Norms about childcare, working hours, and fathers’ uptake of parental leave in South Korea

ABSTRACT The number of leave-taking fathers in South Korea has quintupled in the last five years, but those taking leave are still a minority of the total eligible fathers. These minority fathers, however, take some of the longest leave in the world. Driven by such notable trends, this paper inquires how norms about childcare and working hours shape Korean fathers’ decision to take leave as well as work-family balance after leave. I find that fathers are often pushed to take leave as a last resort in light of poor work-family balance and absence of more desirable alternatives to care for a young child. While these conditions continue to constrain parents after the end of the fathers’ leave, fathers respond in divergent ways: making continued effort to balance employment and family life, reverting to work-centric lifestyles with grandparental support, or going through career changes to address continued childcare needs. My findings highlight that incentives targeting fathers to take leave need to go hand in hand with more fundamental reforms to working hours and reliable ECEC to sustainably support gender equality and work-family balance of dual-earner parents.


Introduction
Over the past few decades, a growing body of literature has examined the potential of men's uptake of leave to promote more gender-equal division of paid and unpaid labour and work-family balance by "undoing gender" (Deutsch, 2007;Deutsch & Gaunt, 2020;Risman, 2009;West & Zimmerman, 1987). Considerable research has documented the behavioural and attitudinal changes experienced by fathers in transitioning to a primary caregiver during, and to an extent, after leave (Brandth & Kvande, 2018;Bünning, 2015;Choi et al., 2019;Doucet & McKay, 2020;Kang, 2013;Kim & Kwon, 2015;Na, 2014;Rehel, 2014;Schober, 2014;Tamm, 2019; see also the volume edited by O'Brien and , especially a solo leave beyond a couple of months. It is in this background that governments seeking to promote work-life balance and gender equality at home have introduced policy incentives to encourage fathers to share leave equitably with their partners. Fathers' uptake of leave in South Korea (hereafter Korea) stands out for its huge chasm between the policy developments encouraging fathers' uptake of leave and strong cultural pressure that prevent them from taking it up (Byun & Won, 2020). Namely, Korea is characterised by one of the longest working hours yet one of the most unequal gendered division of unpaid labour in the OECD. In addition, Korean parents' low trust towards its ECEC (early childhood education and care) system creates sharp structural constraints for dual-earning couples to balance employment with childcare. In such a setting, recent statistics suggest that only around two percent of fathers take leave in Korea, despite rapid increases in recent years. Nonetheless, these fathers take some of the longest leave in the world, of over 200 days on average, which far exceeds the first three-months of "daddy months" during which payments are generous (Statistics Korea, 2020c), meaning many of these fathers take extended leaves beyond three months despite social and financial costs.
This paper seeks to explain such unique patterns of Korean fathers' uptake of leave as well as its aftermaths with a focus on how fathers' uptake of leave interacts with the cultural and institutional norms surrounding childcare and employment. Specifically, I aim to answer two questions. First, how do norms around childcare and employment factor in fathers' decision to take leave, often for substantial periods? Second, how do these norms shape fathers' post-leave work-family balance? To address these questions, I position myself within a body of literature emphasising the importance of studying (parental leave) policy not in isolation but in interaction with cultural, institutional, and other policy conditions. I draw on and triangulate two sources of qualitative data, in-depth interviews with fathers and blog posts of parents on their fathers' leave experience, to capture a diversity of family profiles and narratives. In the following sections, I first present a review of the theoretical and empirical literature which serve as the basis for this research. Then I introduce the Korean context and my data and methods in more detail, present my main findings, and end with some discussions and policy implications for gender equality and work-family balance.

Theoretical background and literature review
Why is it important to study fathers' uptake of leave in relation to cultural, institutional, and policy norms about childcare and working hours? The theoretical approach that this paper takes may be located within broader theoretical perspectives in feminist welfare state and family policy scholarship which highlight the importance of studying a single policy measure in relation to broader cultural, institutional, and policy conditions. A notable contribution is Pfau-Effinger's theorisation of "gender arrangement'" (1998,2012) as the interaction between family policies and the wider societal context to study the gendered outcomes of family policies. Pfau-Effinger posits that family policies alone do not explain considerable differences across countries. Instead, gender cultures, defined as the "values, models, and belief systems which relate to the gender relationship of the family to employment and childcare" (2012, p. 533), must be considered in addition to institutional, economic, and social factors when explaining the impact of policies. For instance, Pfau-Effinger points to cultural norms about organising childcare for children under three as a salient factor shaping the employment rates of women with young children (2012).
'Institutional complementarities' (Hall & Soskice, 2001;Thévenon, 2016) is another useful conceptual framework to study family policies in interaction with their institutional backgrounds rather than in isolation (see also Lee and Zaidi, 2020). Thévenon (2016) notes that the influence of a single policy varies across welfare state regimes due to heterogeneities in the context in which the policy is being implemented. If a package of policies is highly complementary to one another, they can create synergies by maximising the positive impact or minimising the negative outcome of one policy measure. Thus, to examine the effectiveness of family policies, they should be examined in bundles or packages to account for the complementarity between the policies of interest. Here, we may understand the idea of a policy package as "a combination of policy measures designed to address one or more policy objectives, created in order to improve the effectiveness of the individual policy measures, and implemented while minimizing possible unintended effects, and/or facilitating interventions' legitimacy and feasibility in order to increase efficiency" (Givoni et al., 2013, p. 3).
I apply such perspectives to Gornick and Meyers's proposal (2008) for a gender-equalising work-family reconciliation policy package to actualise an earner-carer society. First, family leave provisions should entitle parents to the right to care for children in the form of short-term maternity and paternity leave, longer-term parental leave, and more ad hoc family leave. Leave policies should be generous, flexible, non-transferable, and genderegalitarian, granted to both women and men while incentivising men's uptake. Second, working hours should be reduced to enable workers to reallocate their time to caregiving. Standard full-time employment should be limited to under 40 h per week and 48 weeks per year while guaranteeing choice to and fair treatment of part-time employment and flexible working arrangements. Third, high quality and publicly subsidised early childhood education and care (ECEC) should be available for all parents so that they can engage in employment. ECEC services should be provided and financed by the government to equalise access and cost and ECEC schedules should match the parents' working hours. Here, it is important to stress that it is the interaction of childcare leave policies with the broader cultural and institutional norms about employment and childcare which matter for gender-equalising work-family reconciliation.
Extant studies inquiring the influence of fathers' workplace cultures broadly on leave uptake have found that the salience of masculine workplace norms based on assumptions of the ideal worker constrain a father making full use of his right to leave (Beglaubter, 2017;Brandth & Kvande, 2019;Evertsson et al., 2018;Haas & Hwang, 2019;Harvey & Tremblay, 2019;Kim & Kim, 2020;Meil et al., 2019;Moen et al., 2019;Moran & Koslowski, 2019;Närvi, 2019;Reimer, 2020;Stronik, 2019). In the Korean context, fathers who do go on leave expect and/or experience being penalised in promotions or evaluations, which acts as a further deterring factor (Hong, 2018). On the other hand, motivations for fathers taking leave where it is not common to is mainly explained by fathers' individual values, such as commitment to involved fatherhood and/or gender egalitarian ideologies, including fairness in sharing care, supporting their wives' career, or simply spending time with their children (Meil et al., 2017;Nakazato, 2017;O'Brien & Twamley, 2017;Valarino, 2017). In short, these studies suggest that enabled and agentic fathers take leave, while masculine workplaces pose a structural barrier. Such workplace norms also matter for whether the changes to the gendered division of care responsibilities that are often reported during leave will necessarily persist after the leave is over (Wall and Leitão, 2017;. For instance, Miller reports, based on a longitudinal study (2011) of seventeen British men with short paternity leave experience that fathers "slip back into the familiar territory and patterns of the workplace" (p. 144) once their leave finishes, suggesting the difficulty of bringing about fundamental changes in the re-allocation of these fathers' work and care patterns.
Equally important are contributions which focus on how parental leave policy intersects with ECEC services. For instance, Moss (2012) stresses that where parental leave and ECEC policies are conceptually and structurally separated, it becomes difficult to progress seamlessly from parental leave to ECEC service. Mazzucchelli et al. (2019) examine the gap between statutory leave and ECEC in four European countries, finding no gap for Germany and Sweden, a gap of one to two years in the UK, no ECEC entitlements in Italy. They conclude that the Swedish and German systems are conducive to the flexible transition to and redistribution of care duties with their well-aligned parental leave and ECEC provisions, while the British and Italian systems less so. Farstad (2015) offers a study of parental strategies to fill a nine-month of care gap in Iceland between the nine months of paid parental leave (three months of non-transferable leave for each parent plus three months of shared leave) and public daycare provision which typically begins when the child around 18-months. A gendered pattern of parental leave use arises in this process, where it is the mothers who typically stretch their leave, feeling six months is insufficient, while fathers consider three months to be sufficient. This demonstrates how despite a model of parental leave scheme which is designed to promote gender equality, insufficient ECEC provisions for infants contributes to gendered outcomes.
While such prior contributions have examined how workplace cultures and ECEC each relate to fathers' leave to an extent, not enough attention has been paid to how fathers' uptake of parental leave simultaneously interacts with norms about ECEC and working hours. Building on prior theoretical and empirical contributions, this paper responds to the call to conceptualise parental leave within broader structural and cultural contexts (Mazzucchelli et al., 2019) by studying how Korean fathers' uptake of leave and its consequences are simultaneously moderated by norms about childcare and working hours. The most notable features of Korean fathers' pattern of leave uptake are that the uptake rate is very low (at around two percent), yet the average period of leave is around seven months, much longer than 131 days in Sweden, 70 days in Iceland, or 40 days in Norway (NOSOSCO, 2017). Given that only three 'daddy months' are well enumerated in Korea, it is puzzling why Korean fathers often opt to take very long leaves up to the maximum period of one year or even beyond, even though it is not socially or financially conducive for them to do so. In this context, I will inquire how Korean fathers' decision to take leave, often for substantial periods, is shaped by cultural and institutional norms around employment and childcare. I will furthermore examine how these norms continue to shape the work-family balance of these fathers after their leaves come to an end.

The Korean context
Regulations regarding working time, ECEC, and parental leave are all currently undergoing policy reform and changes in Korea. Despite the recent developments, the three intersecting areas of work-family balance policies for parents with young children are still laden with tensions and inadequacy.

Parental leave
Korea's current leave policy can be summed up as flexible and generous in duration but not in income replacement. The eligibility criteria and maximum duration of leave differ depending on whether the parent is a private sector employee or a public servant, as employees are subject to separate legislative, administrative, and financial jurisdictions by sector when taking parental leave. 1 All public servants are eligible for up to three years of parental leave per child and private sector employees who have been employed for more than six months are eligible for up to one year of leave for each child, up to the child being eight years old (Statistics Korea, 2020a). 2 Because private sector employees need to be registered under the national employment insurance to be eligible for paid leave, those who are not eligible for employment insurance cannot receive leave benefits. 3 In total, 72 percent of all fathers (and 34 percent of all mothers) of children born in 2019 are currently eligible for paid leave in Korea (Statistics Korea, 2020a).
A 'father's bonus' was first introduced in 2014, through which the second parent to take leave, the father in 88 percent of the cases (Ministry of Employment and Labour, 2020), could receive up to 1,500,000 won (approximately $1265) in the first month of leave. The payment ceiling and period of coverage have incrementally expanded in the years that followed, allowing fathers today to receive up to 2,500,000 won (approximately $1209) for the first three months if his wife has already been on leave. Fathers then receive up to 1,200,000 won (approximately $1012) in the months that follow the first three months, or during the entire period of leave if they are the first parent to take leave. Even with the introduction of the "daddy months", the average income replacement rate over the total duration of leave is a mere 42.0 percent of previous earnings (OECD, 2021c), low compared to Norway's 95.5 percent, Iceland's 77.9 percent, or Sweden's 75.5 percent. For these reasons, although the number of Korean fathers taking leave has increased by fivefold in the last five years, still only around two percent of all fathers take leave (Statistics Korea, 2020a; see also Table A1 of the appendix).
In 2019, 31,665 fathers took leave in total, of which 70 percent were private sector employees (127,448 in the case of mothers, of which 65 percent were private sector employees) (National Assembly Budget Office, 2019; Statistics Korea, 2020a). While these numbers reflect a dramatic increase in both the public and private sectors as a result of the governments' incentives for fathers to take leave, the uptake rate still amounts to only about two percent of all fathers. Of all leave-takers, 71 percent of fathers (and 63 percent of mothers) were employed in large companies with more than 300 employees, and in contrast only 4 percent of fathers (and 5 percent of mothers) were hired in small companies with four or less employees (Statistics Korea, 2020a). Among parents of children born in 2010, while more than 60 percent of mothers took leave before the child turned one, roughly half of fathers took leave when their child is between six and eight and one-thirds of fathers when their child was three or younger (Statistics Korea, 2020a). Mothers working in the private sector took 9.8 months of leave, from when the child was 0.9 years old on average and fathers took 7.3 months of leave, from when their child was 3.2 years old on average (National Assembly Budget Office, 2019). In short, the majority of fathers taking leave are either public servants or employees of large companies and fathers typically take leave after their wives have done so.

Time spent on paid and unpaid labour
Korea is further characterised by a particularly long work time regime. In 2020, an average Korean spent 1908 hours in paid work, among the longest in the OECD (2020a). Each week, Koreans on average work more than ten hours longer than the British, nine hours longer than the Swedish, and four hours longer than the OECD average. The widespread norm of long working hours, which is based on a collectivist culture which stresses the interest of the group (company) over that of individual employees , undoubtedly puts much strain on family life, making it especially difficult for dual-earner couples with young children.
More specifically, a national survey of childcare found that fathers spend over 59 h in employment and commuting each week (over 48 h of employment and over 11 h of commuting) and mothers over 48 h (over 39 h of employment and nearly 9 h of commuting) . The same survey found children of working mothers between 0 and 5 years old to be spending an average of 7 h and 42 min in ECEC, which was more than one hour short of these parents' desired ECEC use hours of 8 h and 54 min, suggesting a mismatch between parents' working and childcare schedules. While it is technically possible for parents to work reduced hours, only 550 fathers (3,260 in the case of mothers) used this policy in 2018 (National Assembly Budget Office, 2019).
At the same time, Korea is among the countries with the most imbalanced gendered division of unpaid labour in OECD (OECD, 2021e). Time use studies indicate that even in the case of dual-income couples Korean husbands spend 54 min per day on domestic labour while wives spend nearly 3.5 times this amount of 187 min (Statistics Korea, 2020c). The disproportionately high share of unpaid labour that Korean working moms burden frequently results in women's career discontinuities in childbearing ages, represented by an accentuated M-shape curve outlining women's labour force participation rate by age group (OECD, 2021b). Due to this, the ratio of dual-income households is the lowest at around 45 percent when children are in their preschool ages (Statistics Korea, 2020b).
To address the problems of long working hours posing difficulties to maintaining workfamily balance, the Korean government introduced legislation in 2018 to limit private sectors employees' working time to 52 h per week, before which it was legal to work up to 68 h per week. The regulation was first imposed on large private sector companies with 300 or more employees from July 2019 and is expected to be gradually expanded to small and medium-sized companies in 2021. 4 While a reduction of work hours is a welcome change, there exist critiques that it is poorly regulated and ineffective. Put simply, working long hours has for long been an institutionally endorsed norm in Korea, and despite the recent change, there exists some chasm between regulation and culture.

ECEC
Korea has made some significant progress in its ECEC to provide universal "free" childcare services. A notable feature is the universal childcare vouchers which, since 2013, have been provided to all children between 0 and 5 (regardless of the employment status of parents) intended to contribute to covering ECEC expenses. However, much of this progress was made possible by the state outsourcing its duties to market-based service providers, which came with its share of problems, both for kindergartens and childcare centres. 5 Today, roughly 63 percent of children between 0 and 2 and 85 percent of children between 0 and 5 are enrolled in formal ECEC, which is among the highest rate among OECD countries, but only 17 percent of these children are enrolled in public ECEC services (OECD, 2021d;Statistics Korea, 2020d).
The number of ECEC facilities have expanded dramatically over the past several decades, from 1,919 in 1990 to 35,352 in 2020 in the case of childcare centres. 6 The need for a rapid expansion of ECEC service was addressed by relaxing licencing rules, which resulted in the proliferation of private ECEC services owned by individuals, rather than corporations. In addition to directly providing vouchers to the users (parents), the Korean government sought to make childcare affordable by subsidising private ECEC providers a portion of what is subsidised to public ECEC providers. 7 The government subsidy to ECEC providers have been consistently criticised as being insufficient, especially in covering the wages of ECEC teachers; according to Kim (2019), the state needs to increase its support by 43∼62 percent to be consistent with current levels of consumer price inflation and minimum wage.
In this context, private ECEC providers run by individuals have demanded exemption from the centralised accounting monitoring system, based on grounds that the establishment and maintenance of private ECEC services require substantial private investment, properties, and expenses which they need to retrieve (Hwang, 2021;Jang, 2015). The unique position private ECEC providers take up in Korea has provided the basis for many of these providers to circumvent monitoring, and a series of special audits between 2013 and 2017 found 5951 cases of financial irregularities and accounting fraud in 1878 private kindergartens, where fees were used for personal purposes rather than for quality ECEC, creating a huge scandal (Jo, 2018).
In addition, the poor wage and labour conditions for teachers particularly in private childcare have resulted in high turnover rates as well as recurring news about child abuse or neglect at ECEC institutions, further evoking parents' distrust. Hundreds of child abuse cases at centres were reported each year between 2012 and 2015, and surprisingly, 138 of these centres had been accredited as excellent by the Ministry of Health and Welfare (Cho, 2015). In this context, the demand for public childcare always far exceeds supply because it is considered more reliable and better managed, offers higher quality care, often operates longer hours, and is affordable compared to its private counterpart . Parents of young children under six report waiting an average of 7.6 months for a place at public childcare, 9.2 months in the case of those residing in large cities .
In short, while Korea has made impressive quantitative expansions in its ECEC services, its market-mediated means of service delivery have come with ample side effects. These factors together contribute to an oversupply of, but distrust towards particularly private ECEC services as well as a preference for caring for children under three at home. According to a 2016 study conducted by the Ministry of Gender Equality and Family (MOGEF, 2016), more than 75 percent of parents responded that the family should be the sole caregiver for children under three, and an additional 22 percent responded that the family should be the primary caregiver. Another study found that 99 percent of parents believed home care to be the most ideal form of care for children under one and 86 percent responded the same for children between one and two . In the case parental care was not possible, more than 80 percent preferred extended family as an alternative to parental care (Seo et al., 2012), and feelings of anxiety about their child receiving care from strangers was the most cited reason for preferring family support . In response to such problems, the Korean government has recently announced plans to improve the quality of ECEC services by scaling up the number of public providers, evaluating all childcare centres every three years, and mandating private kindergartens to use the state-managed accounting system. However, it is unclear to what extent the expansion in quantity will be able to address the problems of quality of childcare (Chung, 2018).
Taken as a whole, the work-family imbalance of Korean parents with young children has for long been shaped by long work hours, highly gendered division of unpaid labour, and a preference for family care for children under three. At the same time, the various policy institutions governing employment and childcare are currently undergoing significant changes. This makes for a particularly interesting and timely locus to study fathers' uptake of leave in relation to such norms on working hours and childcare.

Data and methodology
This research benefited from the rich utilisation and triangulation of two complementary qualitative data sources: semi-structured qualitative interviews of fathers and posts from online parenting blogs, both where the father had taken leave. The data was collected from May to October 2020 and ethical approval was obtained from the relevant committee at my home institution. Below I discuss in more detail the data collection and analysis processes.
First, I interviewed 51 fathers who had taken leave or were on leave at the time. Participants were recruited through personal networks as well as various online parenting spaces. 21 interviews were conducted in person, 27 on the phone, two through email, and one through Zoom depending on the preferences of the interviewees and constraints posed by the pandemic. The participants were highly diverse in their demographic background as well as how their leave was taken (see Tables A2 and A3 of the appendix). All interviews were audio-recorded with consent, transcribed verbatim, and the participants' names have been anonymised.
In addition, I analysed 49 blogs that documented the family life of dual-earning couples before, during, and after fathers' leave. These blogs were first identified by searching blogs containing the keyword 'fathers' parental leave' on Naver, Korea's largest search engine, portal, and platform for blogs then narrowed down based on their level of relevance and detail. 19 of these blogs were run by moms, 29 by dads, and one jointly by the couple. I searched and read all posts including the relevant keywords as well as posts providing contextual information on the couples' work-family balance and extracted and compiled key posts containing relevant information onto a separate word document. While I use codes in the main text when quoting the blogs for readability, I present the quoted blog names in original Korean, Romanised Korean, and English translation as well as links to their blog posts in Table A4 of the appendix, in line with .
The blog data complement the qualitative interview data in that the two sources of qualitative data have different strengths. While I interviewed just the fathers, the inclusion of blogs run by dads as well as moms provided me with some of the mothers' perspectives. Moreover, while the interviews were conducted at just a single point in time and thus the interview participants had to provide retrospective information for some of the questions, the chosen blogs had multiple posts related to fathers' uptake of leave written at multiple time points. Hence, the blog posts provided snippets of longitudinal documentation of private and intimate life events as well as a reflection about those events (Hookway, 2017), both before, during, and after the fathers' uptake of leave. This type of longitudinal data was particularly important, as I was interested in the full span of time, from when fathers decide to take leave to after the end of their leave. There are some further general benefits that blogs offer to qualitative research, such as unobtrusive, naturalistic, and spontaneous generation of narratives and often photographs and other visual images which accompany the writings (McGannon et al., 2017). Together, blogs offer a rich multi-media archive of everyday life which provide insights into the participants' own language and reflection which is not led by the researcher. On the other hand, while I had to work with the limited and often scant information that was available on the blog posts, interviews allowed me to gather more precise and focused data on the demographic background of the participants as well as to guide the conversation and pose follow-up questions to probe for information directly relevant for the research project.
My analytic approach was simultaneously inductive and deductive; while some themes were guided by research questions or theories, other themes arose from the fathers' narratives. It was also abductive in the sense that I was 'simultaneously puzzling over empirical materials and theoretical literatures' (Schwartz-Shea & Yanow, 2012, p. 27; see also Doucet & McKay, 2020). I particularly referred to Doucet's application of the Listening Guide method (2019) 8 and van Manen's analytic strategy (1990) 9 in reading, coding, and analysing the data through multiple stages. Though I opted to adapt based on what was most suitable for my research rather than following these pre-existing guides, the guides were helpful in that they serve as models of how the researchers may focus on different aspects of the data in each reading. Building on the common idea of multiple readings and a multi-level analysis scheme, I first read both the archived blog posts and transcribed interview transcripts multiple times vertically to extract each fathers' general narrative on motivations, experiences, and aftermaths of leave uptake and to identify and colour-code the main themes which emerged. I created short summaries of each interview, highlighting the most prominent points which stood out for each father. Then the thematically colour-coded materials were coded in more detail to distinguish reasons for taking leave, feelings about returning to work, and balancing work and family life after the end of leave (see Table 1). Finally, I compared horizontally across fathers, paying attention to identifying general patterns, similarities, and differences.

Fathers' reasons for taking leave
First, what circumstances shape fathers' decision to take leave in a context where it is uncommon and culturally inconducive? At the basic and most general level, fathers wanted to spend more time with family and improve their relationship with children. On one hand, this signalled aspirations toward the involved fatherhood ideal as in previous studies (Meil et al., 2017;Nakazato, 2017;O'Brien & Twamley, 2017;Valarino, 2017), but at the same time it also implied that fathers were not able to spend enough family during normal times due to their career-centred lives. Fathers regularly articulated being burnt out and wanting to take some time off from work as a motivation for taking leave. The degree of fatigue from overwork varied by fathers, but many looked back on how they had worked for years without ever having a proper break, and several reflected on how they were too tired and stressed out after a long day at work to be good fathers or husbands. Below I offer two examples of fathers whose main motivation for leave was a mixture of wanting a break from overworking and wanting spend family time. In this sense, the norm of long working hours, a norm of overworking and the resulting fatigue and work-family imbalance were key contextual factors shaping fathers' decisions to take leave, often for long periods. If overwork preventing personal and family wellbeing was one prominent reason for fathers taking leave, another major reasonmainly for fathers who took leave when the child was under threewas parents' preference for caring for children at home. Such a preference was driven by reasons such as disinclination about "strangers" caring for their baby and concerns about child abuse. Parents wanted to start using formal care when the child could communicate their feelings or conditions, in case an accident happens at childcare, as illustrated by the two dad bloggers who took leave right after their wives finished theirs.
"Even if my wife uses her maximum parental leave period of one year, our baby will be around fifteen months old. People don't send their child to preschool at this time unless it's inevitable. Any parent wants to send their child off when the child is capable of basic communication." -Blogger 2 Disadvantage at work 4 Change in employment status 11 Note: These themes are not necessarily mutually exclusive, and in cases where fathers mentioned more than one theme, all themes mentioned were all included in the count.
"Parents who send their one-year-old child to preschool will feel like they are walking on thin ice every day. A child has to be at least three years old to be able to express wanting or not wanting to go to preschool.
[…] It's not easy sending a one-year-old baby off alone these days, with child abuse happening so frequently." -Blogger 3 A related theme which was mentioned regularly albeit less frequently was that it was logistically impossible for a dual-earning couple to coordinate full-time employment and schedules for their child's care or school even if they stuck to 9-6 or 8-5 working hours. This was either because children needed to be picked up earlier and/ or parents did not want to keep young children at care or after-school programmes for prolonged periods. For example, interviewee B, quoted below, had two children aged seven and two at the time of his leave who had each just begun attending primary school and childcare. While it would technically be possible, although not desirable, to keep the two-year-old at childcare the whole day, the interviewee explained that school ends at half past noon in the case of primary school first graders. For such reasons he also started his leave after his wife finished her leave but mentioned towards the end of the interview that he would not have done so if there were viable alternatives.
"If we are both working, we can send the children off [to childcare] but not pick them up on time.
[…] If it were possible for our parents to sacrifice themselves and volunteered to take care of our children, I wouldn't necessarily have had to take leave." -Interviewee B As interviewee B had also suggested, receiving help from the child's grandparents or other extended family was considered necessary and desirable by many dual-working parents. However, fathers who took leave often stated that was not a feasible option for reasons including the grandparents' bad health, sudden injury, old age, distance, employment, or other care obligations. Interviewee C quoted below, is such an example. The interviewee and his wife had to urgently find a way to look after their infant when his mother was suddenly unable to help, and it was interviewee C rather than his wife who ended up applying for a six-month leave for his one-year-old child because taking leave was easier at his company than his wife's. In other cases, the parents either just felt too bad about asking their parents to care for their grandchildren, because it is a lot of work and entails certain sacrifice on their part. "My mother had been coming over on the weekdays to take care of the baby even though she lives far away, but she suddenly injured her leg […] so she couldn't help anymore. The baby was eight months old then and sending a baby that's not yet even one year old to childcare didn't feel right." -Interviewee C In such circumstances, some parents further considered the option of hiring a nanny or sitter or one parent quitting work, but the father going on leave was a more sensible choice from an economic perspective; paying a full-time nanny would require a considerable proportion of one parent's salary and sustaining a family with just the fathers' wage is not realistic. On the other hand, commitment to gender equal parenting was seldom explicitly articulated as a motivation for taking leave. Instead, it was regularly implicitly embedded in fathers wanting to support their wives in childcare or return to work, or at least not wanting their wives to discontinue their careers, though this rationale is also partially based on economic grounds. Hence, the father going on leave seemed to be a more sensible choice, especially given that the fathers could receive a modest stipend during their leave. This is implied in the quote from the post by blogger 4, who describes his choice to take leave as somewhat a "lesser of the evils", made in an absence of better alternatives.
"It was not that I took leave because I go to a 'dream workplace' but it was rather a case of 'even though'. […] Our house didn't have enough space for a resident babysitter, and both our parents lived too far away to receive their help. Ultimately, I had two options. Either I make my wife take the sole responsibility of childcare, or I become a co-carer. The former option was even scarier than applying for parental leave." -Blogger 4 While some fathers, especially those working in the public sector and/or female-dominated occupations such as teaching, described their process of applying for leave as not difficult at all, more often than not fathers faced some level of friction or dissuasion at work. In short, most fathers explained their decision to take leave as a response to difficulties in reconciling employment and family life. For those with young children, this meant a lack of better options to arrange childcare for a young child, oftentimes even after their wives took leave for the maximum possible period. This also explained why most fathers tended to take considerably long leaves; so that they could fill the care gap for as long as feasible and so that the child could be cared at home for as long as possible until deemed "old enough" for formal care. Even for those with older children, this meant a lack of ways to address the gap that emerged between their time at work and the child's time either at childcare or school. More generally, it meant fathers being unable to spend enough time with their family as a result of overwork.

Concerns about returning to dual-earning patterns
Consistent with prior studies, fathers typically described being in charge of childcare during leave as not always easy, but worthwhile. Thanks to their leave, fathers were able to build stronger bonds with their children and realise the importance of sharing responsibilities for unpaid labour with their wives. Importantly, for parents whose children were now "old enough" for full-day childcare or schooling, or those who were able to arrange help from extended families at the end of the fathers' leave, the father taking leave enabled them to get through a critical period of childcare relatively smoothly. These fathers were fairly content with how their leave went in general and looked forward to returning to work.
However, also notable were parents who voiced concerns and anxiety about arranging childcare as dual earners as fathers were nearing the end of their leave. It is particularly important to note that many of these fathers were returning to circumstances of poor work-family balance which drove their leave uptake in the first place. Parents whose children were still under two or three at the end of the fathers' leave expressed continued qualms about enrolling their baby in preschool due to reasons outlined earlier. For those who could not receive help from extended family, the time gap between employment and childcare hours was again a problem. Several bloggers wrote about feelings of despair and frustration because they did not know how to arrange childcare when the fathers' leave ended. As bloggers 5 and 6 illustrate, for these parents, the father taking leave could postpone, but not fundamentally solve their problems in balancing employment and childcare.
"Even if one is courageous enough to take parental leave of one year after his wife, it doesn't solve all problems. Even if the mom and dad go on leave for two years in total, the child continues to need the parents' care and attention. At childcare facilities cases of child abuse resurface now and then." -Blogger 5 "My husband's parental leave of eight months is coming to an end. And I feel stifled. I have no idea what to do with our baby from next month. I don't want to send [child] to preschool. […] The problem is that preschool is from 8 to 3. No company's working hours are from 8 to 3." -(Mom) blogger 6 On the other hand, fathers whose children were "old enough" by the end of the fathers' leave had a different concern, about reverting to their old work-centric lifestyles upon returning to work. Interviewee D is a good example of such a father; he had explained that his main motivation for taking leave was improving his relationship with his two children, aged seven and nine, with whom he was not able to spend enough time during his busy and stressful work life. D was doubtful of whether he could maintain the improved relationship once he became busy, exhausted, and stressed at work again. Blogger 7 expressed similar sentiments, having been in a "weekend couple" arrangementwhere he lived in a different city during the weekdays for work and returned to his family for just the weekendsfor eight years prior to his leave. He emphasised how close fathers' leave brought him and his family together, but at the same time had fears about returning to being a "weekend daddy".
"I am already worried about returning to work, that I may revert [to work-centred lifestyle] when I am busy and stressed again." -Interviewee D "I am scared of returning to work. More than adjusting to work, I am worried that everything I have built up with the children for the past one year will disappear and that I will return to my life before [leave]." -Blogger 7 Fathers with younger and older children had diverging concerns, but it seemed that fathers' leave coming to an end was giving these parents renewed concerns. The renewed concerns centred around the difficulty in balancing family life with employment given long working hours as well as reluctance about sending young children to childcare, especially for extended periods. These concerns that parents expressed as the fathers' leave was coming to an end were not fundamentally different from the structural constraints which drove the fathers to take leave in the first place.

Balancing care duties with employment after the end of the leave
Given that some parents approached the end of the fathers' leave with renewed concerns, I finally examine how fathers respond to their concerns about work-family balance after the end of their leave. Three divergent themes emerged, fathers (1) adjusting their work schedules, (2) reverting to overworking, or (3) experiencing changes in their employment status.

Making adjustments to balance care duties with employment
The first group of fathers made a conscious effort to continue their family engagement after returning to full-time employment. The earlier mentioned case of blogger 7 belongs to this category of fathers, whereby he actively addressed his concerns by utilising various policies available at his workplace including flexible working hours to be able to come home to his children every evening as opposed to just the weekends. Bloggers 8 and 9, both of whose children were under three, also adjusted their work schedules to coordinate childcare. As could be seen in these quotes, discourses stressing parental care for children under three continued to shape fathers' decisions to adjust work schedules to a degree, though this was not an issue for fathers whose children were older by the end of their leave.
"I volunteered for the morning shift. I wake up at 4 am and go to work at 5:30. I officially get off at 2 pm. That allows me to return home to spend time with my wife and child. I want to keep this schedule up until my child is 36 months old. [… ] This way, [my wife and I] are coordinating our schedules to raising our child together." -Blogger 8 "Originally, both of us worked from 8 to 5, but then [child] would have to be at preschool for eleven hours. So [wife] decided to work from 9 to 6 by pushing back an hour. Then she can send [child] to preschool and go to work. And I decided to work from 7 to 4, pushing forward an hour. Then I could pick [child] up at five and the time [child] spends at childcare shortens to nine hours, from 8 to 5." -Blogger 9 Others, such as interviewees A (mentioned earlier) and E, cut down on their after-work arrangements to come home as early as possible. A and E were similar in that they both described themselves as having been very work-oriented prior to taking leave, but their leave experience helped them to realise the importance of being more engaged at home. Because these fathers had reoriented their priorities, they showed sustained efforts to strike a balance between employment and family life. One factor making this possible could be that most fathers in this category were public sector employees, which often implies better job security and work-life balance than those hired in the private sector.
"I still work in the same team so there's a lot [of work], but still, I try to come home early in the evening and try my best not to make arrangements or have a drink with people." -Interviewee A "I had thought it is natural for a working man to [overwork]. […] After a year [of parental leave], I try not to make arrangements after work hours." -Interviewee E However, a few of these fathers, such as the blogger 8 (mentioned earlier) as well as interviewee F also mentioned being unfairly side-lined or discriminated in the valuations after returning to work, despite both being public sector employees. Thus, the effort that these fathers made to orient their lives toward their family by taking leave and making adjustments after their leave should not necessarily be understood as made possible solely due to conducive work environments; fathers were also actively resisting or opting out of workplaces cultures which are governed around norms of the ideal worker.
"I thought this would never happen to me […] but my position was gone when I returned to work.
[…] I started to be discreetly side-lined from tasks." -Blogger 8 "I mentioned earlier that I was discriminated against in the valuation. After going through that I'm like, why should I do this for the company, why I should be dedicated?" -Interviewee F 5.3.2. Reverting to original work-centred lives A second group of fathers reverted to their overworked lives upon return to work. It was not uncommon for fathers to be coming home in the late evening, which meant the entire family could spend only a few hours together each day during the week. Several fathers explained being too fatigued after work to be able to spend proper time with their child, consistent with the concerns outlined earlier. A good example is interviewee G, who was on his second leave and explained that even though he changed a lot during his first leave, such memories and rhythms had faded once he returned to work.
"I forgot everything when I returned to work. I would be at work by 8 [am] and get off at around 7 during busy periods and on days with staff dinner it would end at 9, 10 or 11 [pm]. So gradually the memories and senses would disappear and I went back to my old self of doing major things in the weekends because I was too tired in the weekdays." -Interviewee G Such dual-earning parents typically resorted to filling any gaps in childcare by receiving help from grandparents if possible, as in the case of interviewee H. Without the availability of help from extended families, a child could be left at childcare for up to eleven hours, as demonstrated by blogger 10. Furthermore, in cases where the fathers reverted to work-centric lives, the duty to pick up and care for the children often fell on the wives who typically got off work earlier and/or lived closer to her workplace, as (mom) blogger 11 illustrates.
"The first three months were difficult, but now everyone's lives have adjusted.
[…] We moved closer to my in-laws, and my mother-in-law is helping with everything.
[…] I typically help out at the weekends and on the weekdays I come home pretty late, even nowadays." -Interviewee H "Since my wife and I both started working again, [child] spends almost 11 h at preschool, from 8:30 am to 7 pm. Besides time spent sleeping or at preschool [child] only has 3-4 h to spend with mom and dad each day. We had to send [child] to preschool because of multiple circumstances but I am unsure if my wife and I are doing the right thing." -Blogger 10 "As always, today I hurriedly got off work at 6 pm and rushed to my parent's place where they are taking care of the kids. My husband would probably come home late again.
[…] It's been over a month that he has been doing this.
[…] It's difficult to see my husband day and night, and I focus on work intensely during the day so that I can get home as soon as possible. My parents are sacrificing so my husband could be freed from housework." -(Mom) blogger 11 These stories demonstrate why and how the "undoing gender" effect which is frequently documented during a father's leave is partially undone and fathers "redo gender" when they return to employment. Even though most fathers looked back on their leave experience and time with family with much fondness, many reverted to their pre-leave work-centric lives, in part because they were absorbed by the institutional inertia but also to an extent, because they were eager to make up for their time and ambition for career building which had been on hold during their leave.

Change in employment status to meet childcare demands
While most fathers maintained their jobs, a few reported changes in employment following their leave, becoming a freelancer or househusband to better meet childcare needs. 10 These fathers commonly stressed that providing reasonable care for their children was impossible without one person quitting their job. The reasons why these fathers rather than their wives opted for a career change varied, including because their wives had better-paying jobs (in the case of interviewee I), were public servants and eligible for pensions (in the case of interviewee J), or had medical conditions (in the case of interviewee K). As interviewee K aptly put it, these career changes were results of 'a mixture of voluntary choice and push from circumstances' but sub-optimal in that fathers were faced with an either-or situation between work and family life. These stories again demonstrate the overall structural incompatibility between employment and childcare that dualearner couples face in raising a young child without support from extended families.
"I returned to work and [my wife and I] were both working for eight months. There was a period, we were both so busy that we reached a limit. I had originally planned to move to another company, but the kids happened to be sick very frequently then.
[…] So I thought I should [take time off] completely, and that just continued. I was a full-time househusband for 1.5 years." -Interviewee I "There was a point when my wife and I were both working for a few months. We had no one to look after the kids for us so we sent the kids to their aunt's place in [another city]. Every night we would video chat with the kids. That's when I thought, do I have to go this far to make money?" -Interviewee J "The range of employment options you could choose from as a caregiver becomes incredibly narrow.
[…] So it was a mixture of voluntary choice and push from circumstances.
[…] After much agony, my wife and I ultimately decided to put care as the constant." -Interviewee K

Discussion and conclusion
This paper has sought to examine how cultural and institutional norms around childcare and employment influence fathers' decision to take parental leave and work-family balance after leave. First, many fathers articulated uptake of leave as an inevitable decision, structured around a lack or absence of other ways to organise childcare while both parents were in full-time employment. Distrust toward formal care and reluctance to keep children under three at care for extended periods were key reasons why parents of young children preferred family care. This also explains why Korean fathers often take long leaves of or nearing one year, to look after their child until they are "old enough" to send to formal care. Even for those with older children, the time gap between employment and childcare or schooling was an issue. More broadly, fathers expressed being overworked and not able to spend enough quality time with their children as a motivation for taking leave; they wanted to use the time to rest and build closer relationships with their children. In short, fathers' uptake of (long) leave was often a response to high levels of strains and incompatibility between employment and family life for dual-earning couples.
Naturally, dual-earning parents had to revisit much of these concerns when the fathers' leave came to an end. While some fathers arranged their work hours or cut down on afterwork arrangementssometimes at the expense of career disadvantagesothers reverted to their previous work-centric lives, leaving the primary duty for childcare to the grandparents and/or the mother. In a few cases, fathers quit their jobs upon judgement that their full-time employment was simply incompatible with childcare. In all these cases, the strains which shaped the fathers' decision to take leave continued to shape their work and family lives after the end of the leave.
What implications does the interaction between fathers' uptake of leave and norms about childcare and long working hours have for gender-equalising work-family reconciliation? First, fathers' uptake of leave does indeed redistribute duties to care for a young child between the couple and support mothers' return to employment and continuation of career through transferring some of the mothers' strains in balancing employment and childcare and the career disadvantages that may follow, onto the fathers. Although fathers may not explicitly draw on discourses of gender equality when articulating their reasons for taking leave, broader themes of gender equality were often embedded and implied in their decisions to take leave. In this sense, policy incentives for fathers to take leave do result in more gender-equal divisions of leave and work-family balance more generally. However, this was done more by straining and disadvantaging the fathers in their wives' replacement rather than enabling both parents to balance employment with childcare with less strain on the whole. Therefore, under Korea's cultural and institutional conditions, fathers' uptake of leave may be better understood as responses to the difficulties of balancing employment and childcare rather than means to bring about fundamental changes for workfamily balance. Furthermore, the "undoing gender" effect that is often observed during the fathers' leave often disappears or weakens once the fathers return to their busy cycles of employment and work-family imbalance. Thus, the outcome of fathers' leave is not only dependent on the fathers' agency and whether he takes the leave solo and for how long but moreover very much moderated by the institutional and cultural norms which are embedded into the context of employment and childcare.
Therefore, while fathers' leave is indeed an important entry point, it would be naïve to consider it as the panacea to gender equality or work-family balance (see also Doucet & McKay, 2020;Moss & Deven, 2015). In Korea's case, the strong preference for home-based care and long working hours are contributing to the increase in fathers' uptake of (long) leave, but simultaneously also limiting the impact of fathers' uptake of leave on gender equality. My findings suggest that the change that could be brought about by the development of a single social family policy would be limited if other interrelated institutions and cultural norms are not complementary. Put another way, while fathers' uptake of leave may have radical potential, it is the configuration of institutions and cultures in which leave policy is located which determine whether and to what extent such radical potential may be actualised.
The main policy implication of this research would thus be to reduce the contradictions or tensions and to enhance the complementarity between different cultures and/or institutions to enable parents to better balance employment and childcare. Regarding working hours, the recent legislative amendment in Korea introducing a 52-hour limit on working time is a change in the right direction, but the 52-hours limit is still longer than the typical upper limit of 48 h in the OECD. Moreover, it may take some time for the reduced working hours to become a widely accepted cultural norm. Furthermore, more effort should be taken to encourage parents, especially fathers, to work flexible or reduced hours, as the number of fathers who utilise this policy is even smaller than those who take fathers' leave. In ensuring wider access to quality ECEC for all, the emphasis should be placed on addressing the root cause of parents' lack of trust toward formal childcare. Guaranteeing better working conditions for the preschool teachers and introducing better regulatory measures for monitoring the running of ECEC would be a much-needed step and may offer a challenging but more fundamental solution than the "old-wine-in-new-bottles" approaches such as accrediting private ECEC facilities, which has been proven to be ineffective.
In conclusion, this research goes to show the importance of consistency or coordination within a bundle of policies as well as the need to study specific policies, not in isolation but in relation to relevant social norms and institutions. While the promotion of father's uptake of leave is consistent with Nancy Fraser's call for a Universal Caregiver paradigm (1997), its potential for gender equality can be actualised only when the institutions which leave policy interacts with are simultaneously redesigned 'so as to eliminate the difficulty and strain' (p. 611) of balancing childcare and employment.
The originality of this research lies in three aspects. First, it responds to recent calls to empirically study fathers' leave in relation to broader cultural and institutional contexts. By documenting the experiences and voices of fathers at the three-way intersection of fathers' leave, working hours, and ECEC, I have demonstrated how much norms around employment and childcare simultaneously shape both fathers' decision to take leave as well as aftermaths. Second, existing studies often imply that fathers taking leave is a sign of enabling circumstances or fathers' agency, while masculine workplace norms act as a barrier for fathers' uptake of leave. On the other hand, this research found that for Korean fathers, uptake of leave can, in certain circumstances, be a response to structural constraints coming from difficulties in balancing employment and childcare. It is important to recognise here that fathers are actively exercising agency within the structural constraints posed by institutional and cultural norms. Finally, this research documented the rich and vivid experiences and narratives of one hundred cases of fathers' leave by creatively triangulating two sources of qualitative data, interviews and blog posts, the latter of which is an underutilised data source with much room to further explore.
However, this research is not without limitations. For one thing, it must be recognised that some, though not all, fathers were on leave during the COVID-19 pandemic, during when the data collection for this study took place. While the pandemic was, on whole, not a prominent theme even when the fathers were explicitly asked about it during the interviews, a part of the data may potentially be biased because of the pandemic setting, during which parents generally had to burden increased labour for childcare due to the closure of some childcare facilities. Second, although I have sought to complement limitations of the data by triangulating my data, the shortcomings of each types of data should nonetheless be recognised: that the interviews did not collect longitudinal information from both the father and his wife, and that the blog data had missing and/ or unclear information that I had to work with. I thus invite more studies based on longitudinal couple interviews to track, in particular, fathers' transition back to the workplace following the end of their leave.
Finally, even though I have tried to base the research on a sufficiently large number of cases (51 interviews and 49 blogs), the qualitative data is not representative of the population in any means. Hence, it would be more accurate to understand the findings of this research as reflective of the experience of some, but not all, Korean parents or fathers with leave experience. For instance, while I have focused on the salience of norms about employment and childcare in this paper, only a third of Korean fathers take leave when their child is three or younger. The childcare-related constraints that I have discussed are particularly relevant to those who take leave for a young child under three, but less so for those taking leave for an older child. Nonetheless, this paper has demonstrated the importance and value in studying fathers' leave experience in tandem with employment and childcare norms. Hence, while being cautious about generalising the findings from this research as representing the experiences of all Korean fathers who take leave, I invite further research which locate fathers' leave within broader structural and cultural contexts. Notes 1. Because the parental leave statistics are also administered separately by sector and most publicly available statistics to date are based on private sector employees only, presenting a holistic picture which is inclusive of both sectors is difficult (see Table A1 of the appendix). 2. In addition, female workers are eligible for 90 days of paid maternity leave and male workers ten days of paid paternity leave (formerly five days, of which three were paid). 3. This includes non-standard contracted workers, self-employed workers, those who work under 15 h per week and/or in the informal economy, as well as casualised workers and/or employees of small businesses who do not join employment insurance (Statistics Korea, 2018). 4. 83 percent of private sector employees in Korea work in small or medium sized companies with less than 300 employees. This means that most workers were not subject to working hour reductions in 2020, when the data collection for this research took place. The same goes for the majority of the interviewed fathers; only five interviewees were employed in large companies at the time of interview. 5. Korea has a split ECEC system, with kindergartens categorised as educational institutions governed under the Ministry of Education and typically providing half-day education for children aged 3 to 5. In contrast, childcare centres provide institutional or home-based all-day care and education for children aged 0-5 and are under the jurisdiction of the Ministry of Health and Welfare. However, the distinction between the two institutions and the service they provide are becoming more blurred these days, with the introduction of the Nuri curriculum in 2011 intended to harmonise the two services (Hwang, 2021). 6. In the case of kindergartens, the number expanded rapidly during the 1980's but since the 1990s has been relatively stable at around 8,000 to 9,000 (Statistics Korea, 2020e). 7. In the case of private kindergartens, 45% of their annual budget is funded by public spending (Hwang, 2021). 8. Doucet suggests conducting a total of four readings, each time paying attention to readerresponse reflexivity, central protagonist narrative, general narrative, and social relationships and subjectivities, and socio-structural contexts at the macro-level (see also Brown & Gilligan, 1992;Doucet, 2018;Gilligan et al., 2003). 9. van Manen proposes first developing a holistic picture of each interviewee, then purposefully focusing on certain topics of interest, and then paying attention to the details of people's narratives (see also Rosenblatt & Wieling, 2019). 10. Among the fathers who experienced changes in their occupations or employment status after the end of the leave, in some cases this had been planned out since before the father applied for leave (for instance, one father had planned to start his own business after finishing his leave and did so) and it was not always clear whether the change in employment status was made more for reasons relating to the fathers' career ambitions, reasons relating to the fathers' changed state of mind, or reasons more to do with acquiring better work-family balance-sometimes it was a mixture of these. Hence, while I present the total number of fathers who experienced changes in their career or employment status in Table  1, I must note that not all of these fathers changed their employment status for reasons that the quotes of interviewees I, J, and K suggest.

Disclosure statement
No potential conflict of interest was reported by the author(s).  (2019), Statistics Korea (2020a) Note: In producing this table, I have referred, wherever possible, to statistics inclusive of both the private and public sectors but rows (f), (g), and (h) are based on the private sector due to data limits. The number of parents who took parental leave in year t for a child born in year t / the number of children born in year t * 100 The number of parents eligible for parental leave in year t / the number of children born in year t * 100 The number of parents who took parental leave in year t for a child born in year t / the number of parents eligible for parental leave for a child born in in year t * 100 The number of parents who took parental leave between year t and year t+9 for a child born in year t / the number of children born in year t * 100