Dogs of the FTSE: the highest UK yielders cut their dividends nearly twice as often as the rest

Dogs of the FTSE is the British version of a famous American stock screen, a fixed rule for picking shares. Once a year you buy the ten highest-yielding shares in the index and hold them until the next reset. You get a big income straight away, from companies the market has gone off, and some of those recover.

But a yield that high is often the market telling you it expects the dividend to fall. Investors call that a yield trap. We wanted to know how often that happens, so we tested the screen on ten years of UK data.

The ten highest yielders cut their dividend within 12 months 29.0% of the time. Every other large UK company did so 14.2% of the time.

What we tested

We took every UK-listed company worth at least $2bn and looked at it every six months from 2015 to 2024, ranking the companies by yield each time and taking the top 10. Each company on each date counts as one observation. The classic screen resets once a year, and re-ranking every six months gave us twice as many snapshots to test.

For every observation we checked whether the company paid less in dividends over the next 12 months than over the 12 months before, by more than 2%. One-off special dividends were left out on both sides. We compared the top 10 with every other company on the list on the same date.

We ranked on the yield from dividends paid over the past year, divided by the share price on the day. The standard yield in our dataset is wrong for some companies that split or consolidated their shares. A broken yield usually looks very high, so a top 10 screen picks those errors out first. We left out 425 observations where the two yield measures disagreed by more than double, could not be checked against each other, or sat above 25%.

The test was written down before it ran, along with the size of effect the yield-trap idea would need to show: 10 percentage points. It is test 2 in our list of every strategy we have tested.

What we found

Those rates come from 170 observations of the top 10 and 1,694 of everyone else. The gap is 14.8 percentage points, so the top 10 cut nearly twice as often.

The gap showed up in both halves of the decade, and it held when we relaxed the data checks. Allowing for chance, we are 90% confident the true gap lies between 8.9 and 21.7 points. That range is well clear of zero. Its lower end falls just short of the 10 points we set as the bar before running the test, so the gap could be a little smaller than that bar.

What it means for your income

Take Priya, a hypothetical saver in her late fifties with a stocks and shares ISA who wants income now. Say she had followed the screen at any point in our test and held its ten shares for a year.

On these numbers, close to three of her ten companies would have cut. The typical holding paid her the same dividend a year later as the year before, a median change of 0%. Across the rest of the market, the typical company raised its dividend by 4.9%.

So Priya starts with a bigger income and a much higher chance that part of it shrinks. Whether the bigger cheque makes up for the cuts depends on share prices as well, and this test did not measure those.

What this test does not tell you

• Returns. It measured dividends. A share can cut and still be a good buy if the price had already fallen to reflect it. Total returns need a separate test.
• Takeovers. A company that is bought out stops paying, and that counts as a cut here. No top 10 observation lost its share price within the year. In the comparison group 26 did, through a takeover or a delisting.
• Currency. A cut is measured in sterling as paid. Some UK-listed companies, such as Shell and AstraZeneca, declare their dividends in dollars or euros, so part of a fall in sterling can be the exchange rate. Measured in each company's own currency, the gap is a little smaller, and the top 10 still cut far more often than the rest.
• Index membership. $2bn is a size line. It approximates the FTSE 100 and the larger end of the FTSE 250, but we did not have the index's own membership history to test against.

The method note lists every exclusion and lets you download the data behind the chart.

How to use this if you buy for yield

Treat a top-ten yield as a reason to look harder before you buy. Check whether earnings and free cash flow cover the dividend, whether the company has cut before, and whether its debt is climbing. The Dividend Safety Score checks each of these for a company.

Frequently asked questions

What is the Dogs of the FTSE strategy?

A simple income screen: buy the ten highest-yielding shares in the FTSE, hold them for a year, then reset. It borrows the idea from the Dogs of the Dow in the US. The bet is that unloved companies with big yields pay well while you wait for them to recover.

Do high-yield shares cut their dividends more often?

In our test, yes. Between 2015 and 2024 the ten highest-yielding UK companies worth at least $2bn cut their dividend within 12 months 29.0% of the time, against 14.2% for the rest. The gap held in both halves of the period.

Does this mean high-yield shares are a bad investment?

Not on its own. The test measured dividends, not total returns, and a share whose price already reflects a likely cut can still do well. It does show that a top-ten yield is often a warning, so check cover, debt and dividend history before you buy.

Why rank on dividends paid rather than the quoted yield?

The quoted yield in our data is wrong for some companies that split or consolidated their shares. A broken yield usually looks very high, so a top 10 screen picks those errors out first. We used the dividends paid over the past year divided by the share price, and dropped rows where the two measures disagreed by more than double.

The takeaway

Between 2015 and 2024, the ten highest-yielding large UK companies cut their dividend nearly twice as often as everyone else: 29.0% against 14.2% within a year.

If you buy the screen anyway, plan your income as if part of it will not last.

This is research on past data, not financial advice. Past dividends say nothing certain about future ones, and your own circumstances matter.