Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2020 Apr 1.
Published in final edited form as: J Marriage Fam. 2018 Nov 12;81(2):468–486. doi: 10.1111/jomf.12545

Paternity Leave and Parental Relationships: Variations by Gender and Mothers’ Work Statuses

Richard J Petts a,✉, Chris Knoester b
PMCID: PMC6407703  NIHMSID: NIHMS991900  PMID: 30858623

Abstract

Objective:

This study examines the associations between paternity leave and parents’ reports of relationship satisfaction and relationship conflict, and whether the associations vary by parent gender and mothers’ work statuses.

Background:

Paternity leave research in the U.S. has focused on implications for father involvement, but paternity leave may also help to strengthen parental relationships by promoting a more equitable division of domestic labor. Given gender gaps in childcare, the association between paternity leave and parental relationship outcomes may also vary by gender and mothers’ work statuses.

Method:

The sample consists of 4,700 couples (i.e., parent dyads) from the Early Childhood Longitudinal Study, Birth Cohort. Multilevel models are used to assess the associations between paternity leave and both relationship satisfaction and conflict, and whether these associations vary by gender and mothers’ work statuses.

Results:

Paternity leave-taking is positively associated with parents’ reports of relationship satisfaction, but length of paternity leave is only positively associated with mothers’ reports of relationship satisfaction. Also, among mothers who worked prior to the child’s birth, paternity leave-taking and length of leave are negatively associated with their reports of relationship conflict. In contrast, among mothers who did not work in paid labor pre-birth, paternity leave is positively associated with mothers’ reports of relationship conflict.

Conclusion:

Paternity leave may have implications for parental relationships (and especially mothers’ perceptions of their relationships with fathers).

Keywords: conflict, family policy, family roles, fatherhood, satisfaction, work-family balance

Work-family balance has become more challenging for parents in recent decades as greater demands have been placed on both employees and parents, resulting in increased work-family conflict (Aumann, Galinsky, & Matos, 2011; Gerson, 2010; Jacobs & Gerson, 2004). This conflict has numerous consequences for families, including reduced paternal involvement, lower parental well-being, and lower parental relationship quality (Daly, 1996; Fellows et al., 2016; Gerson, 2010). As such, it is important to examine policies that may help to reduce work-family conflict and consequently improve family outcomes.

One policy that has received limited attention in the literature is paternity leave. Paternity leave may enable fathers to balance work and family life by allowing fathers to maintain traditional breadwinning roles while also benefitting from time off work and the ability to pursue new fatherhood ideals that emphasize being a nurturing, engaged father (Galinsky, Aumann, & Bond, 2011; Marsiglio & Roy, 2012). In particular, paternity leave may help fathers become more involved in parenting tasks, have more opportunities to learn parenting skills, and develop strong coparenting relationships with the mother following a birth (Almqvist & Duvander, 2014; Bünning, 2015; Huerta et al., 2014; Rehel, 2014). Not surprisingly, paternity leave-taking and longer periods of leave are associated with more frequent father involvement (Haas & Hwang, 2008; Huerta et al., 2014; Nepomnyaschy & Waldfogel, 2007; Pragg & Knoester, 2017).

Yet, research has not fully considered how paternity leave may be associated with other family outcomes such as parental relationship satisfaction and relationship conflict within the context of the United States. By potentially increasing the likelihood that fathers more equally share parenting tasks, and allowing fathers and mothers time to learn how to coparent together from birth, paternity leave may help to strengthen parental relationships (Galovan et al., 2014; Kotsadam & Finseraas, 2011; Newkirk, Perry-Jenkins, & Sayer, 2017; Rehel, 2014). Although there is evidence that paternity leave-taking is associated with lower relationship conflict in Europe, and that more extensive leave policies contribute to parental happiness, research has yet to examine potential links between paternity leave and parental relationships using U.S. data (Glass, Simon, & Anderson 2016; Kotsadam & Finseraas, 2011). Understanding the U.S. context is important because the U.S. is one of only a handful of countries that does not have a national paid parental leave policy (World Bank Group, 2018).

To address this gap, the current study focuses on whether paternity leave-taking and length of paternity leave are associated with parental relationship satisfaction and relationship conflict in the U.S. We also consider whether these associations vary by gender and mothers’ work statuses. In doing so, this study will assess whether findings on paternity leave and parental relationships persist within a society in which access to paid leave is limited (Petts, Knoester, & Li, 2018). Because perceived equity in the division of labor is more likely to influence mothers’ perceptions of relationship satisfaction than fathers’ perceptions, it is possible that paternity leave may be more closely associated with mothers’ perceptions of relationship satisfaction and conflict than fathers’ perceptions (Galovan et al., 2014; Milkie et al., 2002; Newkirk et al., 2017). Increased support from fathers may also be more beneficial to working mothers, compared to mothers who do not work. Indeed, evidence from European studies suggests that paternity leave may help to reduce mothers’ stress and facilitate a quicker return to the labor force for working mothers (Gault et al., 2014; Johansson, 2010; Sejourne et al., 2012). Thus, paternity leave may be more likely to promote relationship satisfaction and reduce conflict in dual-income families, compared to families in which only the father works in paid labor. Overall, understanding the potential associations between paternity leave and parental relationships may aid city, state, and federal policymakers in decisions regarding paid family leave policies.

CONCEPTUAL FRAMEWORK

The arrival of a new child brings about many family changes, as infants are heavily dependent on their parents for physical, social, and emotional care (Waldfogel, 2006). Notably, having a child may shift at least some attention away from pre-existing family members in order to take care of the new baby. Role theory is helpful in understanding how parents may experience the changes brought about by the arrival of a new child (Goode, 1960; Hecht, 2001).

Role theory suggests that having a new child may increase the likelihood of role conflicts (i.e., stress from multiple roles conflicting—such as parent/worker/partner) and role overloads (i.e., stress from too many role responsibilities, which may include role conflicts) (Cowan et al., 1985; Yavorsky, Kamp Dush, & Schoppe-Sullivan, 2015). Role conflicts and role overloads may increase stress, as meeting the expectations of multiple roles can be challenging (Goode, 1960). Because the arrival of a new child creates more work for parents, they prioritize parenting roles due to the substantial effort and attention that young children need (Twenge, Campbell, & Foster, 2003; Waldfogel, 2006). Thus, relationships between romantic partners may suffer. Perhaps not surprisingly, experiencing role conflict and role overload during the transition to parenthood is associated with lower relationship quality and increased conflict among parents (Cowan et al., 1985; Crohan, 1996; Hecht, 2001; Keizer & Schenk, 2012; Twenge et al., 2003).

One strategy that is often employed to manage role conflicts and overloads is role specialization; each parent prioritizes and specializes in particular roles to more efficiently complete family tasks (Becker, 1981). Most commonly, family specialization occurs in the form of role traditionalization; mothers specialize in childcare and housework and fathers specialize in paid labor (Cowan et al., 1985; Yavorsky et al., 2015). Although role traditionalization may theoretically help to alleviate role conflict (Becker 1981), it also contradicts the egalitarian ideology that both men and women support about more equally sharing in domestic and paid labor (Gerson, 2010; Pedulla & Thébaud, 2015).

Because men and women increasingly view the ideal family arrangement as one in which both parents equally share in domestic and paid labor (Gerson, 2010; Pedulla & Thébaud, 2015), role traditionalization may actually weaken, as opposed to strengthen, parental relationships. For mothers, they may feel unsupported by their partners if they are not sharing in domestic labor (i.e., housework and childcare) in an equitable way (Cowan et al., 1985; Dew & Wilcox, 2011; Milkie et al., 2002; Twenge et al., 2003; Wilcox & Nock, 2006). For fathers, they may feel resentment about the lack of time they have to spend with their child and their inability to pursue new fatherhood ideals to the extent that they might desire (Daly, 1996; Gerson, 2010; Marsiglio & Roy, 2012; Townsend, 2002). Moreover, even in families that do not engage in role specialization, mothers still engage in domestic work more frequently than fathers – who often spend more time in leisure (Kamp Dush, Yavorsky, & Schoppe-Sullivan, 2018). These dynamics may contribute to lower relationship satisfaction and greater relationship conflict.

Paternity Leave, Relationship Satisfaction, and Relationship Conflict

The tendency for role traditionalization to occur (or for women to share a greater burden of nonspecialized work) following the transition to parenthood is due, in part, because of cultural practices, public policies, and economic expectations that largely assume and encourage traditional gender roles (Kamp Dush et al., 2018; Pedulla & Thébaud, 2015; Risman, 1998; Yavorsky et al., 2015). Despite an increasing desire to change some of these gendered family roles, the structures and expectations that exist often constrain couples’ ability to enact egalitarian ideals. For example, new fatherhood ideals encourage men to prioritize their roles as engaged fathers, but national and workplace policies in the U.S. provide limited assistance to allow fathers flexibility to fully embrace these ideals (Albiston & O’Connor, 2016; Aumann et al., 2011; Marsiglio & Roy, 2012). Thus, it is important to examine policies that may encourage fathers to more consistently pursue egalitarianism and emphasize new fatherhood ideals.

Paternity leave is one policy that may provide fathers with the time needed to learn how to be a nurturing, engaged father while also maintaining their positions in the labor force (Rehel, 2014; Tanaka & Waldfogel, 2007). Yet, access to paternity leave is limited within the U.S., as national policies only provide unpaid leave to eligible workers, only four states offer paid parental leave, and only 16% of workers have access to paid family leave from their employers (Bureau of Labor Statistics, 2018; National Partnership for Women and Families, 2018).

Access to paternity leave may be particularly important as it provides time for fathers to learn parenting tasks, including those that have traditionally been performed by mothers (Rehel, 2014). Paternity leave, and longer periods of leave, may also provide time for mothers and fathers to learn to coparent, including setting clear expectations about how household labor will (and should) be divided (Rehel, 2014). Thus, paternity leave, and longer periods of leave, may reduce the likelihood of role traditionalization. Bonding with their child from birth may help to strengthen men’s identities as fathers, leading them to place a higher priority on engaged fathering roles relative to fathers who do not take leave (Pasley, Petren, & Fish, 2014; Pragg & Knoester, 2017). Consequently, fathers who take leave (and longer periods of leave) may be better equipped to remain an engaged parent than fathers who do not take leave.

Indeed, paternity leave-taking and length of paternity leave are associated with greater father involvement, more equitable sharing of childcare tasks, and increased participation in housework among fathers (Almqvist & Duvander, 2014,; Bünning, 2015; Hosking, Whitehouse, & Baxter, 2010; Huerta et al., 2014; Petts & Knoester, 2018; Pragg & Knoester, 2017; Tanaka & Waldfogel, 2007). Father involvement is also associated with stronger parental relationships (Keizer & Shenk, 2012; McClain & Brown, 2017). Overall, taking paternity leave, and longer periods of leave, may signal and contribute to fathers’ commitments to a more egalitarian coparenting relationship, which may be associated with higher relationship satisfaction and lower relationship conflict. Indeed, there is evidence from Europe suggesting that paternity leave-taking is associated with lower parental relationship conflict (Kotsadam & Finseraas, 2011), but research has yet to examine this association using U.S. data.

  • Hypothesis 1: Paternity leave-taking will be positively associated with parents’ reports of relationship satisfaction and negatively associated with relationship conflict.

  • Hypothesis 2: Length of paternity leave will be positively associated with parents’ reports of relationship satisfaction and negatively associated with relationship conflict.

Moderating Roles of Gender and Mothers’ Work Statuses

Although paternity leave may be associated with parental relationship satisfaction and conflict, it is also possible that these associations are moderated by gender (i.e., mothers’ vs. fathers’ reports) and mothers’ work statuses (i.e., mothers’ employment in paid labor). First, research suggests that role traditionalization is especially likely to reduce mothers’ perceptions of relationship quality (Cowan et al., 1985; Dew & Wilcox, 2011; Twenge et al., 2003; Wilcox & Nock, 2006). Because mothers are expected to be primarily responsible for childcare and housework while also often engaging in paid labor, they are more likely than fathers to perceive the division of household labor as unfair, especially for couples with egalitarian ideals (Dew & Wilcox, 2011; Milkie et al., 2002; Newkirk et al., 2017). To the extent that paternity leave-taking and length of paternity leave signal fathers’ commitments to being equal coparents (Almqvist & Duvander 2014; Kotsadam & Finseraas, 2011; Pragg & Knoester, 2017), it is possible that mothers are more likely to view the division of household labor as fair if fathers take paternity leave, and take longer periods of leave.

  • Hypothesis 3: Paternity leave-taking and length of paternity leave will be more closely associated with reports of higher levels of relationship satisfaction and lower levels of relationship conflict for mothers as compared to fathers.

The associations between paternity leave and parental relationship outcomes may also vary by whether mothers are employed prior to a child’s birth. The gendered structure of society largely places women at a disadvantage (Pedulla & Thébaud, 2015; Risman, 1998); women earn less money than men, and suffer an additional wage penalty for motherhood, whereas fatherhood is associated with a wage premium (England et al., 2016; Killewald, 2013). Women are also expected to engage in intensive mothering despite being highly involved in paid labor, resulting in women spending approximately twice as much time doing housework and childcare than men (Hays, 1996; Parker & Livingston, 2017; Yavorsky et al., 2015). As a result, women who have been consistently employed prior to a child’s birth, and intend to return to work, are especially likely to experience role conflicts and overloads (Bianchi & Milkie, 2010; Gerson, 2010).

Taking paternity leave, and longer periods of leave, may be especially helpful in alleviating role conflicts and overloads for working mothers. Taking time off work may signal a commitment by fathers to learn how to coparent, and negotiate a balanced work and family life, together (Rehel, 2014). Paternity leave may also enable fathers to help out after returning home from the hospital, which may alleviate some of mothers’ stress and allow mothers to recover physically and mentally from childbirth (Gault et al., 2014; Sejourne et al., 2012). As such, paternity leave-taking and longer lengths of leave may enable mothers to (re)enter the labor force sooner (Johansson, 2010). Given that shared income production is associated with greater relationship stability and satisfaction (Kalil, Ziol-Guest, & Epstein, 2010; Sayer & Bianchi, 2000), and that paternity leave may help to reduce role conflicts (Glass et al., 2016), paternity leave-taking and longer lengths of leave may be especially associated with higher relationship quality and lower conflict among couples in which mothers work prior to the birth of a child.

  • Hypothesis 4: Paternity leave-taking and longer lengths of paternity leave will be more closely associated with reports of higher levels of relationship satisfaction and lower levels of relationship conflict among couples in which the mother is employed in paid labor, as compared to couples in which the mother is not employed.

Other Factors

Although paternity leave may be associated with parental relationship satisfaction and conflict, other factors may confound these associations. In particular, socioeconomic status may be an important confounder. Most employers do not offer [paid] paternity leave, and higher SES positions are more likely to have access to this benefit (Albiston & O’Connor, 2016). SES is also associated with leave-taking practices; fathers with higher SES are more likely to take leave, and longer periods of leave, than low-SES fathers (Huerta et al., 2014; Petts et al., 2018). Moreover, the gendered structure of workplace and economic practices leads leave-taking to be stigmatized, as taking leave violates norms of what a good [male] worker should be (Albiston & O’Connor, 2016). As such, paternity leave-taking is associated with lower performance ratings and lower future income, especially for low-SES and racial/ethnic minority fathers (Rudman & Mescher, 2013; Williams, Blair-Loy, & Berdahl, 2013). Low-SES couples are also more likely to report parental conflict than high-SES couples due to economic stresses (Hardie & Lucas, 2010).

Fathering attitudes and behavior may also confound the association between paternity leave and parental relationship outcomes. Having a salient father identity that emphasizes father engagement, believing that fathers should be actively involved in children’s lives, and being involved prior to the child’s birth may increase the likelihood that men take paternity leave and that they take longer periods of leave (Duvander, 2014; Petts et al., 2018). In addition, having a salient father identity and prioritizing involved fathering may be associated with more positive parental relationship outcomes if fathers are viewed as more equal coparents (Goldberg, 2015; Maume & Sebastian, 2012; Pragg & Knoester, 2017).

Religious parents may be more likely to take leave and report better relationships, but may also be more likely to adhere to traditional gender roles and maintain gendered disparities in domestic work (DeMaris, Mahoney, & Pargament, 2011; Wolfinger & Wilcox, 2008). Additional factors such as family size and child’s age and gender may also affect leave-taking (e.g., married fathers may be more likely to take leave) and parental relationships (e.g., parents with fewer children may report better relationships) (Petts et al., 2018; Twenge et al., 2003).

DATA AND METHODS

Data

Data for this study comes from the Early Childhood Longitudinal Study, Birth Cohort (ECLS-B). The ECLS-B contains a nationally representative sample of children born in the U.S. in 2001. Information was collected from parents when their children were approximately 9 months (W1), 2 years (W2), 4 years (W3), and 5 years old (W4). Data from W1 is used for this study to assess whether paternity leave is associated with parental relationship satisfaction and conflict in the year following a child’s birth. The sample is restricted to couples in which mothers and fathers reside together, both mothers and fathers completed the self-administered questionnaire (which contains the questions on parental relationships), fathers were employed at the time of the child’s birth and at W1 (to be eligible for paternity leave), and mothers answered the questions on paternity leave. These restrictions result in a sample size of 4,700 couples.

Paternity Leave

In the W1 survey, mothers reported on whether fathers took any time off for the birth of their child and how many weeks of leave (paid or unpaid) fathers took. Although mothers were asked specifically about whether fathers took leave and for how long (and reasons provided for not taking leave are consistent with the literature on paternity leave), there is no information on whether fathers took time off using FMLA, an employer family leave program, vacation/sick leave, or some other means. Thus, we conceptualize paternity leave as taking time off work for the birth of a child (regardless of the actual type of leave used), and focus on two indicators. Paternity leave-taking indicates whether fathers took leave (1 = yes). Length of paternity leave indicates the number of weeks of leave that fathers took (no leave, less than one week, 1 week, 2 weeks, 3 weeks, 4 weeks, or 5 or more weeks). Alternative ways of operationalizing length of paternity leave (i.e., series of dummy variables, adjusting the capped length, and including a quadratic term) produced results that are largely consistent with those presented here.

Parental Relationship Satisfaction and Relationship Conflict

We focus on two dimensions of parental relationships. Measures are taken from mothers’ and fathers’ individual responses at W1, but they are treated as dyad-level parent variables in this study. First, relationship satisfaction indicates parents’ ratings of their marriage/relationship with the other parent (1 = not too happy to 3 = very happy). Second, relationship conflict is taken from each parent’s responses to questions about how often (1 = never to 4 = often) they have arguments with their spouse/partner about: chores and responsibilities, your child(ren), money, not showing love and affection, sex, religion, leisure time, drinking, other women or men, and in-laws (α > .78). The mean response is used as the indicator. Supplementary analyses restricting this measure to key areas of conflict (chores and responsibilities, children, and leisure time) produced results that are largely similar to those presented here. Thus, we present results using the more comprehensive indicator of relationship conflict to more fully capture parental relationship dynamics (and the measure used here has higher internal consistency).

Mothers’ Work Status

Variables from W1 were included to indicate mothers’ work statuses before and after the child’s birth. Mother worked prior to birth is a dichotomous variable indicating mothers who worked in paid labor in at least 8 of the 12 months prior to the child’s birth (results using other cutoff points such as 6 of 12 months were largely consistent with those presented here). Mothers’ work statuses at W1 are categorized as (a) mother does not work, (b) works part-time (less than 35 hours a week), or (c) works full-time (35 hours a week or more, used as reference category).

Other Factors

A number of other variables from W1 survey responses were also included. These include separate indicators for fathers’ and mothers’ income (ranging from 0 = less than $10,000 to 8 = $100,000 or more) and for each parent’s educational attainment (1 = did not complete high school to 4 = college degree). Fathers’ work hours is categorized as (a) part-time (less than 35 hours a week), (b) full-time (35–44 hours a week, used as reference category), or (c) more than full time (45 hours a week or more). Fathers’ occupation type is categorized as (a) professional (used as reference category), (b) labor, (c) service, (d) sales, or (e) other occupational type. Race/ethnicity for each parent is coded as (a) White (used as reference category), (b) Black, (c) Latino, or (d) other race/ethnicity. We also control for number of other children, whether the focal child is the father’s first child, whether parents are married or cohabiting (married is used as the reference category), child age, and child gender (1 = male). We also include separate indicators for each parent’s age and religious participation (0 = never to 4 = every week or more).

We also include variables that indicate fathers’ engagement and attitudes regarding parenting roles. Father engagement indicates fathers’ reports of how often (0 = not at all to 3 = every day) they engage in nine activities with their child including reading, telling stories, singing songs, playing games like peekaboo, playing outside, feeding child, putting child to sleep, bathing child, and helping child get dressed (α = .76). The mean is used as the scale score. Fathering competence indicates fathers’ perceptions of themselves as fathers (1 = not very good at being a father to 5 = a very good father). Traditional gender attitudes indicates that fathers agree (1 = yes) that it is more important for fathers to focus on providing while mothers care for the family. Importance of father involvement indicates fathers’ agreement (1 = strongly disagree to 4 = strongly agree) to four statements: (a) it is essential for the child’s well-being that fathers spend time playing with children, (b) the way a father treats his baby has long-term effects on the child, (c) a father should be as heavily involved as the mother in the care of the child, and (d) fatherhood is a rewarding experience (α = .59). The mean response is used. Finally, prenatal involvement is a dichotomous variable indicating whether fathers did all six of the following activities prior to the child’s birth (1 = yes): discuss the pregnancy, see an ultrasound, listen to the baby’s heartbeat, feel the baby move, attend birth class with the mother, and buy things for the baby. As such, this variable captures fathers that had especially high levels of fathering commitments before the child’s birth (95% of fathers engaged in at least 4 activities).

Analytic Strategy

Because we utilize parents’ dyad-level data, multilevel models are used as mothers and fathers are nested within couples (Kenny, Kashy, & Cook, 2006). Initial analyses (not shown) suggest that 60% of the variation in relationship satisfaction and 56% of the variation in relationship conflict occurs within couples, suggesting that a dyadic approach that accounts for nonindependence is warranted. We used the framework of the actor-partner interdependence model (APIM) to assess associations between individual-level (e.g., race/ethnicity, age, education) and shared couple-level factors (which includes variables such as relationship status and child age, as well as information reported by only one parent such as whether mothers worked prior to the child’s birth and father engagement levels) with parents’ reports of relationship satisfaction and conflict (Kenny et al., 2006). Separate predictor variables for each parent are included (when both mothers’ and fathers’ reports are available) to assess the relative influence of both mothers’ and fathers’ characteristics on reports of relationship satisfaction and conflict (e.g., mothers’ characteristics may be associated with both mothers’ and fathers’ reports of relationship satisfaction and conflict). We report results using mothers’ and fathers’ characteristics (instead of actor and partner effects) for ease of interpretation. Gender is included as the distinguishing variable in each model (1 = female/mother) to assess variations in fathers’ and mothers’ relationship satisfaction and conflict (Kenny et al., 2006).

Multilevel ordered logistic models are used to assess whether paternity leave-taking and length of paternity leave are associated with relationship satisfaction; diagnostic analyses suggest that the proportional odds assumption is not violated and thus multilevel ordered logistic models are appropriate. Multilevel linear models are used to assess whether paternity leave is associated with relationship conflict, given the continuous nature of this dependent variable. Interaction effects between gender, mothers’ work statuses, and paternity leave are then assessed; we consider both two-way (e.g., gender × paternity leave) and three-way interactions (e.g., gender × mothers’ work status × paternity leave). Interaction effects that are statistically significant (p < .05) are presented. All continuous control variables were mean centered. Less than 5% of cases include missing data (with the exception of fathers’ income – over 10% of cases are missing). To account for this, multiple imputation from ten imputed models is used in all analyses.

Selection

Although there is reason to believe that paternity leave may influence parental relationships, there may be selection effects that lead fathers to take leave (and longer periods of leave) as well as promote better parental relationships. We consider a couple of approaches to assess this possibility.

First, propensity score matching (PSM) is used to assess whether associations between paternity leave-taking and parental relationship outcomes are due to selection. Ideally, randomized trials are used to determine causal effects as randomization reduces the likelihood that the effect of a treatment on an outcome is due to confounding factors (either observed or unobserved). However, such approaches are often not possible in observational studies. PSM models approximate an experimental design by grouping respondents together who differ in whether or not they receive the treatment (i.e., took paternity leave) but are similar on all other observed confounders (Austin, 2011; Rosenbaum & Rubin, 1983). As such, PSM models minimize the likelihood that the observed relationship between the treatment and outcome is due to systematic (or selection) bias. However, in contrast to randomized trials, PSM models cannot account for biases due to unobserved factors.

To estimate propensity scores, we first ran a logistic regression model using control variables that were likely unchanged between the birth of the child and W1 interview (i.e., education, fathers’ work hours and occupation, parents’ age, race/ethnicity, marital status, number of other children, religious participation, and prenatal involvement) to match fathers in the treatment (took leave) and control (did not take leave) groups and generate propensity scores (i.e., the probability of taking leave). Respondents with the closest propensity scores were matched. We then assessed whether the assumption for common support was met (i.e., that propensity scores overlap between the treatment and control groups), and omitted cases in which this assumption was not met. Pre- and post-tests suggest that balance was achieved (i.e., covariates did not differ statistically between the treatment and control groups). Finally, the propensity scores were used to estimate average treatment effects on the treated for each dimension of parental relationships, and these are reported in the text. We also considered weighting the regression results by the propensity to take leave. Results from these weighted models were similar to those reported here, but these models are not shown due to increased bias in weighted regression estimates (Freedman & Berk, 2008).

Second, augmented inverse propensity weighted (AIPW) estimators are used to assess whether associations between length of paternity leave and parental relationships are due to selection. Similar to PSM, this approach estimates average treatment effects accounting for covariates that may select people into certain treatments (i.e., lengths of paternity leave), yet differs in that AIPW estimators can be used when there are multiple treatments (Cattaneo, 2010). We followed a similar process as described for PSM, and diagnostic analyses suggest that the assumptions needed to utilize AIPW estimators were met. These results are reported in the text.

RESULTS

Summary statistics are included in Table 1, and separate mean values are also reported by whether fathers took paternity leave. Results suggest that reports of relationship satisfaction were high for both mothers and fathers. Also, mothers and fathers reported higher relationship satisfaction when fathers took paternity leave, on average. However, in contrast to our expectations, taking paternity leave did not seem to be related to mothers’ or fathers’ reports of relationship conflict. It is also notable that most fathers took paternity leave (88%), but took relatively short leaves, on average (1.5 weeks). In fact, of the fathers who took leave, the majority took one week or less (63%) and only 5% of fathers took 5 weeks or more off. Moreover, fathers who took leave were more likely to have higher SES, were more involved prior to the child’s birth, and had more positive fathering attitudes than fathers who did not take leave. Finally, mothers reported higher average levels of relationship satisfaction and lower levels of relationship conflict than fathers (t-tests indicate significance at p < .001).

Table 1.

Summary Statistics

M SD Min Min M (no leave) M (took leave)
Relationship Satisfaction (Mother) 2.76 0.45 1 3 2.68 2.78***
Relationship Satisfaction (Father) 2.70 0.49 1 3 2.62 2.71***
Relationship Conflict (Mother) 1.77 0.46 1 4 1.78 1.77
Relationship Conflict (Father) 1.86 0.51 1 4 1.88 1.86
Paternity Leave-Taking 0.88 - 0 1 0.00 1.00***
Length of Paternity Leave 1.35 1.21 0 5 0.00 1.53***
Other Factors
Mother Worked Prior to Birth 0.55 - 0 1 0.52 0.56
Mother Does Not Work 0.47 - 0 1 0.48 0.47
Mother Works Part-Time 0.20 - 0 1 0.17 0.21
Mother Works Full-Timea 0.33 - 0 1 0.35 0.32
Mother Education 3.21 0.89 1 4 2.94 3.25**
Father Education 2.93 1.02 1 4 2.60 2.97***
Father Works Part-Time 0.07 - 0 1 0.10 0.06***
Fathers Works Full-Timea 0.43 - 0 1 0.43 0.43
Father Works more than Full-Time 0.50 - 0 1 0.47 0.51*
Professional Occupationa 0.37 - 0 1 0.28 0.39***
Labor Occupation 0.37 - 0 1 0.45 0.36**
Sales Occupation 0.07 - 0 1 0.08 0.07
Service Occupation 0.17 - 0 1 0.17 0.16
Other Occupation 0.02 - 0 1 0.03 0.02
Father Income 4.04 1.97 1 8 3.51 4.11***
Mother Income 1.93 1.44 1 8 1.85 1.95
Mother Age 29.81 5.86 18 51 29.17 29.90**
Father Age 32.36 6.56 18 66 31.98 32.42
Father is Whitea 0.58 - 0 1 0.48 0.59***
Father is Black 0.07 - 0 1 0.10 0.07*
Father is Latino 0.15 - 0 1 0.25 0.14***
Father is Other Race/Ethnicity 0.20 - 0 1 0.17 0.20
Mother is Whitea 0.56 - 0 1 0.48 0.57***
Mother is Black 0.07 - 0 1 0.09 0.06*
Mother is Latino 0.15 - 0 1 0.24 0.14***
Mother is Other Race/Ethnicity 0.22 - 0 1 0.19 0.22
Cohabiting 0.13 - 0 1 0.20 0.12***
Number of Other Children 1.07 1.13 0 5 1.23 1.05***
First-Time Father 0.37 - 0 1 0.34 0.38
Child Age 10.36 1.78 6 22 10.40 10.35
Child is Male 0.52 - 0 1 0.53 0.52
Father Religious Participation 1.88 1.48 0 4 1.64 1.91***
Mother Religious Participation 2.14 1.54 0 4 1.94 2.17**
Father Engagement 1.83 0.56 0 3 1.76 1.84**
Fathering Competence 4.26 0.85 1 5 4.22 4.27
Traditional Gender Attitudes 0.10 - 0 1 0.14 0.10**
Importance of Father Involvement 3.71 0.33 1 4 3.68 3.72*
Prenatal Involvement 0.37 - 0 1 0.28 0.38***
N 4700 550 4150
Open in a new tab
a

Used as reference category

Significant differences determined by two-tailed t-tests

*

p < .05.

**

p < .01.

***

p < .001

Results from multilevel ordered logistic models that assess the associations between paternity leave and parents’ reports of relationship satisfaction are presented in Table 2. Consistent with the first hypothesis, paternity leave-taking was positively associated with parents’ reports of relationship satisfaction in Model 1 (OR = 1.51, p < .01). This association persisted in PSM models (results not shown; b = .09, p < .001). Results also show that mothers reported higher relationship satisfaction than fathers (OR = 1.66, p < .001).

Table 2.

Results from Multilevel Ordered Logistic Models Predicting Associations between Paternity Leave and Relationship Satisfaction

1 2 3
Variable OR SE B OR SE B OR SE B
Paternity Leave-Taking 1.51** 0.20
Length of Paternity Leave 1.06 0.04 1.00 0.04
Other Factors
Gender 1.66*** 0.10 1.66*** 0.10 1.42*** 0.13
Mother Worked Prior to Birth 1.00 0.10 1.00 0.10 1.00 0.10
Mother Does Not Work 0.95 0.12 0.95 0.12 0.95 0.12
Mother Works Part-Time 0.84 0.10 0.84 0.10 0.84 0.10
Mother Education 0.89 0.08 0.89 0.08 0.89 0.08
Father Education 1.02 0.09 1.02 0.09 1.02 0.09
Father Works Part-Time 0.78 0.15 0.78 0.15 0.78 0.15
Father Works more than Full Time 1.04 0.10 1.04 0.10 1.04 0.10
Labor Occupation 0.80† 0.11 0.80† 0.11 0.80† 0.11
Sales Occupation 1.15 0.22 1.15 0.22 1.15 0.22
Service Occupation 0.80 0.12 0.80 0.12 0.80 0.12
Other Occupation 1.31 0.44 1.29 0.44 1.29 0.44
Father Income 1.04 0.04 1.04 0.04 1.04 0.04
Mother Income 0.93† 0.04 0.92* 0.04 0.92* 0.04
Mother Age 0.98 0.01 0.98 0.01 0.98 0.01
Father Age 0.99 0.01 0.99 0.01 0.99 0.01
Father is Black 0.57† 0.17 0.56† 0.17 0.56† 0.17
Father is Latino 0.98 0.19 0.97 0.19 0.97 0.19
Father is Other Race/Ethnicity 0.79 0.12 0.79 0.12 0.79 0.12
Mother is Black 0.65 0.20 0.65 0.20 0.65 0.20
Mother is Latino 0.87 0.17 0.87 0.17 0.87 0.17
Mother is Other Race/Ethnicity 1.18 0.18 1.18 0.18 1.18 0.18
Cohabiting 0.36*** 0.05 0.36*** 0.05 0.36*** 0.05
Number of Other Children 1.00 0.06 1.00 0.06 1.00 0.06
First-Time Father 1.20 0.16 1.19 0.16 1.19 0.16
Child Age 0.96 0.02 0.96 0.02 0.96 0.02
Child is Male 0.90 0.08 0.90 0.08 0.90 0.08
Father Religious Participation 1.17*** 0.05 1.17*** 0.05 1.17*** 0.05
Mother Religious Participation 1.02 0.04 1.02 0.04 1.02 0.04
Father Engagement 1.39*** 0.13 1.39*** 0.13 1.39*** 0.13
Fathering Competence 2.04*** 0.11 2.04*** 0.11 2.04*** 0.11
Traditional Gender Attitudes 0.90 0.14 0.90 0.14 0.90 0.14
Importance of Father Involvement 1.08 0.16 1.08 0.16 1.08 0.16
Prenatal Involvement 1.20† 0.12 1.21† 0.12 1.21† 0.12
Interactions
Length of Leave × Gender 1.13* 0.06
Log-likelihood −3425 −3425 −3424
Open in a new tab

N = 4700 (9400 observations).

†

p < .10.

*

p < .05.

**

p < .01.

***

p < .001

In contrast to the expectations of the second hypothesis, length of paternity leave was not associated with parents’ reports of relationship satisfaction, as shown in Model 2 of Table 2. However, there is evidence that this association varied by gender. As shown in Model 3 and in support of the third hypothesis, longer periods of leave were more likely to be associated with more positive reports of relationship satisfaction among mothers than among fathers (OR = 1.13, p < .05). These results persisted in AIPW models and suggest that an additional week of paternity leave was associated with approximately a 3% increase in mothers’ perceptions of relationship satisfaction (results not shown). This finding is illustrated in Figure 1, which shows predicted probabilities of each parent reporting that they are very happy in their relationship by length of paternity leave (with all other factors set to mean values). As shown in Figure 1, longer leaves are associated with a greater likelihood that mothers reported being very happy in their relationship (.83 predicted probability in families in which fathers took at least 4 weeks of leave compared to .77 predicted probability in families in which fathers did not take leave). In contrast, length of leave was unrelated to fathers’ perceptions of relationship satisfaction. Overall, results from analyses on relationship satisfaction provide support for hypotheses 1 and 3.

Figure 1.

Figure 1.

Open in a new tab

Predicted Probabilities of Parents Reporting they are Very Happy in their Relationship by Length of Paternity Leave

Results from multilevel models that assess the associations between paternity leave and parents’ reports of relationship conflict are presented in Table 3. In contrast to Hypotheses 1 and 2, neither paternity leave-taking nor length of paternity leave was directly associated with reports of relationship conflict, as shown in Models 1 and 2. Nonsignificant results were also found in selection models (results not shown). Mothers also reported lower relationship conflict than fathers (b = −0.09, p < .001).

Table 3.

Results from Multilevel Linear Models Predicting Associations between Paternity Leave and Relationship Conflict

1 2 3 4
Variable B SE B B SE B B SE B B SE B
Paternity Leave-Taking −0.01 0.02 −0.00 0.03
Length of Paternity Leave −0.01 0.00 0.01 0.01
Other Factors
Gender −0.09*** 0.01 −0.09*** 0.01 −0.14*** 0.03 −0.09*** 0.01
Mother Worked Prior to Birth 0.01 0.01 0.01 0.01 0.01 0.04 0.04* 0.02
Mother Does Not Work −0.07*** 0.02 −0.07*** 0.02 −0.07** 0.02 −0.07** 0.02
Mother Works Part-Time −0.00 0.02 −0.00 0.02 0.00 0.02 0.00 0.02
Mother Education 0.04** 0.01 0.04** 0.01 0.04** 0.01 0.04** 0.01
Father Education −0.01 0.01 −0.01 0.01 −0.01 0.01 −0.01 0.01
Father Works Part-Time 0.05† 0.03 0.05† 0.03 0.05† 0.03 0.05† 0.03
Father Works more than Full Time 0.03* 0.01 0.03* 0.01 0.03* 0.01 0.03* 0.01
Labor Occupation 0.02 0.02 0.02 0.02 0.02 0.02 0.02 0.02
Sales Occupation −0.00 0.02 −0.00 0.02 −0.00 0.02 −0.00 0.02
Service Occupation 0.01 0.02 0.01 0.02 0.01 0.02 0.01 0.02
Other Occupation −0.02 0.04 −0.02 0.04 −0.02 0.04 −0.02 0.04
Father Income 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00
Mother Income −0.00 0.01 −0.00 0.01 −0.00 0.01 −0.00 0.01
Mother Age 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00
Father Age −0.00* 0.00 −0.00* 0.00 −0.00* 0.00 −0.00* 0.00
Father is Black 0.02 0.04 0.02 0.04 0.02 0.04 0.02 0.04
Father is Latino −0.06* 0.03 −0.06* 0.03 −0.06* 0.03 −0.06* 0.03
Father is Other Race/Ethnicity 0.03 0.02 0.03 0.02 0.03 0.02 0.03 0.02
Mother is Black 0.07 0.04 0.07 0.04 0.07 0.04 0.07 0.04
Mother is Latino −0.02 0.03 −0.02 0.03 −0.02 0.03 −0.02 0.03
Mother is Other Race/Ethnicity −0.01 0.02 −0.01 0.02 −0.01 0.02 −0.01 0.02
Cohabiting 0.06** 0.02 0.06** 0.02 0.06** 0.02 0.06** 0.02
Number of Other Children 0.01 0.01 0.01 0.01 0.01 0.01 0.01 0.01
First-Time Father −0.07** 0.02 −0.07** 0.02 −0.07*** 0.02 −0.07*** 0.02
Child Age 0.01** 0.00 0.01** 0.00 0.01** 0.00 0.01** 0.00
Child is Male 0.03* 0.01 0.03* 0.01 0.03* 0.01 0.03* 0.01
Father Religious Participation −0.02** 0.01 −0.02** 0.01 −0.02** 0.01 −0.02** 0.01
Mother Religious Participation 0.01 0.01 0.01 0.01 0.01 0.01 0.01 0.01
Father Engagement −0.00 0.01 −0.00 0.01 −0.00 0.01 −0.00 0.01
Fathering Competence −0.09*** 0.01 −0.09*** 0.01 −0.09*** 0.01 −0.09*** 0.01
Traditional Gender Attitudes 0.05* 0.02 0.05* 0.02 0.05* 0.02 0.05* 0.02
Importance of Father Involvement −0.07*** 0.02 −0.07*** 0.02 −0.07** 0.02 −0.07** 0.02
Prenatal Involvement −0.01 0.01 −0.01 0.01 −0.01 0.01 −0.01 0.01
Interactions
Paternity Leave × Gender 0.05 0.04
Paternity Leave × Mother Worked 0.00 0.05
Gender × Mother Worked 0.10* 0.05
Length of Leave × Mother Worked −0.02* 0.01
Paternity Leave × Gender × Mother Worked −0.10* 0.05
Log-likelihood −3716 −3716 −3711 −3714
Open in a new tab

N = 4700 (9400 observations).

†

p < .10.

*

p < .05.

**

p < .01.

***

p < .001

Yet, there is some support for Hypotheses 3 and 4 in Table 3. As shown in Model 3, the association between paternity leave-taking and parents’ reports of relationship conflict appears to be moderated by gender and whether mothers worked prior to the child’s birth (b = −0.10, p < .05). This association is illustrated in Figure 2, and shows that among families in which mothers consistently worked in the year prior to birth, paternity leave-taking was associated with lower conflict as reported by mothers, on average. In contrast, if mothers did not work in the year prior to birth, they reported higher levels of conflict if fathers took leave, compared to if fathers did not take leave. Also, leave-taking appears to be unrelated to fathers’ reports of relationship conflict (and this association does not vary by mothers’ work status). Overall, results in Model 3 of Table 3 (and Figure 2) provide additional evidence that paternity leave-taking appears to be more closely associated with mothers’ – and especially employed mothers’ – reports of relationship conflict than fathers’ reports.

Figure 2.

Figure 2.

Open in a new tab

Predicted Values of the Associations between Paternity Leave-Taking and Parents’ Reports of Relationship Conflict by Mothers’ Work Statuses

As shown in Model 4 of Table 3, there is also some evidence that although length of paternity leave was unrelated to relationship conflict, this association varied by mothers’ work status (b = −0.02, p < .05). This association is illustrated in Figure 3, and shows that among families in which mothers consistently worked in paid labor in the year prior to the child’s birth, length of paternity leave was negatively associated with parents’ reports of parental conflict. In contrast, when mothers did not work in paid labor pre-birth, length of paternity leave was positively associated with parents’ reports of relationship conflict.

Figure 3.

Figure 3.

Open in a new tab

Predicted Values of the Association between Length of Paternity Leave and Parents’ Reports of Relationship Conflict by Mothers’ Work Statuses

Sensitivity Analyses

Although we anticipated that paternity leave would be especially associated with lower levels of relationship conflict when mothers were employed in paid labor, we did not anticipate that paternity leave would be associated with higher levels of relationship conflict when mothers were unemployed, on average. There was also a need to more fully consider whether mothers’ employment transitions before and after birth mattered. For example, perhaps mothers who planned to transition into paid employment post-birth would especially appreciate paternity leave-taking. Thus, we sought to further investigate the evidence of interaction effects by conducting sensitivity analyses that involved carefully considering parents’ pre- and post-birth work statuses (e.g., unemployed prior to birth and full-time work post-birth, continually unemployed, etc.), categorizing families into different combinations of work statuses, verifying the distributions of families into each of these categories, and then re-running our analyses.

The results from the sensitivity analyses (results not shown) upheld our presentation of the interaction effects shown in Figure 2; in all situations, regardless of mothers’ post-birth work statuses, paternity leave-taking – including taking longer periods of leave – was associated with higher levels of relationship conflict when mothers had not been consistently employed pre-birth and lower levels of relationship conflict when mothers had been consistently employed pre-birth.

A potential explanation for why leave-taking is positively associated with conflict among unemployed mothers is that paternity leave-taking may be viewed as particularly problematic and/or threatening for mothers who specialize in domestic roles. Unemployed mothers may also have higher standards for domestic work and childcare than their partners, compared to the gaps in standards between employed mothers and their partner. Unemployed mothers may then be more likely to engage in critical or gatekeeping behaviors that may amplify relationship conflict. Alternatively, for single-earner families dealing with the expenses associated with having a new child, paternity leave-taking may contribute to financial stress and conflict between parents.

It is also important to note that mothers who were not employed in paid labor before their child’s birth generally reported lower relationship conflict than employed mothers (regardless of their post-birth work status). Also, mothers who were not employed in paid labor after the birth generally reported lower relationship conflict than mothers who are employed (regardless of whether it is part-time or full-time employment). However, a nuance emerges in interpreting the interaction effects that involve mothers’ work statuses and length of leave-taking. As shown in Figure 3 (and upheld in sensitivity analyses), there is a cross-over point at which longer leaves were associated with reports of relationship conflict that are even lower than reports from couples where mothers were not employed pre-birth (and whose partners did not take leave). Generally, leaves that are slightly more than two weeks long seem to be associated with lower levels of conflict not only among families in which mothers were employed pre-birth—but also in comparison to reports of conflict in families in which mothers were not employed pre-birth.

DISCUSSION

Although research has consistently found evidence that paternity leave is positively associated with subsequent father involvement, few studies have focused on other implications that paternity leave may have for families within the context of the U.S. Understanding the consequences of leave-taking for American families is important as cities, states, and the federal government consider legislation on paid family leave. This study partially addresses this gap in the literature by focusing on whether paternity leave-taking and length of paternity leave are associated with parental relationship satisfaction and relationship conflict, and whether these associations vary between mothers’ and fathers’ reports and by mothers’ work statuses. Overall, there is mixed evidence supporting our hypotheses. However, there is fairly consistent support for the expectation that paternity leave-taking and longer lengths of leave are associated with more positive perceptions of relationships among mothers—especially in families in which mothers were consistently employed prior to their child’s birth.

First, there is limited evidence in support of the first and second hypotheses that anticipated that paternity leave-taking and longer leaves would lead to higher levels of relationship satisfaction and lower levels of relationship conflict among parents. Paternity leave-taking is positively associated with parents’ reports of relationship satisfaction, but unrelated to parents’ reports of relationship conflict in the additive models. Length of paternity leave is also unrelated to both relationship satisfaction and conflict. Nonetheless, taking paternity leave may signal a commitment by fathers to not just fulfill the breadwinner role, but to also be an engaged parent and coparent (Albiston & O’Connor, 2016; Almqvist & Duvander, 2014; Pragg & Knoester, 2017; Rehel, 2014). Paternity leave may also provide mothers and fathers with time to learn how to coparent together and establish expectations about what they feel the division of labor within the household should be (Rehel, 2014). As such, paternity leave-taking may increase the likelihood that mothers perceive fathers as committed to being an involved parent, and both mothers and fathers may also perceive the division of household labor as more equitable, resulting in higher relationship satisfaction (Kotsadam & Finseraas, 2011; Milkie et al., 2002).

Second, although paternity leave-taking and length of paternity leave are not found to be directly associated with relationship conflict, there is evidence in support of the fourth hypothesis that anticipated that the associations between paternity leave and parental relationship outcomes would vary by mothers’ work status. Specifically, paternity leave-taking appears to only be associated with mothers’ reports of lower relationship conflict in families in which mothers worked prior to the child’s birth. Similarly, length of leave appears to only be associated with lower levels of parents’ reports of relationship conflict in families in which mothers worked pre-birth. These findings may be due to an increased likelihood of experiencing role conflicts and overloads for working mothers; mothers who are employed prior to the birth of a child face the stresses of balancing when (or if) to return to work (to minimize wage penalties and potential discrimination) with the demands of intensive mothering (England et al., 2016; Hays, 1996; Pedulla & Thébaud, 2015), whereas women who are not employed can focus more energy on mothering and other domestic tasks. These stresses may also increase work-family conflict for fathers (Winslow, 2005). Taking paternity leave, and longer leaves, may be more beneficial in alleviating relationship conflict among couples with greater role conflicts and overloads. In particular, paternity leave may enable fathers to alleviate some of the role conflicts and overloads that working mothers experience by sharing in childcare and other domestic work immediately after a child’s birth (Gault et al., 2014; Kotsadam & Finseraas, 2011; Rehel, 2014). Parents can also use the time offered by paternity leave to negotiate role expectations, which may help to reduce conflict (Kotsadam & Finseraas, 2011; Rehel, 2014). Greater access to [paid] leave policies may be particularly beneficial to low-SES families who face economic and social constraints that make balancing work and family life particularly challenging (Williams et al., 2013).

Finally, results provide considerable support for the third hypothesis that paternity leave is more closely associated with mothers’ reports of relationship satisfaction and conflict than fathers’ reports. For example, length of paternity leave is associated with more positive perceptions of relationship satisfaction among mothers, and paternity leave-taking is associated with mothers’ reports of lower relationship conflict only among mothers who were employed prior to the child’s birth. In contrast, results suggest that paternity leave-taking and length of leave may be largely unrelated to fathers’ perceptions of their relationships with mothers. Previous research suggests that mothers’ perceptions of relationships are more likely to be negatively influenced by role traditionalization than fathers’ perceptions, and this is due to mothers often having to do a greater share of the household and childcare tasks than they feel is equitable (Cowan et al., 1985; Milkie et al., 2002; Newkirk et al., 2017; Twenge et al., 2003). To the extent that paternity leave-taking and length of paternity leave may help to lessen the degree to which role traditionalization occurs by enabling fathers to become more engaged parents immediately starting at the time of the child’s birth (Almqvist & Duvander, 2014; Hosking et al., 2010; Pragg & Knoester, 2017), mothers may be more likely to perceive the division of household labor as fair (Milkie et al., 2002; Yavorsky et al., 2015). Thus, paternity leave may be particularly beneficial for mothers’ perceptions of relationships by reducing role conflicts and overloads and enabling parents to more feasibly pursue the egalitarian ideals that many couples support (Kotsadam & Finseraas, 2011; Pedulla & Thébaud, 2015; Twenge et al., 2003).

Although this study has many strengths, there are some limitations to note. First, mothers did not provide information about the type of leave fathers took or what (if any) paternity leave programs were available to them. Thus, we are unable to determine whether fathers are using workplace paternity leave programs or other types of leave (unpaid leave through the Family and Medical Leave Act, vacation or sick days, etc.). We considered various reasons that may prompt fathers to take time off work in supplementary analyses such as whether the child had an extended stay in the hospital (or NICU) after birth, whether the mother had labor complications, whether the mother had a cesarean section, and whether the mother has health problems that limit her ability to work, and none of these factors altered the associations presented here (in both the full models and selection models). Regardless, having information on specific types of leave is important in assessing the potential policy implications of the evidence of associations between paternity leave and parental relationship outcomes found in this study.

Second, this study attempted to account for the possibility of selection effects due to observed characteristics, but this study does not account for possible selection effects due to unobserved factors (e.g., availability of paternity leave programs). Moreover, the first wave of data was collected approximately nine months after the child was born, and we do not have data on key characteristics (fathers’ work characteristics, parenting attitudes, etc.) prior to the child’s birth. Having longitudinal data collected pre- and post-birth would be ideal, and future research should continue to incorporate additional factors that may confound the association between paternity leave and parental relationship outcomes to further minimize potential selection effects. Nevertheless, we attempted to address potential selection effects to the extent that we were able.

Third, we are not able to account for the influence of coparenting and mothers’ gender and parenting attitudes in this study. Although other studies have examined coparenting using ECLS-B data (e.g., Hohmann-Marriott, 2011), this information is not included in the W1 survey. Additional waves of data were examined in supplementary analyses, but paternity leave was only associated with parental relationship satisfaction and conflict when children are infants. As such, we are unable to incorporate the coparenting measures. We do include father engagement to assess whether this may explain the link between paternity leave and parental relationships, but father engagement was not a mediating factor (full results available upon request). In addition, information on mothers’ gender ideology and parenting attitudes are not available in the ECLS-B. We are able to account for fathers’ attitudes, but the associations between paternity leave and mothers’ perceptions of relationships likely vary by mothers’ gender ideology (i.e., whether she views a traditional division of labor as best) and how mothers view the role of fathers. Future research should incorporate measures of coparenting and mothers’ gender ideology to more fully assess the association between paternity leave and parents’ perceptions of their relationships.

Fourth, this study utilizes data from only one wave. The use of cross-sectional data prevents a comprehensive assessment of reverse causality (i.e., that fathers are more likely to take leave when parents have higher quality relationships). Also, this study only considers short-term associations between paternity leave and perceptions of parental relationships. This may suggest that any relationship benefits stemming from paternity leave may be short-lived, but future research should further assess this question using other longitudinal data. In addition, there is evidence that mothers’ work statuses prior to the child’s birth moderate the association between paternity leave and parental conflict, but interaction terms involving mothers’ employment statuses post-birth were not statistically significant. Given that the first wave of data was collected when children were approximately 9 months old, some mothers may not have returned to work yet. Again, looking at these associations longitudinally would be beneficial.

Overall, this study makes important contributions to our understanding of the potential benefits of parental leave policies by highlighting associations between paternity leave and parents’ reports of relationship satisfaction and relationship conflict after a child’s birth within the U.S. Results suggest that paternity leave-taking and length of leave are especially associated with mothers’ reports of greater relationship satisfaction and pre-birth working mothers’ reports of lower relationship conflict. As such, this study offers evidence that paternity leave may help to reduce gender inequalities and strengthen family relationships. Thus, there may be reason to expand support for paternity leave opportunities for all U.S. families.

Acknowledgements:

Research was supported by the Eunice Kennedy Shriver National Institute of Child Health & Human Development of the National Institutes of Health under Award Number R03HD087875. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health.

REFERENCES

  1. Albiston C & O’Connor LT (2016). Just leave. Harvard Women’s Law Journal, 39, 1–65. [Google Scholar]
  2. Almqvist A & Duvander A (2014). Changes in gender equality? Swedish fathers’ parental leave, division of childcare and housework. Journal of Family Studies, 20, 19–27. 10.5172/jfs.2014.20.1.19. [DOI] [Google Scholar]
  3. Aumann K, Galinsky E, & Matos K (2011). The new male mystique. Families and Work Institute. [Google Scholar]
  4. Austin PC (2011). An introduction to propensity score methods for reducing the effects of confounding in observational studies. Multivariate Behavioral Research, 46, 399–424. 10.1080/00273171.2011.568786. [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Becker GS (1981). A treatise on the family. Cambridge: Harvard University Press. [Google Scholar]
  6. Bianchi SM & Milkie MA (2010). Work and family research in first decade of the 21st centrury. Journal of Marriage and Family, 72, 705–725. 10.1111/j.1741-3737.2010.00726.x. [DOI] [Google Scholar]
  7. Bünning M (2011). What happens after the ‘daddy months’? Fathers’ involvement in paid work, childcare, and housework after taking parental leave in Germany. European Sociological Review, 31, 738–748. 10.1093/esr/jcv072. [DOI] [Google Scholar]
  8. Bureau of Labor Statistics. (2018). Access to paid personal leave, December 2017. Available at: https://www.bls.gov/ebs/paid_personal_leave_122017.htm.
  9. Cattaneo MD (2010). Efficient semiparametric estimation of multi-valued treatment effects under ignorability. Journal of Econometrics, 155, 138–154. 10.1016/j.jeconom.2009.09.023. [DOI] [Google Scholar]
  10. Cowan CP, Cowan PA, Heming G, Garrett E, Coysh WS, Curtis-Boles H, & Boles III AJ (1985). Transitions to parenthood: His, hers, and theirs. Journal of Family Issues, 6, 451–481. 10.1177/019251385006004004. [DOI] [PubMed] [Google Scholar]
  11. Crohan SE (1996). Marital quality and conflict across the transition to parenthood in African American and white couples. Journal of Marriage and Family, 58, 933–944. 10.2307/353981. [DOI] [Google Scholar]
  12. Daly KJ (1996). Spending time with the kids: Meanings of family time for fathers. Family Relations, 45, 466–476. 10.2307/585177. [DOI] [Google Scholar]
  13. DeMaris A, Mahoney A, & Pargament KI (2011). Doing the scut work of infant care: Does religiousness encourage father involvement? Journal of Marriage and Family, 73, 354–368. 10.1111/j.1741-3737.2010.00811.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Dew J & Wilcox WB (2011). If momma ain’t happy: Explaining declines in marital satisfaction among new mothers. Journal of Marriage and Family, 73, 1–12. 10.1111/j.1741-3737.2010.00782.x. [DOI] [Google Scholar]
  15. Duvander A (2014). How long should parental leave be? Attitudes to gender equality, family, and work as determinants of women’s and men’s parental leave in Sweden. Journal of Family Issues, 35, 909–926. 10.1177/0192513X14522242. [DOI] [Google Scholar]
  16. England P, Bearak J, Budig MJ, & Hodges MJ (2016). Do highly paid, highly skilled women experience the largest motherhood penalty? American Sociological Review, 81, 1161–1189. 10.1177/0003122416673598. [DOI] [Google Scholar]
  17. Fellows KJ, Chiu H, Hill EJ, & Hawkins AJ (2016). Work-family conflict and couple relationship quality: A meta-analytic study. Journal of Family and Economic Issues, 37, 509–518. 10.1007/s10834-015-9450-7. [DOI] [Google Scholar]
  18. Freedman DA & Berk RA (2008). Weighting regressions by propensity scores. Evaluation Review, 32, 392–409. 10.1177/0193841X08317586. [DOI] [PubMed] [Google Scholar]
  19. Galinsky E, Aumann K, & Bond JT (2011). Times are changing: Gender and generation at work and at home. New York: Families and Work Institute. [Google Scholar]
  20. Galovan AM, Holmes EK, Schramm DG, & Lee RR (2014). Father involvement, father-child relationship quality, and satisfaction with family work: Actor and partner influences on marital quality. Journal of Family Issues, 35, 1846–1867. 10.1177/0192513X13479948. [DOI] [Google Scholar]
  21. Gault B, Hartmann H, Hegewisch A, Milli J, & Reichlin L (2014). Paid parental leave in the United States: What the data tells us about access, usage, and economic and health benefits. Institute for Women’s Policy Research: Washington, D.C. [Google Scholar]
  22. Gerson K (2010). The unfinished revolution: How a new generation is reshaping family, work, and gender in America. New York: Oxford University Press. [Google Scholar]
  23. Glass J, Simon RW, & Andersson MA (2016). Parenthood and happiness: Effects of work-family reconciliation policies in 22 OECD countries. American Journal of Sociology, 122, 886–929. 10.1086/688892. [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Goldberg JS (2015). Identity and involvement among resident and nonresident fathers. Journal of Family Issues, 36, 852–879. 10.1177/0192513X13500963. [DOI] [Google Scholar]
  25. Goode WJ (1960). A theory of role strain. American Sociological Review, 25, 483–496. 10.2307/2092933. [DOI] [Google Scholar]
  26. Haas L & Hwang CP (2008). The impact of taking parental leave on fathers’ participation in childcare and relationships with children: Lessons from Sweden. Community, Work, and Family, 11, 85–104. 10.1080/13668800701785346. [DOI] [Google Scholar]
  27. Hardie JH & Lucas A (2010). Economic factors and relationship quality among young couples: Comparing cohabitation and marriage. Journal of Marriage and Family, 75, 1141–1154. 10.1111/j.1741-3737.2010.00755.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  28. Hays S (1996). The cultural contradictions of motherhood. New Haven: Yale University Press. [Google Scholar]
  29. Hecht LM (2001). Role conflict and role overload: Different concepts, different consequences. Sociological Inquiry, 71, 111–121. 10.1111/j.1475-682X.2001.tb00930.x. [DOI] [Google Scholar]
  30. Hohmann-Marriott B (2011). Coparenting and father involvement in married and unmarried coresident couples. Journal of Marriage and Family, 73, 296–309. 10.1111/j.1741-3737.2010.00805.x. [DOI] [Google Scholar]
  31. Hosking A, Whitehouse G, & Baxter J (2010). Duration of leave and resident fathers’ involvement in infant care in Australia. Journal of Marriage and Family, 72, 1301–1316. 10.1111/j.1741-3737.2010.00766.x. [DOI] [Google Scholar]
  32. Huerta MC, Adema W, Baxter J, Han W, Lausten M, Lee R, & Waldfogel J (2014). Fathers’ leave and fathers’ involvement: Evidence from four OECD countries. European Journal of Social Security, 16, 308–346. 10.1177/138826271401600403. [DOI] [PMC free article] [PubMed] [Google Scholar]
  33. Jacobs JA & Gerson K (2004). The time divide: Work, family, and gender inequality. Cambridge: Harvard University Press. [Google Scholar]
  34. Johansson E (2010). The effect of own and spousal parental leave on earnings. Institute for Labour Market Policy Evaluation (Sweden), Working Paper 2010:4. https://www.econstor.eu/dspace/bitstream/10419/45782/1/623752174.pdf.
  35. Kalil A, Ziol-Guest KM, & Epstein JL (2010). Nonstandard work and marital instability: Evidence from the National Longitudinal Survey of Youth. Journal of Marriage and Family, 72, 1289–1300. 10.1111/j.1741-3737.2010.00765.x. [DOI] [Google Scholar]
  36. Kamp Dush CM, Yavorsky JE, & Schoppe-Sullivan SJ (2018). What are men doing while women perform extra unpaid labor? Leisure and specialization at the transitions to parenthood. Sex Roles, 78, 715–730. 10.1007/s11199-017-0841-0. [DOI] [PMC free article] [PubMed] [Google Scholar]
  37. Keizer R & Schenk N (2012). Becoming a parent and relationship satisfaction: A longitudinal dyadic perspective. Journal of Marriage and Family, 74, 759–773. 10.1111/j.1741-3737.2012.00991.x. [DOI] [Google Scholar]
  38. Kenny DA, Kashy DA, & Cook WL (2006). Dyadic data analysis. New York: The Guilford Press. [Google Scholar]
  39. Killewald A (2013). A reconsideration of the fatherhood premium: Marriage, coresidence, biology, and fathers’ wages. American Sociological Review, 78, 96–116. 10.1177/0003122412469204. [DOI] [Google Scholar]
  40. Kotsadam A & Finseraas H (2011). The state intervenes in the battle of the sexes: Causal effects of paternity leave. Social Science Research, 40, 1611–1622. 10.1016/j.ssresearch.2011.06.011. [DOI] [Google Scholar]
  41. Marsiglio W & Roy K (2012). Nurturing dads: Social initiatives for contemporary fatherhood. New York: Russell Sage Foundation. [Google Scholar]
  42. Maume DJ & Sebastian RA (2012). Gender, nonstandard work schedules, and marital quality. Journal of Family and Economic Issues, 33, 477–490. 10.1007/s10834-012-9308-1. [DOI] [Google Scholar]
  43. McClain L & Brown SL (2017). The roles of fathers’ involvement and coparenting in relationship quality among cohabiting and married parents. Sex Roles, 76, 334–345. 10.1007/s11199-016-0612-3. [DOI] [PMC free article] [PubMed] [Google Scholar]
  44. Milkie MA, Bianchi SM, Mattingly MJ, & Robinson JP (2002). Gendered division of childrearing: Ideals, realities, and the relationship to parental well-being. Sex Roles, 47, 21–38. 10.1023/A:1020627602889. [DOI] [Google Scholar]
  45. National Partnership for Women and Families. (2018). State paid family and medical leave laws, July 2018. Washington, D.C. [Google Scholar]
  46. Nepomnyaschy L & Waldfogel J (2007). Paternity leave and fathers’ involvement with their young children. Community, Work and Family, 10, 427–453. 10.1080/13668800701575077. [DOI] [Google Scholar]
  47. Newkirk K, Perry-Jenkins M, & Sayer AG (2017). Division of household and childcare labor and relationship conflict among low-income new parents. Sex Roles, 76, 319–333. 10.1007/s11199-016-0604-3. [DOI] [PMC free article] [PubMed] [Google Scholar]
  48. Parker K & Livingston G (2017). 6 facts about American fathers. Pew Research Center. [Google Scholar]
  49. Pasley K, Petren RE, & Fish JN (2014). Use of identity theory to inform fathering scholarship. Journal of Family Theory and Review, 6, 298–318. 10.1111/jftr.12052. [DOI] [Google Scholar]
  50. Pedulla DS & Thébaud S (2015). Can we finish the revolution? Gender, work-family ideals, and institutional constraint. American Sociological Review, 80, 116–139. 10.1177/0003122414564008. [DOI] [PMC free article] [PubMed] [Google Scholar]
  51. Petts RJ & Knoester C (2018). Paternity leave-taking and father engagement. Journal of Marriage and Family, 80, 1144–1162. 10.111/jomf.12494. [DOI] [PMC free article] [PubMed] [Google Scholar]
  52. Petts RJ, Knoester C, & Li Q (2018). Paid paternity leave-taking in the United States. Community, Work & Family. 10.1080/13668803.2018.1471589. [DOI] [PMC free article] [PubMed] [Google Scholar]
  53. Pragg B & Knoester C (2017). Parental leave usage among disadvantaged fathers. Journal of Family Issues, 38, 1157–1185. 10.1177/0192513X15623585. [DOI] [PMC free article] [PubMed] [Google Scholar]
  54. Rehel EM (2014). When dad stays home too: Paternity leave, gender, and parenting. Gender and Society, 28, 110–132. 10.1177/0891243213503900. [DOI] [Google Scholar]
  55. Risman BJ (1998). Gender vertigo: American families in transition. New Haven: Yale University Press. [Google Scholar]
  56. Rosenbaum PR & Rubin DB (1983). The central role of the propensity score in observational studies for causal effects. Biometrika, 70, 41–55. 10.1093/biomet/70.1.41. [DOI] [Google Scholar]
  57. Rudman LA & Mescher K (2013). Penalizing men who request a family leave: Is flexibility stigma a femininity stigma? Journal of Social Issues, 69, 322–340. 10.1111/josi.12017. [DOI] [Google Scholar]
  58. Sayer LC & Bianchi SM (2000). Women’s economic independence and the probability of divorce: A review and reexamination. Journal of Family Issues, 21, 906–943. 10.1177/019251300021007005. [DOI] [Google Scholar]
  59. Sejourne NV Beaume M, Goutaudier N, & Chabrol H (2012). The impact of paternity leave and paternal involvement in child care on maternal postpartum depression. Journal of Reproductive and Infant Psychology, 30, 135–144. 10.1080/02646838.2012.693155. [DOI] [Google Scholar]
  60. Tanaka S & Waldfogel J (2007). Effects of parental leave and working hours on fathers’ involvement with their babies: Evidence from the UK Millennium Cohort Study. Community, Work, and Family, 10, 409–426. 10.1080/13668800701575069. [DOI] [Google Scholar]
  61. Townsend NW (2002). The package deal: Marriage, work, and fatherhood in men’s lives. Philadelphia: Temple University Press. [Google Scholar]
  62. Twenge JM, Campbell WK, & Foster CA (2003). Parenthood and marital satisfaction: A meta-analytic review. Journal of Marriage and Family, 65, 574–583. 10.1111/j.1741-3737.2003.00574.x. [DOI] [Google Scholar]
  63. Waldfogel J (2006). What children need. Cambridge: Harvard University Press. [Google Scholar]
  64. Wilcox WB & Nock SL (2006). What’s love got to do with it? Equality, equity, commitment, and women’s marital quality. Social Forces, 84, 1321–1345. 10.1353/sof.2006.0076. [DOI] [Google Scholar]
  65. Williams JC, Blair-Loy M, & Berdahl JL (2013). Cultural schemas, social class, and the flexibility stigma. Journal of Social Issues, 69, 209–234. 10.1111/josi.12012. [DOI] [Google Scholar]
  66. Winslow S (2005). Work-family conflict, gender, and parenthood, 1977–1997. Journal of Family Issues, 26, 727–755. 10.1177/0192513X05277522. [DOI] [Google Scholar]
  67. Wolfinger NH & Wilcox WB (2008). Happily ever after? Religion, marital status, gender and relationship quality in urban families. Social Forces, 86, 1311–1337. 10.1353/sof.0.0023. [DOI] [Google Scholar]
  68. World Bank Group. (2018). Women, Business and the Law 2018. Washington, D.C.: World Bank. [Google Scholar]
  69. Yavorsky JE, Kamp Dush CM, & Schoppe-Sullivan SJ (2015). The production of inequality: The gender division of labor across the transition to parenthood. Journal of Marriage and Family, 77, 662–679. 10.1111/jomf.12189. [DOI] [PMC free article] [PubMed] [Google Scholar]

RESOURCES