What the gender pay gap number doesn't tell you

It isn’t an equal-pay measure

Paying a man and a woman differently for the same work has been unlawful in the UK since the Equal Pay Act 1970. The gender pay gap is a different, broader measure: it lines up every man’s hourly pay, takes the midpoint, does the same for women, and reports the difference. A company can pay every role equally by sex and still report a large gap, if men and women are not evenly spread across those roles.

It’s a role-mix measure, not a same-role measure

Because the figure is workforce-wide, it mostly reflects who holds which jobs, not what any individual is paid to do a specific job. An organisation with a female-dominated front line and a male-skewed leadership team will report a gap driven almost entirely by that structure. One filed example is 91% female overall but has a smaller share of women in its best-paid quarter than its worst-paid one — that gap between the two shares is most of the story, more than the single headline percentage is.

A shrinking gap can mean promotion — or attrition

A gap can narrow because women moved into better-paid roles, which is the outcome the figure is meant to encourage. It can also narrow because women left a lower-paid part of the business, changing the mix without anyone being promoted. The headline number can’t tell these apart on its own. The pay-quarter breakdown each employer also files — the share of women in the lowest- and highest-paid quarter of the workforce — is what starts to separate the two. See what the figures mean for how Remunify presents that breakdown alongside the headline gap.

One year is noise; a multi-year trend against the sector is signal

A single filing can jump for reasons that have nothing to do with pay policy: a restructuring, an acquisition, a change in who is counted as an employee. The more informative comparison is how an employer’s gap has moved over several years, set against how its own sector moved over the same years. A gap that widens while every peer narrows is a different finding from a gap that is simply high. Four real examples of that comparison, and the sector-level version of it, are on this site.

What it doesn’t cover

The legal duty to report generally applies to employers with 250 or more staff, so the published data does not describe the whole UK labour market, only its larger employers. Figures are also self-submitted and occasionally corrected. None of this makes the measure meaningless — nine years of it, compared consistently, is one of the more detailed public records of how work is distributed by sex in Britain. It just isn’t the single verdict a headline percentage is often treated as.