How to Read a Wage Ladder Without Mistaking It for a Job Count

The answer to the central question is now clear: similar job-to-job switching can produce different wage ladders because a switch records movement, not the value of the destination or the return it generates. Men and women can change employers at nearly the same rate while differing in employment continuity, hours, occupations, amenities, starting wages, and bargaining positions. This final synthesis connects those elements so readers can judge a labor-market change without treating one wage statistic as the whole story.

The ladder must be read across several comparisons

The first comparison is a level versus a path. The Bureau of Labor Statistics chart places the wage dimension in view by reporting median usual weekly earnings for full-time wage and salary workers. In the second quarter of 2026, the all-occupation figures were $1,380 for men and $1,131 for women, while management, professional, and related occupations showed $1,928 and $1,476. Service occupations showed $920 and $730. These figures matter: the occupation in which people work is associated with very different weekly earnings, and the male-female contrast remains visible within the broad categories.

But a current occupation wage is a position, not a route. The chart does not show what workers earned before reaching that occupation, whether they changed employers, whether they moved into part-time work, or which amenities and productivity characteristics accompanied the job. It cannot tell us whether the gap was created at entry, widened through later raises, or partly offset by a more valued work arrangement. The chart is therefore well suited to describing the wage landscape and poorly suited to describing the climb through it.

The second comparison is a rate versus a return. The Chicago Fed paper finds nearly identical job-to-job transition rates for men and women across the CPS, SCE, and SIPP, including with the paper's demographic and occupation controls. This is evidence that the frequency of direct employer-to-employer movement is similar under the study's definitions. It is not evidence that the moves have the same destinations, wages, hours, amenities, or future consequences.

The surrounding flows make that limitation concrete. Women separate from employment more often and re-enter less often; in the paper's CPS sample, the employment-to-population ratio is 66.5% for women and 79.1% for men. Among direct job-to-job moves, roughly 25% of women's transitions lead to part-time work, compared with about 10% of men's. These differences can affect current weekly earnings and the position from which later growth is measured even when the E-to-E headline is similar.

The third comparison is a raw change versus a change from a comparable starting point. The paper reports almost no gender gap in year-over-year pay growth before prior pay is taken into account. After controlling for prior pay, the female coefficient is −0.110 for log hourly-wage growth and −0.141 for log weekly-earnings growth. The estimates support a flatter female wage ladder in the studied sample, but they are regression comparisons of average log changes, not dollar losses assigned to every woman and not a randomized estimate of the effect of a particular switch.

The fourth comparison is an occupation measure versus an individual measure. Conditional occupation-level wage growth in the occupation-switch analysis shows a gender gap of about −4.9 log points, while individual wage growth shows a gap of about −11 log points. The two values answer different questions. The occupation measure captures the average wage associated with a destination occupation; the individual measure can reflect raises, movement between firms, changing responsibilities, or other changes that occur within an occupation. The difference is not a disagreement to be averaged away. It is evidence that aggregation hides part of the ladder.

The fifth comparison is wage versus job value.