The intergenerational sources of the U-turn in gender segregation

In the early 1970s, the balkanization of the US labor market into “men’s occupations” and “women’s occupations” began to unravel, as women entered the professions and other male-typed sectors in record numbers. This decline in gender segregation continued on for several decades but then suddenly stalled at the turn of the century and shows no signs of resuming. Although the stall is itself undisputed, its sources remain unclear. Using nearly a half-century of data from the General Social Survey, we show that a resurgence in segregation-inducing forms of intergenerational transmission stands behind the recent stall. Far from serving as impartial conduits, fathers are now disproportionately conveying male-typed occupations to their sons, whereas mothers are effectively gender-neutral in their transmission outcomes. This segregative turn among fathers accounts for 47% of the stall in the gender segregation trend (between 2000 and 2018), while the earlier integrative turn among fathers accounts for 34% of the initial downturn in segregation (between 1972 and 1999). It follows that a U-turn in intergenerational processes lies behind the U-turn in gender segregation.

In all late-industrial countries, women and men continue to occupy very different types of jobs, with women concentrated in service and nurturing occupations (e.g., sales clerk and nurse) and men concentrated in manual and analytic occupations (e.g., laborer and computer scientists). The standard sociological account of such segregation treats it as a premodern relic that should gradually disappear as educational opportunities are equalized, egalitarian gender attitudes diffuse, overt and covert forms of employer discrimination are rooted out, and family-friendly workplaces and related labor market reforms are instituted (1, 2). Between 1970 and 1990, these types of institutional reforms indeed seemed to be bearing fruit, with most late-industrial countries experiencing sharp declines in segregation (3).This egalitarian turn proved to be short-lived. By the end of the 20th century, the decline in gender segregation had stalled in many late-industrial countries, including the United States (4–7). In Fig. 1, we present the US trend using the General Social Survey (GSS), the main data source for our analyses (8). The trend lines in Fig. 1 reveal that the downturn stalled well before integration was achieved (and Fig. 1B additionally suggests a resegregative uptick off this already high baseline). To completely eliminate segregation, ∼60% of US workers would have to shift to a different occupation, a stark result given the stated commitment to gender equality and equal opportunity in the United States. This hypersegregation translates into profound gender gaps in pay, authority, working conditions, and much more (9–12). Because occupational segregation generates such a wide range of unequal outcomes, it has become a conventional policy target among those who seek to reduce gender inequality.Open in a separate windowFig. 1.The U-turn in occupational gender segregation shown by (A) the index of dissimilarity and (B) a margin-free measure of the gender-by-occupation association. Analysis is based on the 1972 to 2018 GSS using women and men in the labor force, ages 25 to 64 (inclusive), with nonmissing data on age, gender, and occupation. n = 44,640. Occupations are coded with 1970 Census Occupational Classification (see SI Appendix, Appendixes B and D, for details). Trend is smoothed with locally smoothed regression (LOESS) with the span set at 0.75. For the definition of D and A, see Materials and Methods.The long line of research on the sources of the stall proceeds in part from this widely shared (albeit not universal) commitment to reduce gender inequality. Although far from conclusive or exhaustive, the available research suggests that 1) the persisting double burden of domestic work erodes the willingness or capacity of women to hold jobs that entail long hours or overwork (12–14); 2) the persistence of norms against women outearning their partners (in marriages between women and men) leads to settling and underachievement (15); 3) the persistence of essentialist beliefs about the types of occupations that women and men are qualified to undertake locks in a conventional division of labor (1, 16, 17); and 4) the continuing tendency of employers to discriminate against mothers (18) and those who participate intermittently in the labor force (19, 20) makes it difficult for women to break into male-typed occupations.^* Across these assorted supply-side and demand-side accounts, a common theme is that the easy desegregative gains have been creamed off, that a new equilibrium or natural rate (22) has been reached, and that puncturing this equilibrium may require different or more aggressive countermeasures.This interpretation rests on a distinction between 1) overt and easy-to-attack inequalities and 2) covert and difficult-to-attack inequalities. It is often argued, for example, that overt forms of hiring discrimination against women can be reduced through antidiscrimination law and open hiring practices, whereas subtler or covert forms are not as easy to address, at least with legal remedies. Likewise, the extreme essentialist view that women are utterly lacking in rational faculties was an early casualty of the gender revolution, whereas it has proven more difficult to root out a subtler form of essentialism that presumes that men are more likely than women to be “brilliant” or “geniuses” (23). By this logic, the gender revolution will only be reenergized by developing new interventions that are fine-tuned for the subtler, albeit still very consequential, mechanisms that lie behind gender inequality.This conclusion may well be warranted. It would, however, be premature to adopt this interpretation without a fuller understanding of the institutional sources of the stalling out. It is striking in this regard that none of the foregoing accounts directly features the role of intergenerational processes. To be sure, conventional accounts routinely invoke the segregative effects of gender-biased training, norms, and culture, but the social processes through which gender-biased training is delivered or gender-biased norms and culture are instilled are left unclear and thus may or may not have intergenerational sources. This is an unfortunate omission because it undermines our capacity to remediate well. As it stands, most currently popular interventions do not target intergenerational processes (24), an understandable state of affairs given that relatively little is now known about their role in the stalling out.There is, then, a worrying omission in our research on the gender U-turn. Although this omission might seem surprising, it has to be remembered that the contemporary family is often seen as a prime carrier of new forms of gender egalitarianism and accordingly an unlikely source of the stalling out. The new gender-egalitarian family features mothers who are increasingly likely to work, to commit to bona fide careers, and to espouse liberal attitudes (25). In this conventional characterization of familial change, it is notable that mothers are represented as a main force for change, whereas fathers remain in the background largely carrying on. This raises the possibility that fathers may not always be unfettered agents of egalitarian change and may, to the contrary, be implicated in the stalling out. The purpose of our paper is to explore this segregative-father hypothesis by melding models of segregation and intergenerational transmission.The results will show that the U-turn in the segregation trend is indeed linked to a U-turn in intergenerational processes. As we will discuss, a variety of mechanisms may lie behind the rise of segregative reproduction, yet most of them involve familial processes in some fashion. It follows that insofar as desegregation is a policy objective, it may be strategic to build interventions that address contemporary family processes.The two streams of mobility research upon which our analysis will capitalize are those examining 1) the net effects of mothers and fathers on outcomes and 2) the effects of intragenerational mobility on segregation. The first of these two streams is of course especially relevant when examining the intergenerational sources of segregation (26, 27). In recent decades, there has been a backlash against treating the father’s or “family head’s” occupation as a satisfactory measure of class origins (28), a long-overdue development given that a rising share of children are now growing up in more complicated families with many potential role models (29, 30). This newer line of multiple-parent mobility research has, however, focused mainly on examining the relative size of the effects of mothers and fathers, a type of “horse race” analysis that comes naturally when one seeks to rectify decades of scholarship that ignored mothers altogether. As important and influential as this research is, it cannot directly answer the research question that we are taking on, given that the implications of this horse race for occupational segregation depend on the extent to which class and gender-typing reproduction come together. To understand how intergenerational transmission contributes to gender segregation, we need to know 1) whether mothers are passing on female-typed occupations disproportionately to daughters and 2) whether fathers are passing on male-typed occupations disproportionately to sons. Although the intergenerational transmission of gender-typed aspirations is well understood (31, 32), our research takes the next step of building gendered mobility models that uncover whether actual reproductive practices are affected by the gender of the parent, the gender typing of the occupation, and the gender of the offspring (33).The second stream of mobility research relevant to our analysis examines the effects of intragenerational mobility on gender segregation (34). This line of research, which has a long history (35), examines how women and men move in and out of the labor force over their career, how women and men move in and out of gender-typed occupations over their career, and how such mobility has gendered sources (e.g., sexual harassment, gender discrimination, and the “second shift”). Within this literature, many scholars (36) have examined whether the gender of managers affects hiring and promotion decisions (and hence downstream segregation), a line of questioning that is formally similar to our own interest in whether the gender of the parent affects occupational outcomes. Because parents are engaged in ongoing socialization, aspiration development, human capital investment, and network provisioning, their effects on segregation may well be more consequential than the pinpoint hiring and firing decisions of managers. This is the core rationale for building intergenerational processes into a segregation trend analysis.