Model 4 added information on the family situation and work investments in order to investigate whether the implications of the normalization of flexible working for work-to-family conflicts are mediated by adjustments of work investments and family formation.To additionally check for influential sectors and occupations every sector (see Online Appendix Tables D1-D4 and Figure D) and every occupation (see Online Appendix Tables E1-E4 and Figure E) was deleted once from the analysis (Jackknife procedure; Rodgers, 1999).

Results
Table 1 shows the results of the individual fixed effects regression models on changes in the frequency of time-based work-to-family conflicts after transition to parenthood.Model 1 reveal that time-based work-tofamily conflicts increased for men after transition to parenthood but decreased for women.Thus, hypothesis 1a, which stated that transition to parenthood increases time-based work-to-family conflict, is only partly supported 6 .Model 2 reveals that the decrease in women's time-based work-to-family conflict after transition to parenthood can be explained partly by reductions in work investments, especially by working fewer hours.This provides evidence for hypothesis 1b, which predicted that work-to-family conflicts were more likely to increase for men than women after transition to parenthood.

6
The sensitivity analyses provided in the Online Appendix (Tables D1-D2, E1-2 and Figure D and E) indicate that this result is relatively stable.Only when sector 4 (Social, private and public services) or occupation 3 (Technicians and associate professionals) are excluded from the analysis the negative effect of transitions to parenthood for work-to-family conflicts among women reduces in size and significance.This can, however, also be driven by the large reduction in sample size because many women seem to be employed in this sector and occupation.Table 2 presents the results of the individual fixed effects regression models examining the moderating effects of the use of flexitime and flexiplace on the consequences of transition to parenthood for time-based work-to-family conflict revealed in Table 1.Model 1 shows that whereas the use of flexitime contributed to a reduction in time-based work-to-family conflicts for both women and men, the use of flexiplace did not.However, for men, the effect was significant only with p < .1.Model 2 shows that flexitime additionally shaped the consequences of transition to parenthood, but only for women 7 .Figure 1 shows that time-based work-to-family conflicts have not decreased to the same extent after transition to parenthood for women who used flexitime in comparison to women who did not use flexitime.Nevertheless, work-to-family conflicts have not increased for them either.Work-to-family conflicts instead remained rather stable despite of the transition to parenthood among those who used flexitime around the transition to parenthood. Figure 1 also shows that the missing decrease in work-to-family conflicts after transitions to parenthood among those who used flexitime around the transition to parenthood can partly be attributed to their weaker workto-family conflicts already before the transition to parenthood.Moreover, Model 3 shows that the larger decrease in work-to-family conflicts among those who did not use flexitime around the transitions to parenthood can be attributed to their more pronounced reductions in work investments around transitions to parenthood.Thus, some evidence is provided for hypothesis 2a, which predicted that the use of flexible 7 This moderation remains relatively stable in the sensitivity analyses in the Online Appendix.However, significance levels also seem to depend on the included sector and occupation and the involved sample size.
working would reduce the likelihood that transition to motherhood would increase time-based work-tofamily conflicts.Cross-sectional analyses would have only allowed for the conclusion that work-to-family conflicts do not differ between mothers who use or do not use flexitime which is also shown in Figure 1.No evidence was found for hypothesis 3a, which was formulated from a demand perspective and predicted that the use of flexitime or flexiplace would increase the likelihood that transition to motherhood would increase time-based work-to-family conflicts.This, however, only indicates that demands involved in flexible working do not out weight that flexible working is a resource to better align work and family.Increasing demands due to flexible working might still be existent as the analyses do not allow to disentangle both dynamics.
The results for men show no significant moderation of the consequences of transition to parenthood for time-based work-to-family conflict by the use of flexitime or flexiplace.Thus, the results show no evidence in support of hypothesis 2b, which predicted that flexible working would reduce time-based work-to-family conflicts after transition to fatherhood, or of hypothesis 3b, which predicted that flexible working would increase time-based work-to-family conflicts after transition to fatherhood.Table 3A and B show the results of the individual fixed effects regression models examining the normalisation of the use of flexitime and flexiplace in the organisation and its consequences for work-tofamily conflict once for women and once for men.The results provided indicate that increases in the use of flexiplace (p<.1) but not flexitime decreased time-based work-to-family conflicts for women (Model 1 in Table 3A).However, as already indicated in Table 2 and Figure 1, the frequency of time-based work-tofamily conflicts was generally lower in work organisations where flexitime was available.Moreover, the descriptive shows that they are already a much more common work practice.Model 2 showed that the effect of the normalization of flexiplace was partly explained by the individual use of flexible working.Including survey year in Model 3 revealed that the conflict-reducing effect of the normalisation of flexible working was, however, not also part of a general trend over time.Instead the effect of increases in use of flexiplace gained significance in Model 3. Including family situation and work investments in Model 4 indicated that the conflict-reducing effect of the normalisation of flexible working was instead partly mediated by changes in work investments (mainly reductions in overtime hours) 8.
The results provided for men in Table 3B indicate that increases in the use of flexitime (p < 0.05) decreased time-based work-to-family conflicts for men (Model 1).This effect seems in parts be mediated by the individual use of this work arrangement (Model 2) and seems to be part of a general time trend (Model 3).Model 4 indicates that this time trend also relates to changes in work investments (work experience and overtime hours). 98 The sensitivity analyses in Table D3 and E3 in the Online Appendix showed that the results provided are highly dependent on the sectors and occupations included.Sector 2 (Wholesale, retail trade, transport) and 4 seem to have suppressed (Social, private and public services) the conflict reducing effect of the normalization of flexiplace.Moreover, especially dynamics in occupation 2 (Professionals) and occupation 4 (Clerical support workers) seem to have caused the conflict reducing effect of the normalization of flexiplace.9 The sensitivity analyses in Tables D4 And E4 in The Online Appendix revealed also for men some evidence for the conflict reducing effect of the normalization of flexiplace in Model 2 (p<.1).Sector 1 (Manufacturing, Electricity, Water supply) and occupation 7 (Craft and related trades workers) seemed to have suppressed this conflict reducing effect.The sensitivity analyses further showed that the conflict reducing effect of the normalization of flexitime is highly dependent on the sector and occupation included.A weakly significant effect (p<.1) irrespectively of individual use only remains in Model 2 when all sectors are included Thus, only partial support was found for hypothesis 4, which stated that increases in the use of flexiplace and/or flexitime in the work organisation decrease the likelihood of work-to-family conflict irrespective of the individual use of these arrangements.D4).Moreover, weakly significant effect (p<.1) irrespectively of individual use only remained in Model 2 when occupation 6 (Skilled agricultural, forestry and fishery workers), 8 (Plant and machine operators, and assemblers) or 9 (Elementary occupations) were not included.One explanation could be that in these sectors and occupations flexitime and flexiplace are less feasible and workers hold less power to negotiate flexible working.

Discussion
Flexible working arrangements that give employees control over when and where they work have been described as a resource for better aligning demands in the domains of paid work and family against the backdrop of the rise in dual-earner couples.However, demand perspectives suggest that flexible working additionally involves new work demands and thus increases rather than reduces work-to-family conflict.Previous research on the consequences of the use of flexitime and flexiplace for work-to-family conflict has been mainly cross-sectional, and its findings have been inconsistent.Contributing a gendered life course and organisational change perspective to existing research, the present study investigated how flexible working in the form of the use of flexitime and flexiplace moderates the consequences of transition to parenthood for time-based work-to-family conflicts among partnered men and women, and whether the normalisation of the use of flexible working in organisations contributes to organisational change towards more family-friendliness, thereby reducing work-to-family conflicts for all employees, irrespective of whether they use these arrangements or not.
Based on analyses of three waves of linked employer-employee data of coupled employees in large work organisations in Germany, I conclude, first, that the consequences of transition to parenthood for timebased work-to-family conflicts are gendered.The frequency of time-based work-to-family conflicts increased after transition to parenthood for men but decreased for women.The fact that this is mainly explained by reductions in women's work investments -especially reductions in their working hours -is in line with existing research revealing gendered adaptations of work to parenthood with a retraditionalisation of the gendered division of labour (e.g., Grunow & Evertsson, 2016).
Second, the results show that flexitime functions as a resource to reduce work-to-family conflicts among women but not men.The study of Lott (2019) provides the explanation that men are more likely to realize high work demands with flexible working than women which might have absorbed the conflict reducing consequences for them.Flexiplace instead shows no significant association with time-based work-to-family conflicts neither for women nor men.It seems as if flexiplace is more often not only a resource but also involves new work demands as indicated in the demand perspective (see also Chung & Van der Lippe 2020 with a similar conclusion).Nevertheless, the longitudinal design provides no evidence for the argument that flexible working increases work-to-family conflicts due to work intensification (Chung 2019) or increasing difficulties to manage the boundaries between the work and the family sphere (Clark 2000).
Third, the results allow the conclusion that women use flexitime to deviate from the traditional strategy of adapting to parenthood by reducing work investments.Flexitime seems to be not an additional but an alternative resource to decrease the likelihood of more frequent time-based work-to-family conflicts after transition to parenthood among women.In this case, reductions in work investments that additionally reduced time-based work-to-family conflicts were less common.This is in line with existing research showing that family-friendly workplace arrangements enable women to stay in full-time employment and in human-capital-intensive jobs (Chung & van der Horst, 2018;Fuller & Hirsh, 2018;Lott 2019;Piasna & Plagnol, 2017).Moreover, women who used flexitime around the transition to parenthood already experienced fewer work-to-family conflicts before the transition.Thus, it might have been less necessary for them to establish stronger (mental) boundaries around their family life after the transition to parenthood indicated in gendered boundary management strategies (Pleck, 1979;Shockley et al., 2017).It seems as if these women work in more family-friendly work organizations not only making flexible working more available but being in general more family-friendly.This argument offers an additional explanation for the steeper decrease in work-to-family conflicts after transition to parenthood among women who did not use flexitime.Experiencing larger work-to-family conflicts before the transition to parenthood, these women had to install stronger mental boundaries after the transition to adhere to norms of intense mothering and to protect family-life of being invaded by the work sphere ending up with similar levels of work-to-family conflicts as mothers who used flexitime.Thus, from a cross-sectional perspective we can conclude that time-based work-to-family conflict does not differ between mothers who use and do not use flexitime.However, weather parenthood fundamentally alters women's work life courses and involved work-to-family conflicts seems to be partly dependent from the provision of flexitime and the family-friendliness of the organization which would have been invisible in a cross-sectional design on the use of flexible working among parents.
That the use of flexiplace does not appear to have contributed to a reduction in women's and men's work-to-family conflicts after transition to parenthood might be explained by the fact that it was less commonly used in the years under study (2012)(2013)(2014)(2015)(2016)(2017)(2018)(2019), which in turn makes it difficult to detect a significant moderation in fixed effects analyses.Moreover, the use of flexiplace often seems to go hand in hand with the use flexitime making it difficult to disentangle the implications of flexiplace from the more common flexitime arrangement.Alternatively, boundary management is more difficult for flexiplace than flexitime or involves more work intensification in return (Chung & Van der Lippe 2020).
The results for men show that neither flexitime nor flexiplace significantly moderated the consequences of transition to parenthood for time-based work-to-family conflicts.It could be that resource and demand dynamics outweigh each other as interests in active fatherhood contradict with expectations of being the ideal worker in the organization.This is in line with research showing that men use flexible working to realize high work demands (Lott 2019).
Third, it can be concluded that the normalisation of flexible working in organisations depicts organizational change towards more family-friendly structures and practices.The results showed that the normalization of flexitime and flexiplace in parts contributed to reduced work-to-family conflicts.The conflict reducing effect of the normalization of flexible working was attributed to the applicability of and responsiveness to normative and economic pressures to be more family-friendly in the background of the rise in dual earner families and political change towards the support of dual earner dual carer families.However, only weak support is provided for the application of policy feedback theory suggesting that the implementation of policies aimed at supporting the integration of work and family contributes to setting a more general norm of family-friendliness that undermines the norm of the ideal worker and contributes to more family-friendly structures and practices.The conflict reducing effect instead was explained by individual use of flexible working and changing work investments.It might be that societal change has increased normative and economic pressures on large German work organisations to be more familyfriendly, thereby contributing to a decrease in work-to-family conflicts and an increase in the use of flexible working without these processes influencing each other.Reductions in overtime hours seem to be part of this general trend but not in all organizations.That the importance of the normalization of flexitime and flexiplace highly depends on the sectors and occupations included indicates that the experience of normative and economic pressures to become more family-friendly largely varies among them.The same applies to conflicting pressures resulting from globalization and market volatility increasing the likelihood that flexible working is used in the flexibility interests of employers (Chung 2019).
The present study also has some limitations.First, the conclusions made are valid only for partnered men and women in large work organisations in Germany.Additional research is needed to test whether these conclusions apply also to small and medium-size work organisations and to other national contexts, where the normative and economic pressures pushing organisations to offer family-friendly workplace arrangements are likely to differ.Second, this research was not able to take the intensity of flexible working into account.It thus remains an open question whether the intensity of flexible working is even more important for work-to-family conflicts after transitions to parenthood than the general use.Finally, future research is required to investigate the rationales that drive the decision to use or not to use flexible working, as well as the dynamics of organisational change.
Overall, this study holds important implications for family research.Flexitime is of direct importance for family well-being as it contributes to reduced time-based work-to-family conflict at least among women.Flexi-time seems to allow women to respond to predictable and unpredictable family needs decreasing the likelihood of time-based work-to-family conflicts (Abendroth & Den Dulk 2011;Hill et al., 2010).Thus, parenting warmth and consistency in the parent-child relationship can be sustained despite of work obligations (Verweij, Helmerhorst & Keizer 2021;Cooklin et al. 2014).In the same vein, flexi-time seems to be beneficial for relationship quality among couples as it can be used to better synchronize daily working times facilitating joined activities and decreasing time-based work-to-family conflicts (Cooklin et al. 2014;Byran et al. 2014).In addition, flexitime is of indirect importance for family well-being as it seems to help mothers to sustain their work capacities after transitions to parenthood (see also Chung & Van der Horst 2017) enhancing their satisfaction (Wheatley 2016) which can spill over to other family members (Edwards & Rothbard 2000).Finally, the selection into different organizational work contexts seems to be important for time-based work-to-family conflicts and involved family well being.This is in line with existing research highlighting that the family-friendliness of the organizational culture, e.g.available supervisor support or the adherence to the ideal worker norm, is influential for work-to-family conflicts (Abendroth & Reimann 2018;Van der Lippe & Lippenyi, 2020).Thus, more research is required which applies an organizational comparative perspective on family well-being.

Hypothesis 1a :
Transition to parenthood increases time-based work-to-family conflicts.

Figure 1 :
Figure 1: Time-based work-to-family-conflict among women after transition to parenthood: The importance of the use of flexitime

Table 2 :
Individual fixed effect regressions on the use of flexible working and how it shapes gendered consequences of transition to parenthood for time-based work-to-family conflict Note.Source: LEEP-B3, wave 1-3.Standard errors in parentheses.+ p ≤0.10, * p ≤0.05, ** p ≤0.01, *** p ≤0.001

Table 3A :
Individual fixed effects on time-based work-to-family conflicts and the consequences of the normalisation of the use of flexible working among women

Table 3B :
Individual fixed effects on time-based work-to-family conflicts and the consequences of the normalisation of the use of flexible working among men