Abstract
Enduring and accumulated advantages and disadvantages in work and family lives remain invisible in studies focusing on single outcomes. Further, single outcome studies tend to conflate labor market inequalities related to gender, race, and family situation. We combine an intersectional and quantitative life course perspective to analyze parallel work and family lives for Black and White men and women aged 22–44. Results using sequence analysis and data from the National Longitudinal Survey of Youth (NLSY79) show that White men enjoy privileged opportunities to combine work and family life and elicit specific gendered and racialized constraints for Black men and women and White women. Black women experience the strongest interdependence between work and family life: events in their work lives constrain and condition their family lives and vice versa. For Black men, stable partnerships and career success mutually support and sustain each other over the life course. In contrast, for Black women, occupational success goes along with the absence of stable partnerships. Precarious and unstable employment is associated with early single parenthood for all groups supporting instability spillovers between life domains that are most prevalent among Black women, followed by Black men. The findings highlight a sizeable group of resourceful Black single mothers who hold stable middle-class jobs and have often gone unnoticed in previous research. We conclude that economic interventions to equalize opportunities in education, employment, and earnings, particularly early in life, are more promising for reducing intersectional inequalities in work-family life courses than attempting to intervene in family lives.
Introduction
Work and family lives are closely connected, and this connection varies by gender and race. Most research on work-family inequalities focuses on single indicators, such as wage gaps related to parenthood and marriage (England et al. 2016; Glauber 2007, 2008; Killewald and Gough 2013). This single-outcome perspective ignores how the social, psychological, and economic correlates of single events, such as unemployment, are contingent on other life aspects, including preceding and succeeding employment careers and family situations. Work and family lives are dynamically interrelated in ways that can either reinforce or compensate for disadvantages across life domains and over time (Fasang and Mayer 2020).
Recent life course studies have assessed how labor market (dis)advantages associated with family events accumulate over time (Aisenbrey and Fasang 2017; Florian 2018a, 2018b; Jalovaara and Fasang 2020; Kahn, Garcia-Manglano, and Bianchi 2014; Killewald and Zhuo 2019; Sirniö, Kauppinen, and Martikainen 2017). The vast majority of quantitative studies on work-family inequalities only examine gender differences among White men and women or focus on either women or men while adding race as a control variable (Budig and Hodges 2010; DiPrete and McManus 2000; England et al. 2016; Kahn, Garcia-Manglano, and Bianchi 2014). Separating the effects of gender and race obfuscates specific privileges and disadvantages associated with overlapping categories of gender and race (Browne and Misra 2003; Choo and Ferree 2010; Collins 2015).
We take an intersectional life course perspective on gendered and racialized combinations of work and family lives from early adulthood to mid-life (ages 22–44) and ask the following questions: (1) How interdependent are work and family lives, and (2) what types of work and family lives typically co-occur for each intersectional group? We complement the large literature on single outcomes of work-family inequality by offering a long-term perspective that descriptively maps enduring and accumulated (dis)advantages over 20 years of life. The analysis uses the National Longitudinal Survey of Youth 79 and sequence analysis. Mantel coefficients (Piccarreta and Elzinga 2014; Piccarreta 2017) enable us to directly measure the strength of work-family interdependence as the correlation between longer-term processes in the two life domains for the first time (Bernardi, Huinink and Settersten Jr 2019). Multichannel sequence analysis elicits the most prevalent types of combined work and family lives (Gauthier et al. 2010; Pollock 2007).
Our contribution is three-fold. First, single-outcome studies on work-family inequality risk under- or overstating how (dis)advantages in work and family lives endure, compensate for each other, or accumulate over time. Mapping a larger picture of life course experiences provides new descriptive evidence about these (dis)advantages. Second, we extend recent developments in life course theory (Bernardi, Huinink and Settersten Jr 2019) to hypothesize degrees and types of group-specific work-family interdependence that signify social inequality in life courses. Third, we combine an intersectional and quantitative life course perspective to elicit structural inequalities in work-family life courses that are associated with, rather than caused by, identity categories of gender and race.
The analysis focuses on cohorts born between 1957 and 1964 who experienced early to mature adulthood in the 1980s, 1990s, and 2000s. These cohorts entered the labor market during the recession following the oil crises in the 1970s and built careers during the economic restructuring, de-industrialization, skill-biased technological change, and labor market polarization of the 1980s and 1990s (Kalleberg 2011). The second wave of feminism and the civil rights movement promised more equality for women and non-White men and women as they were coming of age, and they were the first cohorts in which women’s educational attainment surpassed men’s (Buchmann and DiPrete 2006). These cohorts are also the first that can be followed into mature adulthood, meaning that later family events and midcareer moves can be tracked.
Previous Research and Contributions
Most studies on work-family inequality have either focused on how economic and work characteristics affect family lives or how family events impact subsequent work lives and economic rewards. Overall, the evidence that structurally limited economic opportunities (for instance, due to racial inequality in education, jobs, and earnings) determine family trajectories is stronger than for the inverse relation (DiPrete and McManus 2000; Raley and Sweeney 2020). Low education and poverty are associated with early, high, and nonmarital fertility, lower rates of marriage, and greater family instability and complexity (Musick et al. 2009; Sweeney and Raley 2014). These associations tend to be stronger for women than for men (Raley and Sweeney 2020) and disproportionately apply to Black men and women, who are more likely to be poor and lower educated (Wilson 1987). Men’s employment and higher earnings are associated with stable marriage and later (though relatively high) fertility (Ludwig and Brüderl 2018). For women, the opposite is the case: high education and successful careers correlate with delayed fertility and elevated levels of childlessness (Blair-Loy and DeHart 2003; Raley and Sweeney 2020).
Evidence in the other direction—of how family events impact work lives—is weaker and more mixed. Marriage, divorce, and parenthood have only a small (if any) effect on White men’s subsequent employment and earnings but have larger repercussions for White women (Van Winkle and Fasang 2020) and Black men and women (DiPrete and McManus 2000; Smock 1993). Small marriage wage premiums are largely due to the selection of men and women with high earning potentials into marriage (Killewald and Gough 2013; Ludwig and Brüderl 2018). Income and employment changes after divorce are often only temporary, but women’s losses are greater and more enduring than men’s (DiPrete and McManus 2000; Leopold 2018). On average, White men lose very little or even gain economically when they divorce. White women and Black men and women lose more (Smock 1993). Black men suffer greater economic losses than White men upon divorce, because Black women tend to contribute more to family income than White women (Smock 1993).
One exception to the weak associations found between family lives and subsequent careers are the well-documented motherhood wage penalties, which are greater for White women than for Black women (Van Winkle and Fasang 2020; Glauber 2007). Fatherhood premiums are smaller than motherhood penalties, and even smaller for Black men than White men (Glauber 2008, 2018). Parenthood wage gaps are also concentrated in shorter periods of the life course for Black than for White men and women (Van Winkle and Fasang 2020). Black men and women’s labor market disadvantages, including higher wage compression and flatter wage growth, leave less scope for wage differences between parents and childless individuals to emerge (Anderson, Binder, and Krause 2003; Glauber 2007; Van Winkle and Fasang 2020; Willson 2003). For women in high-prestige occupations (who are disproportionately White), motherhood often triggers occupational (re)sorting into lower paid jobs that enable them to reconcile work and care by working part time (Abendroth, Huffman, and Treas 2014; Aisenbrey, Evertsson and Grunow 2009; Gangl and Ziefle 2009; Killewald 2013; Weeden, Cha, and Bucca 2016). Correspondingly, motherhood most deters White women from working full time, but it only moderately and briefly impacts Black women’s employment (Florian 2018a, 2018b; see also Killewald and Zhuo 2019).
Instead of studying the causal effects of specific work events on family events or vice versa, we seek to uncover longer-term interdependence between work and family lives. Enduring and accumulated inequalities in work and family lives remain hidden in studies on single outcomes. For instance, parenthood wage gaps by gender and race conflate inequality between parents and childless individuals with gender and racial inequality (Van Winkle and Fasang 2020). Black women’s motherhood penalties are lower than White women’s, as Black women are more concentrated in low paid, insecure jobs. Lower motherhood penalties indicate that Black childless women are disadvantaged compared to White childless women, not that Black mothers are better off than White mothers (Doren and Lin 2019; Van Winkle and Fasang 2020). Standard measures, such as parenthood wage gaps, are therefore not able to capture the full extent of gendered and racialized inequalities in work and family lives.
We address this gap by descriptively mapping the long-term relations between work and family lives. Specific work and family lives sustain or impede each other in different ways for Black and White men and women, and these differences indicate long-term social inequalities. Overall evidence supports the view that economic (dis)advantages in the family of origin and in work lives are stronger determinants of family lives, while the effects of family lives on work lives are weaker and often only short-lived, especially for White men. Consequently, racial differences in opportunities for combining work and family lives arise because generations of Black men and women have been concentrated in low socioeconomic positions.
Black people have consistently had lower educational attainment; higher unemployment, poverty, and incarceration rates; restricted access to stable, high-prestige jobs; and fewer options in marriage markets (Crowder and Tolnay 2000; Tomaskovic-Devey, Thomas and Johnson 2005). In 2000, approximately 10% of Black men and 15% of Black women between ages 22 and 28 completed college, compared to 31% of White women and 24% of White men (McDaniel et al. 2011, 892). Between 1985 and 1999, White men’s average hourly wage was about 30% higher than that of Black men (Western and Pettit 2005, 556/557). White men’s unemployment rate hovered around 15% compared to 30% for Black men (ibid.). Already at first labor market entrance, Black people are concentrated in low-skilled manual and service occupations that offer low and compressed wages, no parental leave benefits, no job security, few opportunities for upward mobility, flatter wage growth, and often unpredictable schedules that are difficult to reconcile with child care (Carrillo et al. 2017; Ren 2021; Tomaskovic-Devey et al. 2006; Willson 2003). Black men and women’s employment is more uncertain and interrupted—they cycle in and out of low-paid jobs (Kalleberg 2011; Tomaskovic-Devey, Thomas and Johnson 2005). Earnings inequality between Black and White men is primarily formed at labor market entry and only marginally changes in response to later work or family lives (Ren 2021), although income volatility is higher among Black families (Western et al. 2012).
Black people have historically had lower rates of home ownership and restricted access to high-quality loans and have been more exposed to predatory lending practices (Rugh and Massey 2010). When Black people are concentrated in localities that primarily offer lower-skilled manual or service jobs, discrimination in housing reinforces labor market disadvantages (McCall 2002). Black women are particularly overrepresented in nonmanual, poorly paid, insecure service occupations. White women on average hold more prestigious occupations than Black women and Black men (Browne and Misra 2003; Van Winkle and Fasang 2020).
Discrimination in dating and marriage markets further narrows options for family formation, particularly for Black women (Crowder and Tolnay 2000). As marriage has become culturally more optional, it has become more important to reach the marriage bar economically. Yet at the same time, it has become less attainable for Black men and women (Raley, Sweeney, and Wondra 2015). The Black-White gap in marriage and nonmarital birth rates peaked for our study cohorts. In 2012, only 65% of 40–44-year-old Black women had been married, compared to 90% of White women (ibid.). Marriage declined most among highly educated Black women, for whom rates of racial intermarriage remain low. In contrast, highly educated Black men’s higher rates of intermarriage reflect greater family options, while at the same time reducing the pool of economically attractive partners for highly educated Black women (Crowder and Tolnay 2000).
In sum, for Black men and women, gendered structural disadvantages associated with low socioeconomic status are compounded by racial discrimination across the life course. Cumulative racial disadvantages are likely substantially larger compared to disadvantages at any single time point (Pager and Shepherd 2008; Reskin 2012). Nonetheless, only a few studies have analyzed enduring and cumulative (dis)advantages in longer-term work and family lives (Aisenbrey and Fasang 2017; Sirniö, Kauppinen and Martikainen 2017). These studies, along with most quantitative life course research on work-family inequality, have paid limited attention to race beyond using it as a control variable, with few exceptions (Florian 2018a, 2018b; Van Winkle and Fasang 2020; Willson 2003). Most studies only focus on women (Florian 2018a, 2018b; Kahn, Garcia-Manglano, and Bianchi 2014; Killewald and Zhuo 2019) and do not undertake a full comparison by gender and race.
Aisenbrey and Fasang (2017) compared Germany and the United States to assess the probability of men and women experiencing different types of work-family lives into midlife. In the United States, interaction effects by gender and race show similar chances for White men and women to combine high-prestige careers with stable partnerships and parenthood. This privilege does not extend to Black women or men, pointing to intersectional inequalities. Aisenbrey and Fasang (2017) highlight prevalent types of interdependence between work and family lives for the total population, in which the White majority drives the resulting averages (the same applies to Killewald and Zhuo 2019, and the fixed effects models in Kahn, Garcia-Manglano, and Bianchi 2014). Specific combinations of work and family lives prevalent only among Black men or White women remain hidden in the pooled analysis. Below, we apply a similar methodological approach to fully map gendered and racialized inequalities in combining work and family lives separately for each group. In addition, we propose a new conceptualization and measurement of the degree of work-family interdependence next to types of interdependence.
Theoretical Background
Work-Family Interdependence across the Life Course
Following Bernardi, Huinink and Settersten Jr (2019, 5), we define the interdependence between life domains as “individuals’ goals, resources, and behaviors in one domain (such as work, family, education, or leisure) [… being] interrelated with goals, resources and behaviors in other domains. This means that domain-specific sub-processes are correlated with each other both at once and over time.” The various causal and selection effects through which work and family lives continually affect one another (see above) combine to form stronger or weaker correlations between work and family life courses. Single-outcome studies often seek to elicit isolated causal or selection effects in the interplay of work and family life (Killewald 2013; Ludwig and Brüderl 2018). We complement this perspective by mapping the degree and type of work-family interdependence, uncovering enduring inequalities that result from a myriad of mechanisms connecting work and family lives over extended periods of time in different ways for Black and White men and women.
In each life domain (work and family), individuals transition through biographical states at a varying pace. Sometimes, events in work lives, such as unemployment or a promotion, are unrelated to family lives. For example, job loss might have no repercussions for family life due to a continuing relationship with a high-earning spouse. Similarly, family events might be unrelated to parallel work careers, for example, when an employee has a third child, takes no parental leave, and work hours and salary remain unchanged. The first example will more often apply to women who are second earners and the second to men who are the main breadwinners. Events in work and family lives can be entirely unrelated, but often they are not, and the (1) degree of work-family interdependence, as well as (2) typical patterns of co-occurrence of work and family lives will vary with ideal worker and parenting norms, initial resources, and structural opportunities for career advancement and family formation that differ by gender and race.
Strong work-family interdependence implies that events in work and family lives mutually condition and constrain one another over longer periods of time, limiting work and family opportunities. Weak work-family interdependence is advantageous in the sense that (dis)advantages in one life domain operate independently of the other, for example, when divorce, parenthood, or job loss do not condition or constrain opportunities in the other life domain. This structural independence between work and family lives can be considered a valuable good in itself.
We build on Bernardi, Huinink and Settersten Jr (2019) to specify three types of work-family interdependence and analyze their prevalence among Black and White men and women: (1) stable, high-prestige careers and family lives of parenthood in stable partnerships as either competing alternatives, or (2) mutually supporting each other, and (3) instability spillovers between interrupted low-prestige careers and family complexity. Limited resources—in terms of time, money, and mental energy—constrain work and family lives. Activities in two life domains can compete for the same resources and preclude each other. Conversely, activities in one life domain can generate resources that open up opportunities in the other life domain, and these life domains can mutually support each other through resource co-production (Bernardi, Huinink and Settersten Jr 2019, 5).
(1) Competing alternatives exist when, for example, individuals have to make choices about investing time and mental energy either at work or when taking care of children (Blair-Loy 2009). When high-stakes careers and parenthood in stable partnerships strongly compete for finite resources, they preclude each other and cannot be jointly sustained over longer periods of time.
In contrast, (2) there is mutually supporting resource co-production across life domains when, for instance, money earned at work attracts potential partners and enables the outsourcing of care activities, which in turn frees up more time and mental energy for work. Mutually supporting work and family lives can stabilize and sustain each other in the long-term.
Finally, (3) instability spillovers between life domains can occur when mutually supporting resources disappear, or when a general lack of resources and high uncertainty undermine long-term planning (Hitlin and Elder 2007). In this case, a series of short-term adjustments can occur, with many transitions in work and family lives. For example, job loss can lead to a loss of mutually supporting financial resources, triggering stress and conflict in partnerships that can lead to separation; this, in turn, further strains economic resources that can no longer be pooled. Such economic hardship may motivate individuals to accept irregular, temporary, and lower-prestige work or to quickly enter new cohabiting partnerships to pool resources, both of which might not last. Instead, such events can trigger more transitions in both work and family lives, cascading into a series of short-term ad-hoc solutions that continually drain resources of time, money, and mental energy and undermine the stability of either occupational careers or family lives.
An Intersectional Perspective on Structural Inequalities in Work and Family Lives
To develop hypotheses on gender and race differences in work-family interdependence, we adopt an intersectional approach that questions gender and race as simple explanatory constructs with separate contributions to inequality (Choo and Ferree 2010; Crenshaw 1991). Overlapping categories of gender and race are associated with particular work-family life course experiences rather than causing them (Browne and Misra 2003, 490). Too often, inequality research diverts attention from the privileges of dominant groups and focuses on the supposed deficits of marginalized groups, who are considered socially problematic (Sprague 2005, 95, 96). An intersectional perspective does not suggest abolishing the most privileged group, typically White men, as a reference category, but suggests that it be complemented with full group-wise comparisons. Full group comparisons avoid the normalization of the White male experience and acknowledge that different groups’ social positions are related to each other.
Some criticize the early intersectionality literature for being preoccupied with identities and disregarding structural disadvantages (McCall 2005). In this study, we focus on structural inequalities in work-family lives for specific birth cohorts of Black and White men and women. Many early studies on intersectionality took either an anticategorical or intracategorical approach that cannot easily be linked to the structural inequalities addressed in the quantitative work-family literature. The anticategorical view rejects categories per se as too simplistic, and because they reinforce the inequalities that they criticize (McCall 2005, 1773). The intracategorical perspective documents the subjective experiences of one intersectional group but lacks a comparison group to assess the specific groups’ experiences. We follow a third intercategorical strategy that compares Black and White men and women vis-à-vis each other (Collins 2015; McCall 2005).
To formulate hypotheses, we discuss the social policies impacting our study cohorts that set the stage for the gendered and racialized labor and marriage markets documented in previous research (see Previous Research and Contributions). The liberal welfare state of the United States promotes women’s employment through gender equity legislation (Orloff 2009; Zippel 2009) but provides little support for childcare or parental leave and only rudimentary basic means-tested benefits (Gornick and Meyers 2003). In liberal regimes, women have better access to top labor market positions but are also less protected from poverty (Aisenbrey and Fasang 2017; Mandel and Shalev 2009). Greater gender equality among some comes with intensified class-based inequalities that translate into racial inequality, given Black men and women’s lower average socioeconomic status. This is relevant for our study, because middle- and upper-class women (who are disproportionately White) tend to benefit from relatively gender-equal access to high-prestige jobs, whereas lower-class men and women (who are disproportionately Black) are overexposed to poverty and insecurity with little state provision.
High earners can afford family time and private childcare, while low and medium earners depend on parental leave and welfare policies. Our study cohorts experienced two major policy reforms: the federal Family and Medical Leave Act (FMLA), the first nationwide option for unpaid parental leave in 1993, and the 1996 welfare reform, the Personal Responsibility and Work Opportunity Reconciliation Act (PRWORA). Our cohorts were between the ages of 31 and 38 in the mid-1990s. With a mean age of 24 at first birth (Mathews and Hamilton 2002), the FMLA came after their prime childbearing years. Moreover, the FMLA only covered around 46% of the private sector and excluded irregular employees with low work hours (Han and Waldfogel 2003). Only firms with 50+ employees were mandated to provide unpaid leave for employees, who had worked at least 1,250 hours in the past year. Our cohorts effectively had no family leave entitlements during their prime childbearing years, outside of occupational or employer provided schemes. These were mainly limited to full-time jobs in high-prestige occupations in larger firms, which White men and women had greater access to (Han and Waldfogel 2003; Western and Pettit 2005; Willson 2003). Non-White, unmarried, and lower educated mothers were less likely to be aware of existing, usually state-level leave schemes or to take any kind of unpaid leave in the 1990s and 2000s (Milkmann and Applebaum 2013; Han and Waldfogel 2003), further limiting their opportunities to combine work and parenthood.
By the time of the 1996 welfare reform, our cohorts had reached occupational maturity and were in their mid- to late thirties, most of them with young children (Fang and Keane 2004; Iceland 2013). Before 1996, federal social policy guaranteed a minimum level of aid to those in poverty. The PRWORA eliminated the entitlement status of welfare and established time limits on receiving aid, along with imposing compulsory work requirements. Welfare eligibility ended after two years, regardless of employment status. A lifetime limit of five years was set on assistance (Iceland 2013, 126). The reform put low earners (disproportionately Black men and women, White women, and single parents) at risk of poverty and employment instability in a critical midcareer phase (Pal and Waldfogel 2016), reinforcing a trend toward increasing employment insecurity in the labor market (Western et al. 2012).
Taken together, the lack of state-funded parental leave and childcare, alongside only basic welfare provision, renders market dynamics extremely important for our cohorts’ opportunities to combine their work and family lives. Those in secure, well-paid jobs with employer-funded benefits and resourceful spouses enjoy greater opportunities, while other groups are overexposed to market forces that limit their options to combine work and family.
Hypotheses
For men, resources generated through stable, high-prestige employment, such as money and prestige, open up opportunities to enter and maintain stable partnerships, and outsource care to female spouses or the market, which frees up time and mental energy for career development. We expect a mutually supporting relationship between stable, high-prestige employment and parenthood in stable partnerships for men (mutual support hypothesis 1). Facing more restricted access to rewarding occupations, Black men will depend even more than White men on mutually supporting resource co-production in work and family lives to sustain successful careers that open up family possibilities and stabilize partnerships (hypothesis 1a).
Because women usually remain primarily responsible for organizing everyday family life and childcare, sustained occupational success and parenthood in stable partnerships will usually lead to more fierce competition for time and mental energy in this group. We hypothesize a competing relationship between parenthood in stable partnerships and stable, high-prestige careers for women (competing alternatives hypothesis 2); these work and family lives will preclude each other. Due to more limited access to high-prestige occupations and disadvantages in marriage markets, this type of interdependence is likely stronger for Black women than White women (hypothesis 2a). For Black women, occupational success will require even stronger resource commitment to career attainment, and they will face a smaller pool of economically desirable partners, which will further reduce opportunities for parenthood in stable partnerships.
Once women attain high-prestige positions, high earnings, benefits, and job security should enable them to realize stable partnered parenthood by outsourcing care. Entering high-prestige occupations later in the life course can prompt a competing relationship to transform into a mutually supporting relationship between parenthood in stable partnerships and occupational success (transformation to mutual support hypothesis 3). White women, who face lower barriers to obtaining and keeping high-prestige jobs, are more likely to experience this transformation than Black women (hypothesis 3a).
Finally, unstable, low-prestige careers will combine with the absence of stable partnerships, either with or without children (instability spillover, hypothesis 4). The absence of stable partnerships includes remaining single, or frequently re-partnering. Instability spillovers between work and family lives will have gendered characteristics: they will be linked to single parenthood for both men and women and single childlessness for men who do not fulfill breadwinner expectations and thus have lower chances in partner markets. Gendered instability spillovers likely operate for all racial groups, but Black men and women are more likely to hold unstable, low-prestige jobs (see above).
Based on hypotheses 1–4 regarding types of interdependence, and previous research on work-family inequality (see Previous Research and Contributions), we summarize expectations on the overall degree of work-family interdependence for each group as follows: Black men and women’s work-family interdependence will be stronger than White men and women’s, because they occupy lower average socioeconomic positions, limiting their family opportunities (work→family). Women’s work-family interdependence will be stronger than men’s, because their family lives have stronger repercussions for their subsequent work careers (family→work). As a result, work-family interdependence will be weakest for White men and strongest for Black women, with Black men and White women in between (degree of interdependence hypothesis 5).
Data and Methods
The National Longitudinal Survey of Youth (NLSY79)1 is a nationally representative sample of 12,686 respondents born between 1957 and 1964, who were first interviewed in 1979, reinterviewed annually until 1994, and biannually since. We analyzed monthly family and work histories from ages 22 to 44 between 1979 and 2008. The NLSY79 is the only NLS group that can be observed into mature adulthood past age 40. Since the oldest cohort (1957) was 22 at the time of the first interview, the observation window opens at age 22. Individuals who had children and married before age 22 started the observation period in these states.
In 2008, 6,261 respondents participated who were either “Black” or “Non-Black and Non-Hispanic”—the latter were classified as White in our analysis. We constructed complete 22-year-long monthly work and family life courses for 5,283 respondents (84%). The family states comprised “Single, no child,” “Single, 1+ children,” “Partnered, no child,” “Partnered, 1 child,” “Partnered, 2 children,” and “Partnered, 3+ children.” Single parents included both resident and nonresident single parents. “Partnered” comprised coresidential married and unmarried unions. Separated or divorced individuals were reclassified as “Single” with or without children (see the online supplementary material for Appendix I).
The work sequences included four nonemployment states and seven employment states. The nonemployment states were “education,” “family care,” “unemployed,” and “gap/out of the labor force.” “Family care” included any type of care of children or other family members with and without official leave entitlements. “Gap/out of the labor force” covered individuals who reported not working and were not in any of the other categories or for whom information was missing for a maximum of one year. Time in employment was categorized into 7 states, each summarizing 10 prestige points of the Treiman (2013) occupational prestige scale (2013): “10–19,” “20–29,” “30–39,” “40–49,” “50–59,” “60–69,” “70–79” (see the online supplementary material for Appendix I). Class was included in our analysis through occupational prestige, which encompasses a package of labor market rewards, including wages, employer benefits, job security, autonomy, and flexibility. We followed processual approaches to social class (Wright and Shin 1988), assuming that class membership is expressed in typical trajectories of attainment—longer-term biographical experiences that form identities and interests and are associated with specific family lives.
To compare occupational prestige within and across intersectional groups, we created an absolute benchmark of low, medium, and high prestige based on the distribution of average prestige across all trajectories in the total sample and the actual jobs. This was necessary because the highest occupational prestige among Black women might correspond to medium prestige among White women. Absolute high prestige included occupations above 48 prestige points (e.g., business and administration associate professionals = 48), medium prestige was between 40 and 47, and occupations below 40 were classified as low prestige (metal workers = 39). This classification of high, medium, and low prestige served as a reference point for interpreting findings across all intersectional groups.
Methods
Mantel coefficients (Piccarreta and Elzinga 2014; Piccarreta 2017) enabled us to measure the degree of interdependence between work and family lives as the correlation of processes in two life domains for the first time (Bernardi, Huinink and Settersten Jr 2019, see above). For each group, all work sequences were compared to each other with optimal matching (substitution costs = 2, indel costs = 1), yielding a pairwise distance matrix that summarizes the similarity of their work lives. The same was repeated for family sequences.2 If Dw and Df denote two distance matrices for the work and family domain, and d1 and d2 are the vectors of their [n(n − 1)/2] extradiagonal elements, the Mantel coefficient is defined as the correlation between all the possible pair-wise dissimilarities, which is between d1 and d2 (Piccarreta 2017, 257).
High Mantel coefficients indicate that individuals with similar family lives also share similar work lives—for example, if early single parenthood only co-occurs with precarious unstable employment (instability spillover, strong work-family interdependence). Low Mantel coefficients suggest that individuals with similar family lives experience many different work careers without any systematic association and vice versa. Events in the family and work domain do not constrain or condition each other (weak work-family interdependence). We calculated Mantel coefficients separately for the four intersectional groups.
Multichannel sequence analysis (Gauthier et al. 2010; Pollock 2007) elicits prevalent types of work-family interdependence. Mantel coefficients are conservative measures that only pick up global associations across the entire distance matrices (Piccarreta and Elzinga 2014). Types of interdependence identified with multichannel sequence and cluster analysis capture both global and local associations between distance matrices. Multichannel sequence analysis classifies life courses as similar when both work and family lives are similar. The resulting pairwise distance matrix using optimal matching (substitution costs = 2, indel costs = 1) enters a partitioning around medoids (PAM) cluster analysis (Studer 2013). Cluster cut-off criteria provide information on whether a meaningful group structure exists and assess the most appropriate number of clusters (see the online supplementary material for Appendix II figure A1). Some clusters are more homogeneous than others, attesting to the heterogeneity of work-family lives. We explored the variation around the main patterns in each cluster using the silhouette width as an indicator for cluster coherence (Studer 2013) (see the online supplementary material for Appendix II figures A2–A7). Because all analyses are calculated separately for the four intersectional groups, the final analyses do not apply the NLSY weights, which correct for the oversampling of Black men and women. Analyses with weights provided very similar results (available from authors).3
Results
Degree of Interdependence
Figure 1 corroborates weak work-family interdependence for White men (hypothesis 5): the Mantel coefficient of 0.01 is not significantly different from zero. Many different combinations occur empirically for White men, indicating that they have the privilege of combining different types of work and family life. This does not imply that all White men “get what they want.” Possibility is not choice. But disadvantages in the work and family domains operate largely independently from one another. Events in one life domain do not constrain or condition events in the other. For White women and Black men, the Mantel correlations of 0.05 are significantly higher compared to White men, with nonoverlapping confidence intervals (figure 1). As expected, work-family interdependence is strongest for Black women, at 0.09.4 The 95% confidence interval for Black women slightly overlaps with those for White women and Black men, but the point estimates do not overlap with the confidence intervals, indicating statistically sound differences (Austin and Hux 2002). The 90% confidence intervals do not overlap (available from authors).
Mantel coefficients on the degree of work-family interdependence.
Types of Interdependence
Several cluster cut-off criteria support six clusters for Black men and five clusters for Black and White women as the best groupings (see the online supplementary material for Appendix figure A1). For White men, cluster cut-off criteria suggest no discernable grouping. White men’s work-family interdependence is so weak that combinations do not cluster into prevalent types. Figures 2–5 show state distribution plots that aggregate state frequencies at each age. Figure 2 displays the total for White men, which lack a clear grouping. Family lives are presented on the left and parallel work lives on the right. The size of the clusters in figures 3–5 reflects their proportion in the respective intersectional group. The clusters are sorted in descending order with the highest average prestige cluster at the top and the lowest at the bottom. Tables 1 and 2 present descriptive information on education, complexity of work trajectories as an indicator for career stability (Elzinga 2010), average prestige difference between the beginning and end of the trajectory, and the percentage that are upwardly or downwardly mobile. Table 3 synthesizes the main findings from figures 3–5, showing typical family lives that cluster with work trajectories of high, medium, stable, and unstable, low prestige for Black men and women, and White women.
State distribution plot of family (left) and work (right) life courses for White men (view in color, total population due to lack of cluster structure).
State distribution plots of six clusters of family (left) and work (right) life courses for Black men (view in color, cluster average prestige score in parentheses).
State distribution plots of five clusters of family (left) and work (right) life courses for Black women (view in color, cluster average prestige score in parentheses).
State distribution plots of five clusters of family (left) and work (right) life courses for White women (view in color, cluster average prestige score in parentheses).
Descriptive Information for Black and White Men
| Black men | White men | |||||||
|---|---|---|---|---|---|---|---|---|
| Least → most advantaged | ||||||||
| Clustersa | 1) | 2) | 3) | 4) | 5) | 6) | Total | Total |
| N | 180 | 158 | 175 | 107 | 98 | 109 | 827 | 1,757 |
| % | 22 | 19 | 21 | 13 | 12 | 13 | 100 | 100 |
| Average Treiman | 28.9 | 33.7 | 34.0 | 35.7 | 38.1 | 49.2 | 35.6 | 42.2 |
| Prestige change | 0.2 | 3.7 | 2.7 | 3.0 | 4.6 | 6.1 | 3.1 | 5.8 |
| % mobile: up | 29.4 | 41.1 | 45.7 | 40.2 | 53.1 | 45.0 | 41.4 | 47.2 |
| same | 32.8 | 31.0 | 22.3 | 35.5 | 25.5 | 33.0 | 29.1 | 29.9 |
| down | 36.7 | 25.3 | 30.3 | 24.3 | 20.4 | 22.0 | 28.6 | 22.4 |
| NA | 1.1 | 2.5 | 1.7 | 0.0 | 1.0 | 0.0 | 1.2 | 0.5 |
| Complexity work | 20.2 | 19.1 | 19.7 | 15.9 | 14.5 | 14.6 | 17.9 | 15.2 |
| % No HS | 34.3 | 29.7 | 25.7 | 29.0 | 18.4 | 10.1 | 25.9 | 17.7 |
| % Just HS | 49.4 | 32.3 | 49.1 | 41.1 | 39.8 | 19.3 | 40.3 | 35.5 |
| Father Edu years | 9.3 | 10.2 | 10.5 | 9.5 | 11.0 | 11.3 | 10.3 | 11.8 |
| Mother Edu years | 10.3 | 10.9 | 10.9 | 10.6 | 10.9 | 11.9 | 10.9 | 11.4 |
| Child start | 0.6 | 0.01 | 0.5 | 0.7 | 0.3 | 0.2 | 0.4 | 0.2 |
| Child end | 2.2 | 0.3 | 2.2 | 3.5 | 1.9 | 1.5 | 1.9 | 1.7 |
| Black men | White men | |||||||
|---|---|---|---|---|---|---|---|---|
| Least → most advantaged | ||||||||
| Clustersa | 1) | 2) | 3) | 4) | 5) | 6) | Total | Total |
| N | 180 | 158 | 175 | 107 | 98 | 109 | 827 | 1,757 |
| % | 22 | 19 | 21 | 13 | 12 | 13 | 100 | 100 |
| Average Treiman | 28.9 | 33.7 | 34.0 | 35.7 | 38.1 | 49.2 | 35.6 | 42.2 |
| Prestige change | 0.2 | 3.7 | 2.7 | 3.0 | 4.6 | 6.1 | 3.1 | 5.8 |
| % mobile: up | 29.4 | 41.1 | 45.7 | 40.2 | 53.1 | 45.0 | 41.4 | 47.2 |
| same | 32.8 | 31.0 | 22.3 | 35.5 | 25.5 | 33.0 | 29.1 | 29.9 |
| down | 36.7 | 25.3 | 30.3 | 24.3 | 20.4 | 22.0 | 28.6 | 22.4 |
| NA | 1.1 | 2.5 | 1.7 | 0.0 | 1.0 | 0.0 | 1.2 | 0.5 |
| Complexity work | 20.2 | 19.1 | 19.7 | 15.9 | 14.5 | 14.6 | 17.9 | 15.2 |
| % No HS | 34.3 | 29.7 | 25.7 | 29.0 | 18.4 | 10.1 | 25.9 | 17.7 |
| % Just HS | 49.4 | 32.3 | 49.1 | 41.1 | 39.8 | 19.3 | 40.3 | 35.5 |
| Father Edu years | 9.3 | 10.2 | 10.5 | 9.5 | 11.0 | 11.3 | 10.3 | 11.8 |
| Mother Edu years | 10.3 | 10.9 | 10.9 | 10.6 | 10.9 | 11.9 | 10.9 | 11.4 |
| Child start | 0.6 | 0.01 | 0.5 | 0.7 | 0.3 | 0.2 | 0.4 | 0.2 |
| Child end | 2.2 | 0.3 | 2.2 | 3.5 | 1.9 | 1.5 | 1.9 | 1.7 |
Note: Total only for White men due to lack of cluster structure.
aClusters for Black men: 1) Early single parent, unstable low prestige; 2) Single childless, unstable low prestige; 3) Late single parent, unstable low prestige; 4) Repartnered, many children, stable low prestige; 5) Partnered, two children, stable low prestige; 6) Partnered, late one child, high prestige.
Descriptive Information for Black and White Men
| Black men | White men | |||||||
|---|---|---|---|---|---|---|---|---|
| Least → most advantaged | ||||||||
| Clustersa | 1) | 2) | 3) | 4) | 5) | 6) | Total | Total |
| N | 180 | 158 | 175 | 107 | 98 | 109 | 827 | 1,757 |
| % | 22 | 19 | 21 | 13 | 12 | 13 | 100 | 100 |
| Average Treiman | 28.9 | 33.7 | 34.0 | 35.7 | 38.1 | 49.2 | 35.6 | 42.2 |
| Prestige change | 0.2 | 3.7 | 2.7 | 3.0 | 4.6 | 6.1 | 3.1 | 5.8 |
| % mobile: up | 29.4 | 41.1 | 45.7 | 40.2 | 53.1 | 45.0 | 41.4 | 47.2 |
| same | 32.8 | 31.0 | 22.3 | 35.5 | 25.5 | 33.0 | 29.1 | 29.9 |
| down | 36.7 | 25.3 | 30.3 | 24.3 | 20.4 | 22.0 | 28.6 | 22.4 |
| NA | 1.1 | 2.5 | 1.7 | 0.0 | 1.0 | 0.0 | 1.2 | 0.5 |
| Complexity work | 20.2 | 19.1 | 19.7 | 15.9 | 14.5 | 14.6 | 17.9 | 15.2 |
| % No HS | 34.3 | 29.7 | 25.7 | 29.0 | 18.4 | 10.1 | 25.9 | 17.7 |
| % Just HS | 49.4 | 32.3 | 49.1 | 41.1 | 39.8 | 19.3 | 40.3 | 35.5 |
| Father Edu years | 9.3 | 10.2 | 10.5 | 9.5 | 11.0 | 11.3 | 10.3 | 11.8 |
| Mother Edu years | 10.3 | 10.9 | 10.9 | 10.6 | 10.9 | 11.9 | 10.9 | 11.4 |
| Child start | 0.6 | 0.01 | 0.5 | 0.7 | 0.3 | 0.2 | 0.4 | 0.2 |
| Child end | 2.2 | 0.3 | 2.2 | 3.5 | 1.9 | 1.5 | 1.9 | 1.7 |
| Black men | White men | |||||||
|---|---|---|---|---|---|---|---|---|
| Least → most advantaged | ||||||||
| Clustersa | 1) | 2) | 3) | 4) | 5) | 6) | Total | Total |
| N | 180 | 158 | 175 | 107 | 98 | 109 | 827 | 1,757 |
| % | 22 | 19 | 21 | 13 | 12 | 13 | 100 | 100 |
| Average Treiman | 28.9 | 33.7 | 34.0 | 35.7 | 38.1 | 49.2 | 35.6 | 42.2 |
| Prestige change | 0.2 | 3.7 | 2.7 | 3.0 | 4.6 | 6.1 | 3.1 | 5.8 |
| % mobile: up | 29.4 | 41.1 | 45.7 | 40.2 | 53.1 | 45.0 | 41.4 | 47.2 |
| same | 32.8 | 31.0 | 22.3 | 35.5 | 25.5 | 33.0 | 29.1 | 29.9 |
| down | 36.7 | 25.3 | 30.3 | 24.3 | 20.4 | 22.0 | 28.6 | 22.4 |
| NA | 1.1 | 2.5 | 1.7 | 0.0 | 1.0 | 0.0 | 1.2 | 0.5 |
| Complexity work | 20.2 | 19.1 | 19.7 | 15.9 | 14.5 | 14.6 | 17.9 | 15.2 |
| % No HS | 34.3 | 29.7 | 25.7 | 29.0 | 18.4 | 10.1 | 25.9 | 17.7 |
| % Just HS | 49.4 | 32.3 | 49.1 | 41.1 | 39.8 | 19.3 | 40.3 | 35.5 |
| Father Edu years | 9.3 | 10.2 | 10.5 | 9.5 | 11.0 | 11.3 | 10.3 | 11.8 |
| Mother Edu years | 10.3 | 10.9 | 10.9 | 10.6 | 10.9 | 11.9 | 10.9 | 11.4 |
| Child start | 0.6 | 0.01 | 0.5 | 0.7 | 0.3 | 0.2 | 0.4 | 0.2 |
| Child end | 2.2 | 0.3 | 2.2 | 3.5 | 1.9 | 1.5 | 1.9 | 1.7 |
Note: Total only for White men due to lack of cluster structure.
aClusters for Black men: 1) Early single parent, unstable low prestige; 2) Single childless, unstable low prestige; 3) Late single parent, unstable low prestige; 4) Repartnered, many children, stable low prestige; 5) Partnered, two children, stable low prestige; 6) Partnered, late one child, high prestige.
Black men
High-prestige careers (mean = 49) strongly go along with stable partnerships and only having one or two children for a small group of Black men (13%, Cluster 6 in figure 3), corroborating a mutually supporting relationship between successful careers and parenthood in stable partnerships (hypothesis 1a). Medium-prestige careers do not occur in large numbers, highlighting the lack of middle-class careers for Black men (table 3). Instead, the remaining 87% of Black men divide into those experiencing stable and unstable, low-prestige careers (complexity, table 1).
Clusters 5 and 4 combine stable low-prestige employment (mean = 38 and 36, complexity = 14.5 and 15.9, table 1) with either a stable partnership and early fatherhood of two children (cluster 5) or repartnering after early high fertility (cluster 4). Repartnering (cluster 4) is visible in single fatherhood (green), peaking at almost 50% at age 23 but dropping sharply to 14% by age 34. By their mid-thirties, 81% of this group have entered a new cohabiting or married relationship with three or more children (dark blue, percentages from figure 3).
The remaining three work-family types in clusters 3, 2, and 1 combine unstable, low-prestige careers (average prestige of 34, 34, and 29, complexity = 19.7, 19.2, and 20.1) with the absence of stable partnerships into mature adulthood (instability spillover hypothesis 4, table 3). Instead, they experience late single fatherhood (cluster 3), single childlessness (cluster 2), and early single fatherhood (cluster 1). Early single fatherhood (cluster 1) goes along with the lowest prestige, highest employment instability, and the greatest downward mobility (36.7%) (table 1).
For Black men, careers of high prestige, but also stable low prestige, only occur in sizeable numbers in combination with stable first or later partnerships for most of the time between ages 22 and 44 (Clusters 6, 5, and 4). Stable careers, even if they are low prestige, offer family possibilities that in turn sustain career stability, suggesting a “career stability premium” associated with long-term coresidential unions for Black men (mutual support hypothesis 1a). Associations between fatherhood and occupational careers are more complex. High-prestige careers strongly coincide with relatively late partnerships and low fertility, supporting a career advantage of delayed, moderate fertility, though single childlessness is not associated with occupational success. Instead, single childlessness goes along with unstable and low-prestige careers, in line with gendered instability spillovers between work and family lives (hypothesis 4).
Black women
Black women are the only group for whom high-prestige careers do not occur in sizeable numbers, irrespective of their family lives (table 3). For Black women, high-prestige careers remain an exception. In contrast to Black men, the most successful and stable medium-prestige careers among Black women are associated with the absence of stable partnerships: cluster 5 combines late single motherhood, and cluster 4 single childlessness, both with an average prestige of 46 (figure 4 and table 2). Cluster 5 (late single motherhood with stable medium-prestige careers) is the most upwardly mobile one (5.7 prestige point gain, 44.1% upward mobility, table 2). Unlike Black men, for sizeable groups of Black women (22% and 18% in clusters 5 and 4,) we find a competing, not mutually supporting relationship between occupational success and stable coresidential partnerships (hypothesis 2a).
A small group of Black women (14%) combine stable partnerships and motherhood of two children with medium-prestige careers (Cluster 3). The remaining two groups are trapped in precarious, low-prestige employment well into midlife. Cluster 2 experiences low occupational prestige at average employment stability (Treiman average = 39, complexity = 19.6) with repartnering and early high fertility. Similar to Black men, the women in Cluster 1 combine the lowest prestige and most unstable employment (Treiman average = 34, complexity = 21.8) with early and enduring single parenthood. Compared to Black men, this group experiences even more downward mobility (tables 1 and 2) and accounts for a staggering 34% of Black women (22% for Black men, table 3).5
For Black women, single motherhood is associated with either unstable, low-prestige careers (Cluster 1) or stable, medium-prestige careers and upward mobility (Cluster 5). Therefore, instability spillovers between single parenthood and precarious work (hypothesis 4) only apply to Cluster 1, but not to Cluster 5, suggesting that they only operate under certain conditions. Single mothers in Cluster 1 have many children (2.4), enter motherhood early (almost all before age 22), and are mostly single at birth. Single mothers in Cluster 5 have fewer children (1.7) and enter (single) motherhood later and more often through separation (see figure 4). Consequently, the timing and life course context (through birth or separation), rather than single motherhood per se, is decisive in Black women’s parallel career development (Zagel and Hübgen 2018).
White women
Unlike Black women, White women enjoy high-prestige careers in sizeable numbers (table 3, figure 5). The most successful White women surpass the highest prestige group for Black men by four points (54 compared to 49). High occupational prestige careers coincide with either spending most of the time between ages 22 and 44 single and childless (Cluster 5) or having only one child relatively late in a stable partnership (Cluster 4). Together, the two high-prestige groups account for 38% of White women. Cluster 5, the occupationally most successful work-family experience, indicates a competing relationship between occupational success and parenthood in stable partnerships (hypothesis 2), similar to the relatively most successful medium-prestige careers among Black women (hypothesis 2a). Cluster 4 shows a transformation from a competing to a mutually supporting relationship between partnered parenthood for White women once they have entered high-prestige occupations after age 30 (hypothesis 3a) (see Doren 2019).
Descriptive Information for Black and White Women
| Black women | White women | |||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Least → most advantaged | Least → most advantaged | |||||||||||
| Clustersa | 1) | 2) | 3) | 4) | 5) | Total | 1) | 2) | 3) | 4) | 5) | Total |
| N | 299 | 115 | 118 | 153 | 188 | 873 | 236 | 372 | 535 | 473 | 210 | 1,826 |
| % | 34 | 13 | 14 | 18 | 22 | 100 | 13 | 20 | 29 | 26 | 12 | 100 |
| Average prestige | 34.2 | 38.8 | 45.2 | 45.7 | 46.3 | 41.1 | 38.0 | 42.3 | 46.7 | 48.6 | 53.7 | 46.0 |
| Prestige change | −0.3 | 4.5 | 4.4 | 3.1 | 5.7 | 2.9 | 3.1 | 3.8 | 4.1 | 4.7 | 6.1 | 4.3 |
| % mobile: up | 28.4 | 39.1 | 47.5 | 39.2 | 44.1 | 37.7 | 37.7 | 43.3 | 42.4 | 45.0 | 41.9 | 42.4 |
| same | 27.4 | 26.1 | 22.0 | 28.1 | 28.7 | 26.9 | 29.7 | 29.8 | 29.3 | 27.9 | 36.2 | 29.9 |
| down | 36.1 | 33.9 | 30.5 | 32.7 | 27.1 | 32.5 | 29.2 | 25.5 | 28.0 | 27.1 | 21.4 | 26.7 |
| NA | 8.0 | 0.9 | 0.0 | 0.0 | 0.0 | 2.9 | 3.4 | 2.4 | 0.2 | 0.0 | 0.5 | 1.0 |
| Complexity work | 21.8 | 19.6 | 17.4 | 17.5 | 18.4 | 19.4 | 21.5 | 19.8 | 18.3 | 15.8 | 16.4 | 18.2 |
| % No HS | 38.5 | 17.4 | 1.7 | 7.2 | 7.4 | 18.7 | 36.0 | 18.8 | 14.2 | 5.7 | 4.3 | 14.6 |
| % Just HS | 35.8 | 34.8 | 38.1 | 22.2 | 28.7 | 32.1 | 41.9 | 32.2 | 35.7 | 38.3 | 18.1 | 34.3 |
| Father Edu years | 9.4 | 10.0 | 10.7 | 11.3 | 10.4 | 10.3 | 10.2 | 11.7 | 11.5 | 12.2 | 12.7 | 11.7 |
| Mother Edu years | 9.9 | 10.9 | 11.1 | 11.2 | 10.9 | 10.7 | 9.9 | 11.1 | 11.4 | 11.7 | 12.2 | 11.3 |
| Child start | 1.2 | 0.9 | 0.4 | 0.0 | 0.8 | 0.8 | 0.9 | 0.7 | 0.4 | 0.1 | 0.01 | 0.4 |
| Child end | 2.4 | 2.9 | 1.9 | 0.2 | 1.7 | 1.9 | 2.3 | 3.2 | 2.0 | 0.9 | 0.7 | 1.8 |
| Black women | White women | |||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Least → most advantaged | Least → most advantaged | |||||||||||
| Clustersa | 1) | 2) | 3) | 4) | 5) | Total | 1) | 2) | 3) | 4) | 5) | Total |
| N | 299 | 115 | 118 | 153 | 188 | 873 | 236 | 372 | 535 | 473 | 210 | 1,826 |
| % | 34 | 13 | 14 | 18 | 22 | 100 | 13 | 20 | 29 | 26 | 12 | 100 |
| Average prestige | 34.2 | 38.8 | 45.2 | 45.7 | 46.3 | 41.1 | 38.0 | 42.3 | 46.7 | 48.6 | 53.7 | 46.0 |
| Prestige change | −0.3 | 4.5 | 4.4 | 3.1 | 5.7 | 2.9 | 3.1 | 3.8 | 4.1 | 4.7 | 6.1 | 4.3 |
| % mobile: up | 28.4 | 39.1 | 47.5 | 39.2 | 44.1 | 37.7 | 37.7 | 43.3 | 42.4 | 45.0 | 41.9 | 42.4 |
| same | 27.4 | 26.1 | 22.0 | 28.1 | 28.7 | 26.9 | 29.7 | 29.8 | 29.3 | 27.9 | 36.2 | 29.9 |
| down | 36.1 | 33.9 | 30.5 | 32.7 | 27.1 | 32.5 | 29.2 | 25.5 | 28.0 | 27.1 | 21.4 | 26.7 |
| NA | 8.0 | 0.9 | 0.0 | 0.0 | 0.0 | 2.9 | 3.4 | 2.4 | 0.2 | 0.0 | 0.5 | 1.0 |
| Complexity work | 21.8 | 19.6 | 17.4 | 17.5 | 18.4 | 19.4 | 21.5 | 19.8 | 18.3 | 15.8 | 16.4 | 18.2 |
| % No HS | 38.5 | 17.4 | 1.7 | 7.2 | 7.4 | 18.7 | 36.0 | 18.8 | 14.2 | 5.7 | 4.3 | 14.6 |
| % Just HS | 35.8 | 34.8 | 38.1 | 22.2 | 28.7 | 32.1 | 41.9 | 32.2 | 35.7 | 38.3 | 18.1 | 34.3 |
| Father Edu years | 9.4 | 10.0 | 10.7 | 11.3 | 10.4 | 10.3 | 10.2 | 11.7 | 11.5 | 12.2 | 12.7 | 11.7 |
| Mother Edu years | 9.9 | 10.9 | 11.1 | 11.2 | 10.9 | 10.7 | 9.9 | 11.1 | 11.4 | 11.7 | 12.2 | 11.3 |
| Child start | 1.2 | 0.9 | 0.4 | 0.0 | 0.8 | 0.8 | 0.9 | 0.7 | 0.4 | 0.1 | 0.01 | 0.4 |
| Child end | 2.4 | 2.9 | 1.9 | 0.2 | 1.7 | 1.9 | 2.3 | 3.2 | 2.0 | 0.9 | 0.7 | 1.8 |
aClusters: Black women: 1) Early single parent, unstable low prestige; 2) Repartnered, many children, stable low prestige; 3) Partnered, two children, medium prestige; 4) Single childless, medium prestige; 5) Late single parent, medium prestige. White women: 1) Early single parent, unstable low prestige; 2) Partnered, many children, medium prestige; 3) Partnered, early two children, medium prestige; 4) Partnered, late one child, high prestige; 5) Single childless, high prestige.
Descriptive Information for Black and White Women
| Black women | White women | |||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Least → most advantaged | Least → most advantaged | |||||||||||
| Clustersa | 1) | 2) | 3) | 4) | 5) | Total | 1) | 2) | 3) | 4) | 5) | Total |
| N | 299 | 115 | 118 | 153 | 188 | 873 | 236 | 372 | 535 | 473 | 210 | 1,826 |
| % | 34 | 13 | 14 | 18 | 22 | 100 | 13 | 20 | 29 | 26 | 12 | 100 |
| Average prestige | 34.2 | 38.8 | 45.2 | 45.7 | 46.3 | 41.1 | 38.0 | 42.3 | 46.7 | 48.6 | 53.7 | 46.0 |
| Prestige change | −0.3 | 4.5 | 4.4 | 3.1 | 5.7 | 2.9 | 3.1 | 3.8 | 4.1 | 4.7 | 6.1 | 4.3 |
| % mobile: up | 28.4 | 39.1 | 47.5 | 39.2 | 44.1 | 37.7 | 37.7 | 43.3 | 42.4 | 45.0 | 41.9 | 42.4 |
| same | 27.4 | 26.1 | 22.0 | 28.1 | 28.7 | 26.9 | 29.7 | 29.8 | 29.3 | 27.9 | 36.2 | 29.9 |
| down | 36.1 | 33.9 | 30.5 | 32.7 | 27.1 | 32.5 | 29.2 | 25.5 | 28.0 | 27.1 | 21.4 | 26.7 |
| NA | 8.0 | 0.9 | 0.0 | 0.0 | 0.0 | 2.9 | 3.4 | 2.4 | 0.2 | 0.0 | 0.5 | 1.0 |
| Complexity work | 21.8 | 19.6 | 17.4 | 17.5 | 18.4 | 19.4 | 21.5 | 19.8 | 18.3 | 15.8 | 16.4 | 18.2 |
| % No HS | 38.5 | 17.4 | 1.7 | 7.2 | 7.4 | 18.7 | 36.0 | 18.8 | 14.2 | 5.7 | 4.3 | 14.6 |
| % Just HS | 35.8 | 34.8 | 38.1 | 22.2 | 28.7 | 32.1 | 41.9 | 32.2 | 35.7 | 38.3 | 18.1 | 34.3 |
| Father Edu years | 9.4 | 10.0 | 10.7 | 11.3 | 10.4 | 10.3 | 10.2 | 11.7 | 11.5 | 12.2 | 12.7 | 11.7 |
| Mother Edu years | 9.9 | 10.9 | 11.1 | 11.2 | 10.9 | 10.7 | 9.9 | 11.1 | 11.4 | 11.7 | 12.2 | 11.3 |
| Child start | 1.2 | 0.9 | 0.4 | 0.0 | 0.8 | 0.8 | 0.9 | 0.7 | 0.4 | 0.1 | 0.01 | 0.4 |
| Child end | 2.4 | 2.9 | 1.9 | 0.2 | 1.7 | 1.9 | 2.3 | 3.2 | 2.0 | 0.9 | 0.7 | 1.8 |
| Black women | White women | |||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Least → most advantaged | Least → most advantaged | |||||||||||
| Clustersa | 1) | 2) | 3) | 4) | 5) | Total | 1) | 2) | 3) | 4) | 5) | Total |
| N | 299 | 115 | 118 | 153 | 188 | 873 | 236 | 372 | 535 | 473 | 210 | 1,826 |
| % | 34 | 13 | 14 | 18 | 22 | 100 | 13 | 20 | 29 | 26 | 12 | 100 |
| Average prestige | 34.2 | 38.8 | 45.2 | 45.7 | 46.3 | 41.1 | 38.0 | 42.3 | 46.7 | 48.6 | 53.7 | 46.0 |
| Prestige change | −0.3 | 4.5 | 4.4 | 3.1 | 5.7 | 2.9 | 3.1 | 3.8 | 4.1 | 4.7 | 6.1 | 4.3 |
| % mobile: up | 28.4 | 39.1 | 47.5 | 39.2 | 44.1 | 37.7 | 37.7 | 43.3 | 42.4 | 45.0 | 41.9 | 42.4 |
| same | 27.4 | 26.1 | 22.0 | 28.1 | 28.7 | 26.9 | 29.7 | 29.8 | 29.3 | 27.9 | 36.2 | 29.9 |
| down | 36.1 | 33.9 | 30.5 | 32.7 | 27.1 | 32.5 | 29.2 | 25.5 | 28.0 | 27.1 | 21.4 | 26.7 |
| NA | 8.0 | 0.9 | 0.0 | 0.0 | 0.0 | 2.9 | 3.4 | 2.4 | 0.2 | 0.0 | 0.5 | 1.0 |
| Complexity work | 21.8 | 19.6 | 17.4 | 17.5 | 18.4 | 19.4 | 21.5 | 19.8 | 18.3 | 15.8 | 16.4 | 18.2 |
| % No HS | 38.5 | 17.4 | 1.7 | 7.2 | 7.4 | 18.7 | 36.0 | 18.8 | 14.2 | 5.7 | 4.3 | 14.6 |
| % Just HS | 35.8 | 34.8 | 38.1 | 22.2 | 28.7 | 32.1 | 41.9 | 32.2 | 35.7 | 38.3 | 18.1 | 34.3 |
| Father Edu years | 9.4 | 10.0 | 10.7 | 11.3 | 10.4 | 10.3 | 10.2 | 11.7 | 11.5 | 12.2 | 12.7 | 11.7 |
| Mother Edu years | 9.9 | 10.9 | 11.1 | 11.2 | 10.9 | 10.7 | 9.9 | 11.1 | 11.4 | 11.7 | 12.2 | 11.3 |
| Child start | 1.2 | 0.9 | 0.4 | 0.0 | 0.8 | 0.8 | 0.9 | 0.7 | 0.4 | 0.1 | 0.01 | 0.4 |
| Child end | 2.4 | 2.9 | 1.9 | 0.2 | 1.7 | 1.9 | 2.3 | 3.2 | 2.0 | 0.9 | 0.7 | 1.8 |
aClusters: Black women: 1) Early single parent, unstable low prestige; 2) Repartnered, many children, stable low prestige; 3) Partnered, two children, medium prestige; 4) Single childless, medium prestige; 5) Late single parent, medium prestige. White women: 1) Early single parent, unstable low prestige; 2) Partnered, many children, medium prestige; 3) Partnered, early two children, medium prestige; 4) Partnered, late one child, high prestige; 5) Single childless, high prestige.
Overview of Family Life Course Types by Average Occupational Prestige for Black Men and Women and White Women
| Prestige | Black men | % | Black women | % | White women | % |
|---|---|---|---|---|---|---|
| High | 5) Single childlessness | 12 | ||||
| 6) Partnered, late one child | 13 | — | 4) Partnered, late one child | 26 | ||
| Medium | 5) Late single parenthood | 22 | 3) Partnered, early two children | 29 | ||
| 4) Single childlessness | 18 | 2) Partnered, early high fertility | 20 | |||
| — | 3) Partnered, two children | 14 | ||||
| Low stable | 5) Partnered, two children | 12 | ||||
| 4) (Re-)partnered, early high fertility | 13 | 2) (Re-)partnered, early high fertility | 13 | — | ||
| Low unstable | 3) Late single parenthood | 21 | ||||
| 2) Single childlessness | 19 | |||||
| 1) Early single parenthood | 22 | 1) Early single parenthood | 34 | 1) Early single parenthood | 13 |
| Prestige | Black men | % | Black women | % | White women | % |
|---|---|---|---|---|---|---|
| High | 5) Single childlessness | 12 | ||||
| 6) Partnered, late one child | 13 | — | 4) Partnered, late one child | 26 | ||
| Medium | 5) Late single parenthood | 22 | 3) Partnered, early two children | 29 | ||
| 4) Single childlessness | 18 | 2) Partnered, early high fertility | 20 | |||
| — | 3) Partnered, two children | 14 | ||||
| Low stable | 5) Partnered, two children | 12 | ||||
| 4) (Re-)partnered, early high fertility | 13 | 2) (Re-)partnered, early high fertility | 13 | — | ||
| Low unstable | 3) Late single parenthood | 21 | ||||
| 2) Single childlessness | 19 | |||||
| 1) Early single parenthood | 22 | 1) Early single parenthood | 34 | 1) Early single parenthood | 13 |
Overview of Family Life Course Types by Average Occupational Prestige for Black Men and Women and White Women
| Prestige | Black men | % | Black women | % | White women | % |
|---|---|---|---|---|---|---|
| High | 5) Single childlessness | 12 | ||||
| 6) Partnered, late one child | 13 | — | 4) Partnered, late one child | 26 | ||
| Medium | 5) Late single parenthood | 22 | 3) Partnered, early two children | 29 | ||
| 4) Single childlessness | 18 | 2) Partnered, early high fertility | 20 | |||
| — | 3) Partnered, two children | 14 | ||||
| Low stable | 5) Partnered, two children | 12 | ||||
| 4) (Re-)partnered, early high fertility | 13 | 2) (Re-)partnered, early high fertility | 13 | — | ||
| Low unstable | 3) Late single parenthood | 21 | ||||
| 2) Single childlessness | 19 | |||||
| 1) Early single parenthood | 22 | 1) Early single parenthood | 34 | 1) Early single parenthood | 13 |
| Prestige | Black men | % | Black women | % | White women | % |
|---|---|---|---|---|---|---|
| High | 5) Single childlessness | 12 | ||||
| 6) Partnered, late one child | 13 | — | 4) Partnered, late one child | 26 | ||
| Medium | 5) Late single parenthood | 22 | 3) Partnered, early two children | 29 | ||
| 4) Single childlessness | 18 | 2) Partnered, early high fertility | 20 | |||
| — | 3) Partnered, two children | 14 | ||||
| Low stable | 5) Partnered, two children | 12 | ||||
| 4) (Re-)partnered, early high fertility | 13 | 2) (Re-)partnered, early high fertility | 13 | — | ||
| Low unstable | 3) Late single parenthood | 21 | ||||
| 2) Single childlessness | 19 | |||||
| 1) Early single parenthood | 22 | 1) Early single parenthood | 34 | 1) Early single parenthood | 13 |
The majority of White women (49%) combine relatively stable medium-prestige careers with stable partnerships and either two or more children (Clusters 3 and 2). Similar to Black women, the most economically disadvantaged White women combine early single motherhood with precarious employment and welfare dependence into midlife (gendered instability spillover, hypothesis 4). However, this only applies to 13% of White women (34% of Black women, 22% of Black men, table 3). White single mothers’ average occupational prestige is four points higher than that of their Black peers and they are less downwardly mobile (table 2). Unlike for Black women, single motherhood for White women only coincides in sizeable numbers with unstable, low-prestige employment, and not with stable middle-class careers.
Discussion
This study takes an intersectional life course perspective to elicit degrees and types of work-family interdependence over the life course by gender and race. White men have abundant possibilities for combining their work and family lives. On average, they enjoy the highest socioeconomic status and career rewards that open family possibilities, and their family lives have few repercussions for subsequent careers, lowering work-family interdependence over longer periods of the life course. Black men are concentrated in low socioeconomic positions and unstable, low-prestige careers associated with limited family opportunities. In addition, previous research shows that family events, such as separation, have greater negative career consequences for Black men than for White men (Smock 1993). This fuels stronger work-family interdependence for Black men, which is evident in our analysis. For Black men, fatherhood in stable partnerships and high-prestige careers mutually support one another into midlife. Even in low-prestige jobs, Black men enjoy a “career stability premium” associated with partnered parenthood. This is consistent with an initial selection argument (Ludwig and Brüderl 2018): men who are more successful in finding a partner are also more successful in establishing stable careers; subsequently, partnered parenthood and occupational success positively reinforce one another through effective resource coproduction (Bernardi, Huinink and Settersten Jr 2019). Work-family patterns for Black men polarize into high-prestige (13%), stable low-prestige (25%), and unstable, low-prestige employment (62%). There is no common career path for Black men in secure middle-class jobs, irrespective of their family lives. Unstable, low-prestige careers, experienced by a staggering 62% of Black men in our cohorts, strongly coincide with an enduring absence of stable partnerships: these men are either childless or single fathers (gendered instability spillover). Single fatherhood among Black men is likely more nonresidential than among the majority of White and Black single mothers. Future research should provide a more detailed account of shared parenting arrangements.
Black women experience the strongest interdependence of work and family lives. Stable, high-prestige employment, enjoyed by few Black men (13%) and more White women (38%), is not viable for Black women in sizeable numbers, irrespective of their family lives. Gendered work-family interdependence, specific to Black men and women, underlines the added value of an intersectional perspective. For Black women, stable partnerships and labor market success are competing; for Black men, they are mutually supporting resources into mature adulthood, showing how class, gender, and race intersect in typical life course profiles.
Studies focusing on early single mothers in precarious employment with high welfare dependency (Edin and Kefalas 2011) have neglected the heterogeneity of Black women’s work-family experiences (figure 4). The full variety of Black women’s work-family lives debunks the focus on supposed deficits in previous studies, which often center on “socially problematic cases” and neglect the remarkably stable and successful careers in Clusters 3, 4, and 5 (Sprague 2005). For example, the Fragile Families Survey (Reichman et al. 2001), a nationally representative sample of nonmarital births in urban areas, by design excludes Black men and women who enter parenthood in marriage, even if they become single parents through separation later, and have more successful careers. Our study substantiates that racial differences in work-family lives are at least as large among highly educated women as they are among lower educated women (Sweeney and Raley 2014).
Early high fertility and single motherhood are also prime obstacles to high-prestige careers for White women (Abendroth, Huffman, and Treas 2014; Kahn, Garcia-Manglano, and Bianchi 2014). For White women, both single childlessness and late partnered parenthood co-occur with high-prestige careers, whereas only one of these family patterns coincides with relative career success for either Black women or Black men. White women’s privileged options to combine different family lives with high-prestige careers are not viable to the same extent for either Black men or Black women (table 3). For some White women, a competing relationship between partnered parenthood and occupational success transforms into a mutually supporting one once they enter high-prestige positions.
For our study cohorts, labor market dynamics were particularly powerful in stratifying possibilities to combine work and family. The FMLA was introduced after their active childbearing years and the PRWORA 1996 welfare reform increased market dependence in midlife, when many of them had young children and jobs had become more insecure (Western et al. 2012). The 1996 welfare reform targeted low-income workers, disproportionately Black men and women and single parents, and curbed opportunities for upward mobility. Early single parents (Clusters 1 in table 3, 22% of Black men, 34% of Black women, and 13% of White women) remain trapped in precarious unstable careers well into midlife, with likely long-lasting disadvantages in mature adulthood and old age. They combine precarious and unstable careers with childcare responsibilities and lack a stable coresidential partner to pool resources and divide work and care. For them, disadvantage accumulates across the life course—both over time, with enduring career instability, and across the life domains of work and family.
Our findings have limitations that should be addressed in future research. We analyzed a narrow group of birth cohorts (1957–1964) that could be observed into mature adulthood and only included Black and White men and women. Future research should include other combinations of race and gender and should extend to younger cohorts. Younger cohorts are more racially and ethnically diverse and have experienced more family complexity with the rise of (serial) cohabitation and union instability (Cherlin 2010); increased education and debt levels; diminishing returns on education, a rising service sector, great recessions in 2008 and following the COVID-19 pandemic; and continuing high incarceration rates among Black men. A trend toward more gender equality in employment has stalled since the 1990s (Blau, Brinton, and Grusky 2006). On the one hand, these developments might further polarize work-family experiences, increasing the gap between White men’s privilege and the specific constraints experienced by Black men, Black women, and (lower educated) White women. On the other hand, relative educational disadvantages, de-unionization, and the continued erosion of typically male, secure, working-class jobs could strengthen work-family interdependence for White men. Evidence from younger cohorts suggests that deteriorating economic circumstances have had similar effects of increasing single parenthood and family instability among White men and women (Furstenberg 2009) as we see primarily for Black men and women in our analysis. This suggests a class-based mechanism that operates irrespective of race and calls for class- rather than race-targeted policy responses (ibid.). Future research that includes younger cohorts could also address how work-family lives perpetuate across generations and add to the reproduction of gender and race inequalities.
Our study joins others (DiPrete and McManus 2000; Leopold 2018; Raley and Sweeney 2020) in highlighting that initial economic disadvantage in the family of origin and in work lives constrain and condition family trajectories more than the other way around. Consequently, interventions to equalize work-family lives should target racial and gender inequalities in social origin, education, jobs, and earnings and should not attempt to intervene in family lives. Prioritizing Black men’s employment to stabilize Black families and break generation-long cycles of racial disadvantage was already a core recommendation of the Moynihan report (Moynihan 1965), although the report was often cited to promote marriage counseling and criticized as antifeminist for problematizing matriarchal structures among the Black community (Massey and Sampson 2009). Similar to more recent discussions regarding the Moynihan report (Wilson 1987; Furstenberg 2009), our findings underscore the key role of education and employment in reducing race and gender gaps in work and family disadvantages that endure and accumulate over the life course. High-quality employment can grant access to mutually supporting benefits of occupational success and parenthood in stable partnerships, even if it is achieved later in life. Nonetheless, interventions in education and employment early in life (Heckman 2006) are the most promising way to halt the early onset of instability spillovers between work and family lives. Other countries demonstrate that public paid parental leave and basic welfare provisions open up opportunities for families that cannot afford market alternatives (Aisenbrey, Evertsson and Grunow 2009).
Our findings question the efficacy of marriage counseling—for instance, the federal marriage promotion provision of the 1996 welfare reform—in stabilizing families and reducing race and gender inequality in work-family lives (Furstenberg 2009). Especially for Black women, stable (married) partnerships are not associated with occupational success. Quite the contrary, the most occupationally successful Black women of our study cohorts (40%, Clusters 4 and 5 in table 3) had never experienced stable long-term partnerships into midlife. For White women, parenthood in stable partnerships and occupational success only transform from competing to mutually supporting once they have entered high-prestige jobs in their thirties. Moreover, marriage counseling would not reach the most disadvantaged parents, who are single at birth, or the enduringly single childless men in precarious employment. Overall, economic interventions appear most promising. Among family interventions, instead of marriage counseling, our findings suggest sex education and access to effective contraception. Our results confirm the well-known career advantages associated with delayed fertility for men and women. Among Black women, the timing of birth in the life course and how motherhood coincides with education and career development appear more important for occupational success than their relationship status. We conclude that equalizing economic starting conditions and labor market opportunities at various stages of the life course is most promising to reduce intersectional inequalities in work and family lives by gender and race.
About the Authors
Anette Eva Fasang is a professor of sociology at Humboldt University of Berlin and a fellow at the WZB Berlin Social Science Center. Her research interests include social demography, stratification, life course and comparative welfare state research, and methods for quantitative longitudinal data analysis. Recent publications include “Parenthood Wage Gaps across the Life Course: a Comparison by Gender and Race” (with Zachary Van Winkle) in Journal of Marriage and Family.
Silke Aisenbrey is a professor of sociology at Yeshiva University, New York. Her research interests include stratification, gender inequality, and life course research. Recent publications include “The Interplay of Work and Family Trajectories over the Life Course: The United States and Germany in comparison” (with Anette Fasang) in the American Journal of Sociology.
Acknowledgements
Footnotes
The NLSY79 survey is sponsored and directed by the U.S. Bureau of Labor Statistics and managed by the Center for Human Resource Research (CHRR) at The Ohio State University. Interviews are conducted by the National Opinion Research Center (NORC) at the University of Chicago.
Sensitivity analyses with other cost specifications (Hamming Distance, Dynamic Hamming Matching) generated qualitatively similar results.
The analyses used the TraMineR (Gabadinho et al. 2011) and Weighted Cluster Packages (Studer 2013) in R (R Version 4.1.1). The Mantel coefficients were calculated using code provided by Matthias Studer based on Piccarreta and Elzinga’s (2014) proposition.
To date, there is little guidance on assessing absolute Mantel values in life courses. We only interpret the differences between the intersectional groups.
Work-family interdependence is strongest among Black women with no high school or only high school diplomas (available from authors).
References
Milkman, Ruth and Eileen Appelbaum.
