2.14 The chief executive pay ratio is labelled 'Option B' but uses the median of [n] employees at each percentile, not the three best-equivalent employees the Regulations describe
UK Company B · a UK-listed technology and logistics group · Regulator-style question
In brief
The report computes the chief executive officer (CEO) pay ratios of [n]:1, [n]:1 and [n]:1 under 'Option B', but footnote 2 says it identified [n] employees at each of the 25th, 50th and 75th percentiles and used the median of each group as the comparator. Option B in the Regulations identifies three UK employees as the best equivalents of those percentiles, with adjustment allowed to make them representative. The 2024 report used the same approach.
The evidence, as filed
Screenshot withheld to protect the company’s identity: Annual Report 2025, page 150; Annual Report 2024, page 139. Their wording or layout could lead back to the company even with names and figures masked; the evidence list below records what each page shows.
What the report shows, and what the rule requires
| What the annual report shows | What the rule requires |
|---|---|
| What the annual report shows Method stated as Option B; comparator at each percentile = median pay of a group of [n] employees (p.150 footnote 2). FY25 comparators: total pay [amounts]. | What the rule requires Either three best-equivalent employees under Option B (with any adjustment explained), or the method described as Option C with the reasons the Regulations require for that choice. |
The rule
SI 2008/410 Schedule 8, paragraph 19D(4)
Version relied on: Large and Medium-sized Companies and Groups (Accounts and Reports) Regulations 2008 (SI 2008/410), in force from April 2008, as amended (the directors’ remuneration report Schedule 8 as substituted from October 2013); revised text, legislation.gov.uk
Under Option B the company must determine Y25, Y50 and Y75 as follows— (a)as a starting point, use the most recent hourly rate gender pay gap information for all UK employees of the company to identify three UK employees as the best equivalents of P25, P50 and P75
Why it applies here: a median of [n] is not three individual employees; it may be defended as an adjustment under sub-paragraph (4)(c), or it may be closer to Option C, which allows other data.
Why it matters
The statutory ratio may not be computed under the method it is labelled as; comparability with other companies' Option B ratios is reduced.
Exposure
A question on method; no penalty attaches.
Our workings
the ratio table labels every year 'Option B' and footnote 2 describes identifying [n] employees at each percentile from the gender pay gap data and reporting the median of each group. The 2024 report carries the same footnote. Law: paragraph 19D(4) of Schedule 8 says that under Option B the company uses the gender pay gap information 'to identify three UK employees as the best equivalents of P25, P50 and P75', and may then adjust their pay figures so they are reasonably representative.
the printed ratios reproduce from the printed figures ([n] / [n] = [n]; [n] / [n] = [n]; [n] / [n] = [n]), so the arithmetic is sound.
The upper-quartile comparator ([amount]) sits only [amount] above the median ([amount]), which a reader may want explained.
| Percentile | CEO single figure (£'[n]) | Comparator total pay (£'[n]) | Ratio |
|---|---|---|---|
| 25th | [n] | [n] | [n] / [n] = [n] |
| Median | [n] | [n] | [n] / [n] = [n] |
| 75th | [n] | [n] | [n] / [n] = [n] |
Management’s best defence, and our reply
The Committee would say Option B expressly allows adjustment so that the comparators are reasonably representative, and that taking the median of a small group around each percentile is exactly such an adjustment, reducing the effect of one unusual employee. It would add that the method is explained in the footnote and has been applied consistently, so readers are not misled. That is a reasonable reading, which is why this is a question; the answer still needs to be stated, because the Regulations frame Option B around three identified employees and the label alone does not tell the reader that the method departs from that.
The fix
- Either describe the approach as Option C and give the reasons, or state in footnote 2 that the group medians are an adjustment under Option B to make the comparators reasonably representative, and why.
Question for the company
Does the Committee regard the use of the median of [n] employees at each percentile as Option B with an adjustment under paragraph 19D(4)(c), or as Option C, and why is the 75th percentile comparator ([amount]) almost the same as the median ([amount])?
Evidence relied on
| # | What it shows | Type | Where |
|---|---|---|---|
| 1 | FY25 ratios, labelled Option B | Document | Annual Report 2025, page 150 |
| 2 | The method uses groups of [n] employees, not three | Document | Annual Report 2025, page 150 |
| 3 | Comparator = median of each group | Document | Annual Report 2025, page 150 |
| 4 | Comparator pay figures used in the workings | Document | Annual Report 2025, page 150 |
| 5 | The 2024 report used the same method | Document | Annual Report 2024, page 139 |
Prepared by Pramanex from the company’s published annual reports, without the company’s involvement, to show how an annual report review reads. The company is not named, and nothing on this page is a statement about any identifiable company. Observations are not an audit, review or other assurance engagement, and not legal advice. A regulator-style question is a question, not a conclusion.
Want this on your own annual report, before it is published?
Start with three free findings on your latest published annual report, or talk to us about a fixed-fee pre-publication review.
Your annual report will be in draft before the year-end audit.
Have it read before it is published.
Under NDA, with our reviewer on your insider list under UK MAR.