Enduring Inequality: Gender And Employment Income In Late Career

Abstract

Marriage, union membership, and managerial and professional occupations yield no significant earnings advantages for white women, in Ordinary Least Squares (OLS) regression models predicting logged employment income for whites still working in late career, surveyed in the first wave (1992) of the Health and Retirement Study. White men reap earnings benefits on all fronts, however, this is in addition to the benefits of self-employment combined with either marriage or professional occupation. Following Tilly (1998), we suggest that this enduring inequality is rooted in relations between men and women (especially marriage) and relations between men (closed status communities or old boy networks).

Notes

We wish to thank Yong Wang for assistance with the data analysis, which was funded in part by a 2002 Kinley Trust Grant, and Suzanne Bianchi for advice on the glass ceiling and the U.S. Census data.

¹Alessio and Andrzejewski (2000, p. 313) claim that the glass ceiling theory predicts “that the income gap between men and women widens as one moves higher on the income variable.” Current Population Survey (2002) data do not bear this out. Cotter et al. (2001) report, however, that the log-odds (or risk) of earning at or above the 25th, 50th, or 75th percentile of white male earnings diminish, for white women, as one moves from 25th to 75th percentile and diminish (relative to white males) as work experience increases. Since they do not control for marital status or occupation, we can only speculate as to what barriers might explain the predicted lack of white women at the higher levels of earnings. In any case, these data do not address the question of the size of the earnings gap.

²Literally, Tilly (1998, p.10) defines “exploitation” as “when powerful, connected people command resources from which they draw significantly increased returns by coordinating the effort of outsiders whom they exclude from the full value added by that effort.” “Opportunity hoarding” is “when members of a categorically bounded network acquire access to a resource that is valuable, renewable, subject to monopoly, supportive of network activities, and enhanced by the network's modus operandi.”

³This scale follows Wright and Perrone (1977) and Hogan and Perrucci (1998) by recoding years to correspond to educational credentials: “0” = under 8 years, “1” = 8 years, “2” = 9–11 years, “3” = 12 years, “4” = 13–15 years, “5” = 16 years, “6” = over 16 years.

⁴Mean differences in Table 3 (and, for that matter, in Table 2), are sometimes highly significant. Comparing Single Professional and Single Other earnings (in Table 3), for example, yields t = 3.1, p < .0005 (one tail). These statistics for each paired comparison can be computed from the data presented in Table 3 (or Table 2).