Screening Dividend Stocks: Dividend History and Financial Strength

One question, asked several different ways

So this lesson isn't about finding the biggest yield. It's about one question, and everything below is a different way of asking it:

Will this dividend still be paid in five years?

That's the whole job. You're making a forecast about a company you don't run, using numbers it published about a year that's already over. Nothing you're about to read settles that forecast. Each number is a piece of evidence — some strong, some weaker than it looks — and your judgment is what you build out of them.

Hold onto the distinction between income investing — buying stocks for the dividends they pay and keep paying — and growth investing. A growth investor is trying to buy low and sell high, so the price is the point. You're trying to own a payment that arrives whether the price is up or down. That changes what counts as good news. A dividend that survives a bad year is a win even if the share price didn't move at all.

What a payment record proves, and what it doesn't

Start with history, because it's the easiest thing to look up and the easiest thing to over-read.

Yield against a benchmark

A yield on its own is a number with nothing to lean against. To make it mean something, compare it to a benchmark — the index you measure performance against. A broad one like the S&P 500 works, and its average dividend yield is a number you look up rather than memorize. Use it as a yardstick in both directions. A stock yielding less than the index is paying you below-average income, which doesn't disqualify it, but it does mean the income case has to come from somewhere else. A stock yielding several times the index is the more interesting reading: unusually high against a broad average is a warning, not a bargain — it's the market saying it doubts the payment.

Figure

The far-above bar is the one to be careful with. An unusually high yield is a question, not a prize — a point the last lesson made and this one keeps making.

Yield against its own past

The more useful comparison is the stock against itself. Put the current yield next to its five-year average yield. If today's yield sits at or above that average, something has changed — and there are exactly two candidates: the dividend went up, or the price went down. You cannot tell which from the yield. You have to go look.

Yields also move for reasons that have nothing to do with the business: a one-off special payment, a stock split, an accounting quirk in a single quarter. This is precisely why the yield is a starting filter and never a verdict.

The honest problem with streaks

Companies advertise long records of raising the dividend, and investors treat those records as proof. They are genuine evidence of something real: a management team that has repeatedly chosen to protect the dividend when it could have spent the money elsewhere. That culture exists and it matters.

Here's the part nobody puts in the marketing. Every company that ever cut its dividend had an unbroken record right up until the day it didn't. A twenty-year streak is a description of twenty years that already happened. It is not a commitment, it is not a contract, and the board can end it at the next meeting without asking you.

That's the limit of history: it's all backward-looking. Which is why the rest of the lesson is about the money.

Where the money comes from, and who gets paid first

Video coming soon

This lesson explains the idea in full without it.

A dividend is not a bill the company owes you. It's what's left after everyone the company does owe has been paid.

Follow a dollar of sales through a business, in order:

1. Suppliers and payroll — the cost of actually making the thing.
2. Interest on debt. Lenders have a contract. Miss a payment and you're in default.
3. Taxes.
4. What survives is earnings — the company's profit.
5. Out of those earnings, the board decides whether to declare a dividend.

Free cash flow: is the cash actually there?

Earnings are an accounting figure, and accounting involves judgment. Cash doesn't. Free cash flow is the cash a business generated and got to keep:

Free cash flow = cash from operations − capital expenditures

Cash from operations is money the core business actually brought in. Capital expenditures are what the company had to spend on the physical stuff that keeps it running — equipment, maintenance, buildings. Subtract the second from the first and you have money that can go to dividends, to paying down debt, or back into the business.

Say fictional UVWXYZ generated $500 million from operations and spent $200 million on equipment and maintenance. Free cash flow is $300 million. If UVWXYZ paid $150 million in dividends, half its free cash flow went to shareholders and half stayed in the company. If instead it paid $290 million, the dividend consumed nearly everything the business produced — and next year's equipment breakdown has nowhere to come from.

Free cash flow that grows over several years suggests the company's cash position is strengthening. Free cash flow that shrinks while the dividend rises is a countdown.

The payout ratio is the number that carries the weight

If you take one metric out of this lesson, take this one. The payout ratio is the share of earnings paid out as dividends. High can mean unsustainable — and here's the mechanism, because the reason matters more than the rule.

Compare two fictional companies. Both pay their shareholders every quarter. Both have paid for years.

Right now these look similar from the outside. Both are paying. ABCDEF is paying more than twice as much, which makes it the more attractive one on yield.

Now run a bad year. Business slows and earnings fall 30% at both companies, to $2.80 per share.

UVWXYZ absorbs the year. The dividend is paid out of earnings, there's still money left to reinvest, and shareholders may never notice anything happened.

ABCDEF has three options and none of them are good: borrow the difference, drain its savings, or cut the dividend. The first two work once or twice. The third is what usually happens — and dividend cuts tend to hit the share price on the same day, so you lose the income and the value together.

Figure

That's the whole idea. A payout ratio isn't a grade. It's a measure of cushion — how much can go wrong before the dividend has to come from borrowing, from savings, or from your pocket. A company paying out nearly everything it earns has no cushion, and it will look completely fine until the first year it isn't.

The same ratio on a real company

UVWXYZ and ABCDEF were invented, and one of them had to be — its job was to run out of cushion on schedule. Here is the same arithmetic on a real business, and the wrinkle no invented example can produce.

The Campbell's Company (ticker CPB — the soup) paid a $0.39-per-share dividend all four quarters of the year before this snapshot: $1.56 for the year, actually paid, not annualized. For fiscal 2025 it reported earnings per share of $2.01 under GAAP — the standard accounting rulebook every U.S. public company files by. The same release also reports adjusted EPS of $2.97: earnings with items management considers one-time, like an impairment charge, added back. One year, one company, two earnings figures. Now compute the payout ratio with each:

The same real dividend is either consuming most of what the company earns or barely half of it, depending entirely on which earnings figure you divide by. Neither computation contains an error. The fictional table above couldn't teach this, because inventing both earnings numbers means inventing the gap between them — here the gap is real, and deciding which number better describes the business's ability to keep paying is precisely the judgment the ratio was supposed to spare you.

And the cushion wasn't standing still. At the snapshot date, the company's own guidance projected adjusted EPS of $2.15 to $2.25 for fiscal 2026 — a decline it put at 26% to 23%, measured against yet another denominator: not the $2.97 above but $2.91, fiscal 2025 restated as if it hadn't run an extra, 53rd week. Run the $1.56 dividend against the middle of that range and even the adjusted payout ratio lands just above 70%. The bad-year scenario in the fictional table isn't a scare story; companies publish theirs in advance, and the cushion you computed from last year's report can already be scheduled to shrink. None of this says the dividend was safe or in danger — it says a payout ratio is a fraction whose numerator is a board's choice and whose denominator is an accountant's, and both move.

Two more angles: management and debt

Return on equity

Return on equity (ROE) measures how much profit management generates from the money shareholders have put in. You want a business that takes a dollar and reliably makes it more than a dollar, because that's ultimately where dividends come from.

ROE varies widely by sector, so a number that's excellent in one industry is unremarkable in another, and the market-wide average moves with the profit cycle. That makes it a poor thing to memorize and a good thing to look up. The comparison worth making is a company against other companies in its own industry, and against its own ROE over the past several years — a business whose ROE has been sliding for five years is telling you something a single year's figure can't.

ROE has a blind spot, and it's a big one: it ignores debt entirely. A company can raise its ROE by borrowing heavily, because borrowed money isn't shareholder equity. A high ROE and a mountain of debt is a combination that looks like skill and behaves like risk.

Leverage, and why lenders outrank you

Leverage is a company borrowing to fund assets. It isn't automatically bad — it's how most businesses get built. But go back to the order of payments: interest comes before earnings, and earnings come before your dividend. Every dollar of debt service is a dollar that reaches the lender before it can reach you, and the lender's claim is enforceable while yours is a board's preference.

Two straightforward ways to read it:

Current ratio = current assets ÷ current liabilities. This covers the short term — bills due inside a year. Above 1 means the company has at least a dollar on hand for every dollar coming due. Below 1 means it doesn't, and something has to give.

Long-term debt to capital = long-term debt ÷ total capital. This covers the rest. Lower means less leverage. Under 100% means the company's capital exceeds what it owes long-term.

Neither one settles anything by itself, and "normal" leverage is wildly different for a utility than for a software company. If a company sits outside the range you'd expect, that's your cue to read further — check whether it recently made a large acquisition, and compare it to companies in the same sector rather than to the market at large.

Paying a fair price for the income

A durable dividend bought at a bad price is still a bad purchase. The price-to-earnings (P/E) ratio — share price ÷ earnings per share — tells you what you're paying per dollar of profit.

If fictional UVWXYZ trades at $100 and earned $10 per share last year, its P/E is 10. Investors are paying $10 for each $1 of annual earnings.

A high P/E generally means the market expects earnings to grow substantially — that's a growth stock, and it's a different bet than the one you're making. A very low P/E is not the bargain it appears to be either; prices get that low when the market believes something is genuinely wrong with the business, and often the market is right. Income investors usually want the unexciting middle: established, profitable, priced like what it is.

What a written criteria list actually looks like

Here is the point of this section, before the table: the numbers below are one investor's choices, not laws of finance. Nothing in the world says a payout ratio must be under 70%. Someone else, with a different income need and a different tolerance for a cut, would write different numbers and be equally defensible. What is not optional is writing them down — because a threshold you set in advance is a decision, and a threshold you adjust while looking at a stock you already like is a rationalization.

Most brokers offer a stock screener, a search tool that filters the market down by criteria you set. It's the mechanism. This table is the kind of thing you'd feed it:

Read the right-hand column, not the middle one. Every threshold in the middle is arbitrary in a way the reasoning beside it isn't. If you copy this table without being able to say what each line is evidence of, you've adopted someone else's opinions and called it a plan.

Here's the sharpest demonstration this lesson has to offer, and it takes one row. Run the real company from the payout section against "payout ratio under 70%": on GAAP earnings CPB fails the screen at 78%; on adjusted earnings it passes at 53%. Same company, same dividend, same threshold — the screen's answer flips on which earnings figure got typed into the denominator, a choice your screener's data feed made for you, silently, before you ever saw the list. That isn't an argument against screening. It's the reason the output of a screen is a reading list and not a verdict.

A screen also doesn't decide anything. It hands you a shorter list. Everything a screen catches passed a numeric test on numbers a company reported about a period that's already over — and the numbers are the beginning of the reading, not a substitute for it. What you do with that shorter list is the next lesson.

Key takeaways

• Every metric here answers one question: will this dividend still be paid in five years? Dividend history, payout ratio, and financial strength are all evidence toward that single forecast — not a checklist to tick.
• A long streak of dividend increases is real evidence of a management culture that protects the dividend. It is also exactly what every company that ever cut its dividend had, right up until it cut. History is backward-looking; treat a streak as a reason to look closer.
• The payout ratio measures cushion. A company paying out nearly all of its earnings has none — one bad year and the dividend has to come from borrowing, from savings, or from a cut. Compare within an industry, never across — and know which earnings figure the ratio was built on: a real company reports both a GAAP and an adjusted EPS, and the cushion you compute can flip between them.
• Debt payments are contractual and come before earnings; your dividend is optional and comes last. That ordering is why leverage matters to an income investor even when the business looks healthy.
• Screening thresholds are editorial choices, not rules of finance. Write yours down in advance, know what each one is evidence of, and don't loosen them for a stock you've already decided you like.

• Every number in this lesson is trying to answer one question: will this company still be paying five years from now?
• A long record of paying is real evidence of something. It is also exactly what every company that ever stopped had, right up until it stopped.
• Check how much of what a company earns it hands out. Handing out nearly all of it means no cushion when a bad year arrives.
• There is more than one way to measure what a company earned, and the same company can look comfortable by one measure and stretched by the other. Always notice which one you were handed.
• Everyone the company borrowed from gets paid before you do. You are last in line, every time.
• Decide your rules before you find a company you like, and write them down. A rule you bend while looking at something is not a rule.