The popular notion of glass ceiling effects implies that gender (or other) disadvantages are stronger at the top of the hierarchy than at lower levels and that these disadvantages become worse later in a persons career. We define four specific criteria that must be met to conclude that a glass ceiling exists. Using random effects models and data from the Panel Study of Income Dynamics, we examine gender and race inequalities at the 25th, 50th, and 75th percentiles of white male earnings. We find evidence of a glass ceiling for women, but racial inequalities among men do not follow a similar pattern. Thus, we should not describe all systems of differential work rewards as “glass ceilings” They appear to be a distinctively gender phenomenon.
Cotter et al. (2001) studied this question.