Re-thinking Fatherhood in Context

The transformation of gender relations towards greater labor market equality remains uneven and incomplete (England 2010; Gornick/Meyers 2009). Women’s increased participation in employment has narrowed economic inequalities between men and women in many respects. However, women’s employment rates and working hours still lag behind those of men, occupational gender desegregation is stagnating, and gender earnings gaps refuse to disappear in most affluent Western countries (Bernhardt/Morris/Handcock 1995; Bettio/Verashchagina 2008; Petersen/ Penner/Høgnes 2014). What keeps these gender inequalities from fading? One important source of the persistent gender gaps is parenthood (Gornick/Meyers 2009). Inequalities between childless women and men have narrowed or even disappeared, but parenthood continues to exacerbate gender differences in paid (and unpaid) work (Hook/Pettit 2016; Yavorsky/Kamp Dush/Schoppe-Sullivan 2015). While men have increased the amount of time they spend on unpaid work in families, this increase has not matched women’s gains in paid work. Women, especially mothers, continue to provide the greater part of care work in families. As a consequence, mothers often have a more tenuous attachment to employment and fare worse in the labor market compared to other groups of workers, including childless women. This contrasts with fathers’ experiences. Research has shown that many men benefit from fatherhood. For example, unlike women who incur wage penalties for motherhood, many men garner earnings advantages relative to childless men (Smith Koslowski 2011; Hodges/Budig 2010). Both “motherhood penalties” and “fatherhood wage premiums” contribute to the persistence of gendered earnings inequalities (Petersen/Penner/ Høgnes 2014). Nevertheless, research and institutional interventions addressing the nexus between paid and unpaid work have primarily focused on women. Work-family researchers have generated a rich body of literature that seeks to understand why mothers fare worse in the labor market, and how institutional and cultural contexts moderate the impact of children on women’s employment (Gornick/Meyers 2009; Korpi/Ferrarini/Englund 2013; Pfau-Effinger 2004). In contrast, the role of fatherhood in shaping men’s employment has received relatively little attention. Men’s employment patterns are often (implicitly) considered as a seemingly homogenous, yet unexamined background against which women’s employment is evaluated. Furthermore, we are only beginning to understand how socio-political and economic contexts shape the relationship between fatherhood and men’s employment outcomes (Bünning/Pollmann-Schult 2016). This essay argues that in order to further unpack gender economic inequalities in post-industrial societies we need to bring fatherhood and its relationship to men’s employment into sharper focus. This requires us not only to ask how parenthood

shapes men's working lives, but how different groups of fathers fare in the labor market, and how their engagement in paid (and unpaid) work is linked to labor market structures and larger socio-political contexts.In this essay, I use the example of fathers' paid working hours to illustrate the heterogeneity of their employment patterns within and across welfare state and labor market contexts, with a particular focus on the United States (see also Woods 2012).I use longitudinal data from the Panel Study of Income Dynamics and group-based trajectory models to examine men's employment patters before and after the first transition to parenthood.Rather than estimating average employment patterns, this approach allows for an examination of the diversity in men's employment patterns over time.I examine how race and class 1 are linked to different employment pathways among fathers, and how the prevalence of pathways has changed over time.I start by considering how the particular institutional and labor market context of the U.S. may shape men's employment-family interface by comparing the U.S. with three examples of European countries drawing on data from the 2014 European Labor Force Survey and the U.S. Current Population Survey.
Men's Employment in Europe and the U.S.
Men's work-family practices and cultural ideals of fatherhood are changing."New fathers" who are active and involved fathers are increasingly visible in the public space and the media (Kaufman 2013).Time-use research indeed shows that married fathers in the U.S. have more than doubled the average time they spend on housework and childcare from 7.3 hours in 1965 to 16.7 hours a week in 2010 (Bianchi et al. 2012).Nevertheless, children continue to shape men's lives in different ways compared to women, notably with regard to employment.Stable full-time employment and the breadwinner role remain salient markers of "good fatherhood" (Brescoll/Uhlmann 2005;Townsend 2002).This is the case even though full-time work no longer guarantees breadwinner wages for many workers in the U.S. and in countries across Europe (Gautié/Ponthieux 2016;Sweet/Meiksins 2008).Fathers are more likely to be employed and work longer hours compared to childless men and women, and mothers (Misra/Budig/Boeckmann 2011).A comparison of men's weekly working hours in the U.S., Sweden, the Netherlands, and Germany supports this. Figure 1 shows 25 to 45 year old men's weekly employment hours by fatherhood status. 2 Despite distinct labor market structures and public policies in these countries, men who live with children in the same household are more likely to be employed and work full-time (35 hours or more per week) than childless men in all four countries.What could contribute to this pattern?On the one hand, men who become fathers may also possess characteristics that help them to secure (full-time) employment. 3On the other hand, the continuing full-time employment of a majority of fathers is also a manifestation of the persistent significance of breadwinning for fathers, cutting across different socio-political contexts.Nevertheless, there are clear differences in working patterns across countries, in particular in the incidence of very long working hours above 50 and 60 hours per week.Among these four countries, Sweden and the Netherlands are most successful at curbing very long working hours through statutory and collective regulation of working time. 4The Dutch government has also been among the most active in promoting part-time work for both women and men through working time policies and public relations campaigns (Lewis et al. 2008).Even though mothers remain the majority of part-time workers, Dutch men (including fathers) are more likely than their counterparts in other countries to work part-time.German statutory and collective working time regulation also help to curb weekly working hours (Eurofound 2016), but German men are more likely than Dutch or Swedish men to work longer full-time hours.German family policy reforms in the past decade have implemented more incentives for fathers to take up childcare.But change remains slow, and fulltime employment continues to be the norm for the vast majority of German fathers.While regulating normal weekly working hours is usually not considered a core work-family policy, shorter overall work weeks free up more time for other activi-ties, including the care of children or elderly parents.This may ease work-family conflict if paired with regular and predictable work schedules (Fagnani/Letablier 2004;Gornick/Heron 2006).As Figure 1 illustrates, the U.S. is unique among these four countries in the extent that men, and especially fathers, work very long hours.What are contributing factors?
Labor Market Structure and Public Policies in the U.S.
U.S. laws provide only weak regulation of working time compared to other industrialized Western countries, and policies offer little support for parents with young children. 5The 1938 Fair Labor Standards Act set the weekly working hours above which overtime is owed at 40.But certain groups of workers are exempt from this rule and the proportion of the covered workforce has eroded over time. 6As a consequence American parents, especially fathers, spend more hours on paid work on average compared to their European counterparts (Gornick/Heron 2006;Jacobs/ Gornick 2002).Moreover, the U.S. provides limited public support for working parents.Childcare is provided through the market and the cost to parents is high compared to many European countries.Affordable, quality childcare services are often inaccessible for working-class parents (Morgan 2005).The leave entitlement provided by the 1996 Family and Medical Leave Act (FMLA) is unpaid and only about half of U.S. workers are covered, which limits parents' ability to take up leave (Jorgensen/Appelbaum 2014). 7In the absence of stronger public support, it is primarily up to families to solve the quandary between securing care for their children and their economic well-being.Men continue to be the main earners in the majority of two-parent families, even though the share of primary breadwinner mothers has continually risen in the U.S. (Wang/Parker/Taylor 2013).The limited public support for families is compounded by labor market conditions that have over time broadened opportunities for high-skilled workers, but made it hard for workers with fewer educational credentials and skills to find stable, fulltime employment that pays living wages (Kalleberg 2011).In terms of working time, this has also led to a polarization between workers who spend very long hours in employment (especially in professional and managerial occupations), and workers who cannot find enough work to make a living (Jacobs/Gerson 2005).Different groups of American fathers may be therefore confronted with different conditions under which to reconcile employment and family.

Methods and Data
To examine how different groups of fathers fare in the U.S., I draw on longitudinal data from the Panel Study of Income Dynamics, a survey that has followed a sample of U.S. families and their descendants since 1968.I follow a sample of men from the year before they became fathers (for at least three years) up to their first child's ninth birthday. 8Studies have examined average changes in working hours associated with the transition to fatherhood.But we know little about how fathers' employment participation and working hours evolve over time.In particular, whether there is heterogeneity in the employment trajectories fathers follow, and how common they are.The sample consists of 1877 black and white men born between 1943 and 1988, aged 18 or older at the time of the first birth. 9To examine fathers' employment pathways over time, I use group-based trajectory models (Nagin 2005).This is a regression-based method that allows researchers to probe data for the presence of multiple ideal-typical pathways over a given period, and estimate individuals' likelihoods of following these pathways. 10The analysis also considers how race and differences in (pre-parenthood) educational attainment are linked to new fathers' employment pathways, and whether the prevalence of different employment pathways has changed over time.

Inequalities among Fathers in the U.S. Labor Market
The central finding of this analysis is that a majority of American fathers (nearly 75%) follow stable employment trajectories, either working around 42 hours (55%) or around 51 hours a week (20%).For this majority of fathers, employment participation and working hours seem unaffected by parenthood.Only a small minority of men (approx.10%) shows increasing levels of engagement in employment after fatherhood, either increasing working hours from an already high level (around 60 hours a week), or take-up of employment after leaving education.The remaining 15% of the sample represents men with fluctuating employment participation patterns or marginally employed men (either out of employment or working only a few hours).Most of these men work in jobs with hours that are full-time or nearly full-time.But they experience spells out of employment, or significant reductions in their working hours over the course of the observation window.Comparing across successive cohorts of men, the prevalence of pathways involving very long working hours above 50 or 60 hours a week slightly increased over time, as did the incidence of the marginal employment trajectory.In contrast, working the standard 40 hour workweek has become somewhat less common among fathers. 11This is in line with broader changes in the U.S. labor market (Jacobs/Gerson 2005).These shifts in the prevalence of ideal-typical pathways over time are however neither substantial, nor statistically significant.Thus, even with fathers' gains in care work over time, these findings suggest stability in labor market patterns of the vast majority of fathers.However, there are important differences in employment trajectories between black and white fathers, and between men with different levels of education.Men with fewer educational credentials, particularly black fathers, are more likely to follow trajectories with periods out of employment or marginal labor market attachment during early fatherhood.By having limited access to stable, full-time employment, these fathers also have restricted access to socially valued forms of fatherhood, which continue to emphasize the breadwinner role.Although college education protects black fathers from employment instability, white college-educated fathers' racially privileged position in the labor market facilitates their access to professional and managerial jobs, which often means working very long hours in the American labor market.In contrast, college-educated, black fathers are underrepresented in employment trajectories with very long hours and are more likely to work "regular" full-time hours around 40 hours per week.How may these inequalities shape fathers' employment-family interface?

Fatherhood in the Context of Intersecting Inequalities
On the one hand, fewer hours at work open opportunities for greater involvement in the day-to-day care of children, and time for unpaid family work, leisure, and ultimately for a more equitable distribution of paid and unpaid work in two-parent households.On the other hand, black fathers' underrepresentation in long hours' work also highlights the racial disparities in the access to jobs that are most likely to pay true breadwinner wages in the American labor market, i.e. professional-managerial jobs.Then again, very long hours, or "overwork" have been shown to negatively impact fathers' relationships with adolescent children (Crouter et al. 2001), and couples' division of labor.For example, Youngjoo Cha (2010) showed that men's very long working hours increase the risk of female partners leaving employment, especially among parents.Racial and class inequalities among fathers thus intersect with gender inequalities in complex ways.Taken together, the findings suggest that for the majority of American men, engagement in paid work remains largely unaffected by parenthood despite increasing levels of engagement in care work among men in general (Bianchi et al. 2012).In contrast to women, most men still seem to be able to consider family and employment as separate spheres.At the same time, disparities in access to stable (well-paying) full-time jobs limit access to socially valued and rewarded forms of fatherhood for some men.Especially fathers with fewer educational credentials and skills have higher risks of being unemployed or underemployed, and earning low wages.Public discourses in the U.S. frequently portray low-income and marginally employed fathers as "deadbeat dads" who take little financial responsibility for the children they father.Conversely, men with more education and marketable skills in professional jobs are often found at the other extreme, being intensely engaged in employment.Working very long hours likely limits their involvement in the day-to-day care of their children.Nevertheless, in contrast to low-income fathers, they are unlikely to be sanctioned for their breadwinning-centered fatherhood practices.Highlighting shifts in fathers' childcare and housework is important.But the findings of these analyses suggest caution in a too optimistic interpretation of changes in men's lives in connection with parenthood in the context of the U.S.Many men are more actively involved in the care of young children, and spend more time on other unpaid work within households than their fathers.Nevertheless, men's engagement in paid work remains shielded from the impact of care responsibilities, and this contributes to the maintenance of gendered labor market inequalities.

Notes
11 Education is used as a proxy measure for class.12 Fathers are defined as men who live with their children in the same household.The samples are limited to 25 to 45 year old men, the prime years for employment and childbearing.Men who do live with children cannot be identified as fathers in the data, some "childless men" may therefore be non-resident fathers.13 For example, men who are more sociable are more likely to become married and become fathers (Jokela/Kivimäki/Elovainio 2009).If sociability and similar characteristics also predict greater engagement in the labor market (e.g. because employers value these characteristics), average differences between fathers and childless men in employment may be due to the selection of men with greater propensity for more intense engagement in the labor market into fatherhood.14 With average hours slightly above 37 they were among the European countries with the lowest collectively agreed hours in 2014 (Eurofound 2016).In both countries over 80% (Netherlands), respectively 90% (Sweden) of the workforce is covered by collective agreements (Gornick/ Heron 2006).15 In Europe, men in the United Kingdom, a country with weak working time regulation, also have a similarly high incidence of very long working hours.16 Managers and supervisors are exempted, as are workers earning above a threshold, which was last adjusted in 1975 (Gornick/Heron 2006).As a result, the salaried workforce covered by the overtime rule has fallen from 49.6% in 1975 to 9.5% in 2015 (Cooper et al. 2016).17 To be eligible for FLMA, an employee has to work for an employer with at least 50 employees in a 75 mile radius for at least one year and 1,250 hours.Under these criteria an estimated 55.9% of workers were eligible to take unpaid leave of 12 weeks in 2012.An estimated 20% of eligible employees took leave over a period of 18 months covered by the survey (Jorgensen/ Appelbaum 2014).18 Because the analysis focuses on employment pathways over time, men who are observed for less than 3 years in the data are not included.19 Both co-resident and non-resident fathers are included in the PSID sample.Comparisons of groups other than white and black men were not possible due to insufficient sample sizes.10 More detailed information about the data, sample and the method is available from the author.11 Robustness analyses show that the findings are substantively the same regardless of how cohorts were defined, including the common distinction between the early Baby Boom (1943-1955), later Baby Boom (1956-1964), and post-Baby Boom cohorts .

Figure 1 .
Figure 1.Distribution of Fathers Across Different Hours' Brackets (Weekly Work Hours)