Withdrawn: does sector predict dividend cuts among the UK's biggest companies?

Withdrawn 10 October 2026. The first version of this note said that among UK-listed companies worth $50bn or more, the dividend cut rate varied 4.5-fold by sector, from 8.8% for consumer staples to 40.0% at the top. We rechecked it and the data cannot support a claim about sectors. Each sector in the table is between one and six companies, the top of the range is a single company whose dividend record in our data is faulty, and the gaps the note relied on are within what chance would produce. The original figures are kept below for the record. What we found.

What the first version claimed

The note took every UK-listed share with a market cap of $50bn or more at a semiannual observation date from 2015 to 2024 and measured how often its dividend was cut over the following 12 months, split by sector. It reported a spread from 8.8% to 40.0%, said the ranking held at every specials threshold, and said the gap between "stalwart" sectors (financials and energy) and the rest widened to 19.3 points among the highest yielders.

What we found

The note's analysis script was never committed, so we rebuilt its method from the note's own description. The rebuild gives 356 observations and a 28.4% cut rate for the tier, close to the published 348 and 27.3%, which confirms the reading. We then re-ran it on the dataset's own cut label: each company's full payment record, dated by payment date, in split-adjusted amounts. That label matches a fresh recomputation on all 23,056 rows.

Nineteen companies

The whole tier is 19 companies observed every six months. Consumer staples is Unilever and British American Tobacco. Healthcare is AstraZeneca and GSK. Energy is Shell and BP, and BP is counted three times, because its two preference share lines are listed with the company's market cap. The "other" row at the top of the range is National Grid alone. Industrials includes three miners, BHP, Rio Tinto and Glencore, because of how the dataset labels sectors.

A data fault at the top

National Grid's cached record in our data mixes adjusted and unadjusted amounts from one year to the next (its 2018 final appears as 20.8p, where the company paid about 30p), and that is where most of its nine "cuts" come from.

Currency

AstraZeneca, Shell and Unilever declare their dividends in dollars or euros. Our data records the London line in pence, so a fall in the exchange rate of more than 2% reads as a cut even when the dividend in the company's own currency did not change. AstraZeneca held its dividend flat in dollars for most of the period, and its sterling total still swings by more than 10% from one year to the next.

Chance

Resampling companies rather than observations, the gap between the "stalwart" group and the rest runs from -5.7 to +13.4 points at 90% confidence (10 companies against 11), so it cannot be told apart from zero. Among the highest yielders the range is -11.8 to +24.1. For a group of two companies no interval means much: resampling two companies gives only three possible combinations.

Stability

The order changes when the decade is split at January 2020. Healthcare cut 20.0% of the time before and 40.0% after. Financials went from 37.0% to 30.9%, and industrials from 27.1% to 41.2%.

What this does not show

Sector may well matter to dividend risk. Among the UK's largest companies there are too few of them to tell, and any sector reading of these 19 names comes down to a handful of individual dividend records.

Reproduce it

cd dividendmapper && npx tsx scripts/research/ttm-cut-sector-and-magnitude-label.ts

It rebuilds the first version's method, prints the rechecked tables, the company counts, the company-resampled intervals and the split halves, and fails if its recomputed outcome disagrees with the frozen label on any row.