The 20 Best Dividend Companies Are Not the 20 Highest Yields

Search for the best dividend companies and you will get a list sorted by yield. It is the easiest ranking to build and the worst one to act on, because the yield goes up when the share price goes down. A yield ranking is therefore a rough list of the companies the market currently trusts least.

So we built a different one. We took the dividend aristocrats, the companies with the longest records of raising their dividend, and we scored them on five tests. Yield was one of the five, worth 15 points out of 100. The results were not what a yield sort would have given us.

Four famous names that did not make the list

Start with the failures, because they are the point.

• British American Tobacco yields around 6%. Its current run of increases is 7 years, which is short. It did not pass.
• Sanofi yields over 5%. It pays out more than 120% of its earnings, and its streak restarted 2 years ago. It did not pass.
• Unilever yields 3.7%. Its dividend growth over the last year was negative. It did not pass.
• Diageo cut its dividend. It now pays out more than it earns. It did not pass.

Every one of those is a household name that an income screen would rank highly. All four fail a basic durability test. That gap between "pays a lot right now" and "will keep paying" is the whole subject of this article.

The five tests

A company had to pass all five to be scored at all. A failure on any one removed it, whatever its other numbers looked like.

1. A streak of at least ten years

Ten years covers a recession and a rate cycle. A company that has raised its dividend through both has shown you something. A three-year streak in a good market has shown you nothing.

2. No cut, and no going backwards

A dividend cut is the outcome the income investor is trying to avoid, so a company that has just delivered one does not belong on a list of the best. We removed anything with negative dividend growth.

3. The dividend has to come out of free cash flow

Dividends are paid in cash, not in accounting profit. So the honest test is this one:

Free cash flow payout = dividends paid ÷ free cash flow

If that number is over 100%, the company is paying you with money it did not generate this year. It is funding the difference from the balance sheet or from borrowing. That can go on for a while. It cannot go on forever.

4. The payout cannot be stretched on both measures at once

Some companies look expensive on earnings but fine on cash, usually because of a large non-cash charge. That is common and often harmless. What is not harmless is a payout over 100% of earnings and over 90% of cash flow at the same time. There is no accounting explanation left at that point.

5. Debt under four times profits

A dividend and a debt pile compete for the same cash. When earnings fall, the lenders are paid first and you are paid last. Anything above four times EBITDA was removed.

The 20 that passed

Thirty-five companies went into the screen. Thirty passed all five tests. These are the twenty that scored highest.

1. Cincinnati Financial
2. ADP
3. General Dynamics
4. Chevron
5. Chubb
6. Exxon Mobil
7. BAE Systems
8. T. Rowe Price
9. S&P Global
10. Emerson Electric
11. Lowe's
12. Johnson & Johnson
13. Colgate-Palmolive
14. Procter & Gamble
15. Aflac
16. Illinois Tool Works
17. McCormick
18. Novartis
19. Roche
20. Abbott

Two honest observations about that list before you read anything into it.

It is not an income list. S&P Global sits at number 9 on a yield of under 1%. It scores well because it covers its dividend many times over and has raised it for decades, not because it pays you much today. A quality ranking and an income ranking are different things, and anyone selling you one as the other is not being straight with you.

The insurers are flattered. Cincinnati Financial, Chubb and Aflac all report investment gains inside net income, which pushes their payout ratios down and their scores up. The rule that produced the ranking is not perfect on financial companies.

Wolters Kluwer is the near miss worth naming. On the numbers it would have ranked around 14th. Its streak is 9 years. It failed on one year, and the rule did what rules do.

The number that did the most work

If you take one metric from this, take the free cash flow payout ratio. It separated the list more sharply than anything else.

Compare two companies that both look identical from the outside. Both are aristocrats. Both have raised the dividend for more than 50 years.

• ADP pays out 51.7% of its free cash flow, and grew the dividend 10.4% last year.
• Kimberly-Clark pays out 93.5% of its free cash flow, and grew the dividend 2.0%.

Kimberly-Clark yields nearly twice as much today. It also has almost no room left. When a company is already handing over 93 cents of every free dollar, a bad year has to come out of somewhere else, and the growth rate tells you management knows it.

The same pattern shows up across the higher-yielding names that passed. Swiss Re pays out 96.7% of free cash flow. PepsiCo pays 87.1%. T. Rowe Price pays 66.5%. Chevron, on a similar yield to several of them, pays 51.7%. The yield does not tell you which is which. The cash payout does.

Where these rules break down

Two kinds of company had to be pulled out and judged separately, and it is worth knowing why.

Utilities and REITs. Both spend heavily on the assets that generate their income, so free cash flow is negative by design. Test 3 would fail every one of them for the wrong reason. Realty Income shows a payout of 238% of earnings, which sounds alarming and is not: property depreciation crushes reported earnings, and the right measure for a REIT is adjusted funds from operations. A rule applied outside the situation it was built for produces confident nonsense.

And sometimes the alarm is real. Air Products has raised its dividend for more than 40 years. In the period we pulled it also had negative free cash flow and a loss. That is not an accounting artefact. A long streak is a record of the past, and it does not vote on next year.

Where these numbers come from

The universe is the S&P Europe 350 Dividend Aristocrats plus the US aristocrats as published by Sure Dividend. The company figures were pulled from stockanalysis.com in August 2026.

One thing we owe you. Free data has errors in it. In this pull, one page reported British American Tobacco's yield as 0.06% when the dividend and share price put it near 6%, because pence and pounds were mixed. We caught that one. We will not have caught all of them, and neither will any screen you run. Treat a number from a free source as a question to check in the company's own report, not as a fact.

The bottom line

Yield tells you what a company pays today. It tells you nothing about whether it will pay next year, and because it rises as the price falls, a yield ranking is quietly biased toward the companies in most trouble.

Five plain questions do a better job. How long is the streak? Has it ever been cut? Does the cash cover it? Is the payout stretched on both measures? Is there too much debt? None of those requires a spreadsheet, and together they would have kept you away from every one of the four famous names at the top of this article.