It is clear from research on corporate leadership that a glass ceiling prevents many women from occupying top executive or CEO positions. In new research, we find that the glass ceiling is far more extensive than previous literature has determined: women are not just excluded from top leadership positions, but rather they are excluded from nearly all top income positions regardless of occupation. We also find that progress on this front has been stalled for the last twenty years.

Scholars and journalists have paid increasing attention to people at the top of the income distribution, such as members of the top one per cent, because they control significant financial resources. For example, our calculations from the 2016 Survey of Consumer Finances show that top one per cent households receive nearly a quarter of all US income.

We also find that it takes at least $845,000 in household income to be in the top one per cent. This concentration of income is particularly troubling given that members of the top one per cent not only have disproportional financial resources, but they also exercise considerable political and social power.

Although top income earners are attracting more attention, the focus is typically at the household level. That is, discussion of top households tends to focus on differences between top one per cent households and the remaining 99 per cent. While this work is critical for understanding inequality among households, conceptualising the rich at the household level obscures whose income—men’s or women’s—leads to a household being in the top one per cent.

Because individuals who are in the one per cent based on their own income (i.e., without including their spouse’s income) likely accumulate more social, economic, and political power than their spouse, it is critical to distinguish how gender matters in these households.

In our study, we examine the extent to which women’s income contributes to a household’s one per cent status and the pathways associated with men and women making it into the one per cent. Specifically, we examine whether women’s own employment characteristics and education provide a personal pathway to one per cent status based on their own income or whether women primarily have one per cent status via marriage because of their spouse’s income.

We use data from the 1995–2016 Survey of Consumer Finances (SCF) to study these ideas empirically. The Federal Reserve Board’s SCF is a nationally representative cross-sectional survey conducted every three years, and provides highly accurate information on top-income households.

The one per cent glass ceiling

Our findings are stark. We find that married households rarely qualify for one per cent status based on women’s income alone. Specifically, Figure 1 shows that women’s income alone was sufficient for one per cent status in less than 2 per cent of elite households in 1995; the corresponding statistic was still only 4.5 per cent in 2013 and 2016.

We also find that women’s income is rarely necessary for married households to meet the one per cent threshold. Figure 1 shows that women’s income was necessary in about 9.6 per cent and 7.4 per cent of married one per cent households in 1995 and 1998, respectively; in 2016, the percentage had climbed some but nevertheless remained low at 15 per cent.

Figure 1. Women’s income is necessary or sufficient for household 1% status, married households, 1995–2016

Together, these findings indicate that men’s income largely determines a household’s one per cent status because most households would have one per cent status regardless of women’s income. Broadly, these findings indicate a clear male dominance of income resources in families in the upper echelon.

Of course, a minority of women do enter the one per cent as a result of their own personal achievements. We find that having higher education and being self-employed increase women’s likelihood of qualifying for one per cent status based on their own income. In fact, self-employment carries a significantly higher return for women than for men in earning elite-level income.

For example, women who are self-employed have 30 times higher odds of earning top one per cent income than non-self-employed women, but men who are self-employed have only 9 times higher odds of earning top one per cent income than non-self-employed men, a gender difference that is statistically significant.

This, of course, does not imply that self-employed women have higher rates of personal one per cent status than self-employed men (they do not as we also show in our study); instead, it implies that women may need self-employment more than men to reach this elite position.

Our findings reflect the fact that women who are not self-employed have virtually no chance of personally qualifying for one per cent status on their own; thus, self-employment is likely women’s only viable route to earning elite-level income.

Self-employment (via starting a successful business) may free ambitious and competent women from biases and discrimination that otherwise hinder their upward mobility in corporate America, despite the reality that women still face gender-based hurdles related to business start-up and growth. Given that men hold most high-income financial (e.g., hedge fund managers) and leadership positions (e.g., CEOs), men have routes other than self-employment to reaching personal one per cent status (even though self-employment is still an important pathway for men as well).

Although higher education and self-employment are important for women’s own economic mobility, we find that women’s main route to one per cent status is via marriage when their husband’s income is included in the household income total.

Specifically, married women are 992 per cent more likely to be part of a one per cent household than single women, holding constant other important characteristics (e.g., race, education, age). Married men are also more likely to have household one per cent status than single men (71 per cent more likely), however, the positive association is not nearly as strong for men as it is for women.

Clearly, women benefit from marriage by gaining access to their spouse’s income. But why is there a positive association between men’s household one per cent status and marriage? Two potential explanations exist, which future research should confirm.

First, as research on general household populations show, marriage tends to provide men access to their wives’ unpaid labor, which frees men to devote more time to work and take risks with their careers. Second, selection effects are likely at play: the characteristics that make men more likely to financially succeed (and push a household into the one per cent) also make them more likely to marry.

Importantly, we find that not all married women have similar chances of making it to a one per cent household. Rather, there are clear differences among married women in access to elite income positions: married women who have highly educated or self-employed husbands are best positioned to be in an elite household.

Considering that men’s income mostly determines a household one per cent status, men’s membership in a top one per cent household is less strongly tied to their spouse’s characteristics. Ultimately, this highlights the different and unequal pathways associated with men’s and women’s access to these elite positions.

Progress over time? 

Our work shows that women are clearly less likely than men to earn elite-level income, but has this female disadvantage decreased over time? Although the statistics shown in Figure 1 seemingly suggest some gender gains, when we use more rigorous methods and control for important factors (age, education, etc.) we find that women have not made any gains over the past 20 years in closing the gender gap in earning top one percent income. Their chances of earning elite-level income, relative to men, are the same today as they were in the mid-to-late 1990s.

How is this possible? Women now own one-third of all US businesses, and graduate from MBA, MD, and JD programs in comparable numbers as men. They also represent nearly half the workforce in the United States. Despite these gains, the federal government has made no major legislative changes to reduce work/family conflict over the past twenty years, conflict that more frequently negatively affects women’s careers than men’s.

Moreover, although more women own businesses today, the revenues earned by those which are female-owned have barely increased in recent decades; similarly, women continue to employ far fewer employees than male-owned businesses. Perhaps most striking, high-income women have made the least amount of progress in closing the gender gap over the past few decades, such that the gender gap is now largest among high-income workers compared to all other socioeconomic classes. This is partly because occupational segregation among men and women remains stark, in which women have made few inroads into exceptionally high-paying occupations.

Broader implications 

Our research suggests that the glass ceiling is more expansive than previously thought. Women not only face barriers accessing leadership positions, but they also experience difficulty accessing all top income positions. This has broad implications because people in the one per cent have significant influence over our political system and they wield substantial economic power through major business decisions and donations to community organisations.

Women may be part of many married one per cent households, but until they gain parity in income resources, men will likely continue to exercise the majority of substantive power within this group.



Jill E. Yavorsky is an assistant professor of sociology at the University of North Carolina, Charlotte.



Lisa Keister is professor of sociology at Duke University.




Yue Qian is an assistant professor of sociology at the University of British Columbia.