A dividend cut is one of the most expensive single events a value investor can absorb. The stock typically drops 15-30% on the announcement, the income stream you were modeling disappears, and the entire thesis - which was usually anchored on "this is a reliable cash-flow business" - has to be rewritten.
So evaluating dividend safety BEFORE buying is one of the highest-leverage things you can do as a long-term investor. And yet, most retail investors look at one metric (dividend yield) and call it research. Yield is the worst possible standalone signal - a 9% yield often means the market expects a cut.
This post is a transparent walk-through of the five-metric framework we use to compute invest-like's dividend safety A/B/C/D grade, the exact thresholds for each tier, and a comparison to how Simply Wall St scores the same companies differently. By the end you should be able to compute your own dividend safety read on any ticker in under 5 minutes.
What dividend safety actually means
A dividend is safe if the company can keep paying it through realistic downside scenarios. "Realistic downside" is doing a lot of work in that sentence.
A 100% safe dividend would mean the company can pay it through any recession, any leverage shock, any management change, and any cost surge. No dividend is truly 100% safe - even Berkshire's holdings have seen specific positions cut dividends.
A practically safe dividend means:
- The current cash generation comfortably covers the dividend (not just earnings - cash).
- The balance sheet is strong enough to maintain the payout through a recession without taking on dangerous debt.
- Management has a demonstrated multi-year history of prioritizing the dividend.
- The underlying business is structurally cash-generative, not cyclically.
- The payout ratio leaves room for the next 3-5 years of expected dividend growth.
The five-metric framework below tests each of those criteria with specific thresholds.
Metric 1: Free cash flow payout ratio
Earnings payout ratio (dividends / earnings) is the standard metric. It's also the most misleading.
Earnings include non-cash items (depreciation, amortization, stock-based compensation) that don't fund dividends. A company can show "100% earnings payout ratio" while actually being safely cash-flow-positive - or vice versa.
The better metric: Free cash flow payout ratio = Dividends paid / Free cash flow.
Thresholds:
- < 50%: very safe. Plenty of room for dividend growth and reinvestment.
- 50-70%: comfortable. Sustainable, modest growth.
- 70-85%: tight. Growth will be limited; one bad quarter hurts.
- > 85%: at risk. The dividend depends on FCF holding up exactly.
- > 100%: distressed. Company is borrowing or eating cash to pay the dividend.
In invest-like's grade: FCF payout ratio above 85% drops the grade by one tier; above 100% drops it by two.
Worked example: Coca-Cola (KO) - 70-75% FCF payout ratio in 2024. Comfortable. KO has been in this range for over a decade and grows the dividend ~5% annually within it.
Metric 2: Net debt to EBITDA
A dividend can look safe on the current FCF coverage but be at risk because the balance sheet is over-levered. If the company has 5× debt-to-EBITDA, any operational hiccup forces it to choose between debt service and dividend payments. Lenders win that fight.
Thresholds:
- < 1×: very safe leverage profile
- 1-2×: safe for most industries
- 2-3×: industry-dependent (investment-grade utility = fine, consumer discretionary = risky)
- 3-4×: stressed; dividend is at meaningful risk in a downturn
- > 4×: high risk; rating downgrade likely if conditions worsen
In invest-like's grade: net debt to EBITDA above 3× drops the grade by one tier; above 4× drops it by two.
Worked example: AT&T (T) pre-WarnerMedia spinoff - net debt to EBITDA was 3.5×+. Despite a high yield (7%+) and strong nominal FCF coverage, the dividend was cut in 2022 when the leverage became unsustainable post-spinoff. The leverage metric was the leading indicator.
Metric 3: Dividend coverage by recurring cash flow
Some companies generate strong total FCF but get there through one-time items - asset sales, working capital releases, tax refunds. These boost cash flow this year but won't recur.
The right test: can the dividend be covered by recurring operating cash flow (operating cash flow MINUS one-time items, divided by dividends paid)?
This is harder to compute mechanically because "recurring" requires judgment. The proxy we use:
- Operating cash flow trailing twelve months
- MINUS the largest one-time gain in OCF (asset sale, settlement, etc.)
- Divided by dividends paid trailing twelve months
A recurring coverage ratio below 1.0× over multiple years is the strongest single warning sign of an upcoming dividend cut.
Worked example: General Electric (GE) 2017-2018 - reported earnings looked weak but headline FCF was supported by one-time pension funding releases and asset divestiture proceeds. Strip those out and recurring OCF dropped below the dividend payment. GE cut the dividend twice in 14 months.
Metric 4: Dividend growth history through cycles
A dividend's safety is partly a cultural property of the company. Boards that have prioritized the dividend through 30 years of cycles will defend it harder than boards that started paying one in the last decade.
The dividend aristocrats (S&P 500 companies that have raised dividends for 25+ consecutive years) and dividend kings (50+ years) carry an implicit signal: management considers the dividend a near-sacred commitment. Cutting it would be a permanent reputational hit they're motivated to avoid.
In invest-like's grade:
- 25+ years of increases (dividend aristocrat): +1 tier
- 50+ years (dividend king): +1 tier (caps at A regardless of other metrics, with overrides if leverage is extreme)
- < 5 years of dividend history: no boost, slight penalty if leverage is also weak
- Any cut in the last 10 years: significant penalty
The history doesn't make the future, but it shifts the burden of proof. A company that's never cut in 50 years would have to face a truly catastrophic situation to cut now.
Worked example: Procter & Gamble (PG) - dividend king with 65+ years of consecutive increases. Even during the 2020 COVID demand shock, the dividend was raised. The cultural commitment is a real economic factor.
Metric 5: business cyclicality and revenue volatility
Dividend safety depends on the volatility of the underlying cash flows. A company with low-volatility cash flows (consumer staples, utilities, healthcare) can run higher payout ratios safely. A cyclical (homebuilders, semiconductors, autos) cannot.
We measure cyclicality via revenue standard deviation over 10 years:
- CV < 10%: stable (KO, JNJ, PG)
- CV 10-25%: moderately stable (MSFT, AAPL)
- CV 25-50%: cyclical (autos, industrials, retail)
- CV > 50%: highly cyclical (semis, homebuilders, miners)
Highly cyclical businesses get a one-tier penalty in the dividend safety grade regardless of their current FCF coverage. The reason: their current coverage might look fine, but the next downturn could cut revenue 40% and break the payout structurally.
How the A/B/C/D grade is computed
The aggregation is intentionally simple:
- Start at the median of each metric's tier. Most companies land at B by default.
- Apply metric-specific adjustments per the thresholds above.
- Cap the final grade at A (no A+ tier - the marginal distinction adds confusion).
- Floor at D (no F - very few companies are in distressed-cut-imminent territory at scale).
The grade is a SUMMARY. It's not the analysis itself. Every grade comes with the underlying metric scores so you can see why the company landed where it did.
A grade (very safe)
- FCF payout ratio < 60%
- Net debt to EBITDA < 2×
- Recurring coverage > 1.5×
- 25+ years of dividend increases preferred (or 10+ years with no cuts)
- Stable to moderately stable business
Typical A names: PG, KO, JNJ, MSFT, V, MA, COST.
B grade (safe)
- FCF payout ratio 60-80%
- Net debt to EBITDA 2-3×
- Recurring coverage 1.1-1.5×
- 10+ years of consistent payments
- Stable to moderately cyclical business
Typical B names: PEP, MCD, MMM, IBM, HD.
C grade (caution)
- FCF payout ratio 80-100%
- Net debt to EBITDA 3-4×
- Recurring coverage 0.9-1.1×
- Recent cuts or new dividend (< 5 years)
- Cyclical exposure
C grades are companies where the dividend is plausibly safe in normal conditions but at material risk in a downturn.
D grade (at risk)
- FCF payout ratio > 100% OR
- Net debt to EBITDA > 4× OR
- Recurring coverage < 0.9× OR
- Highly cyclical AND high payout
A D grade is a "the dividend is structurally at risk; if you're owning this for the income, reconsider."
Where this disagrees with Simply Wall St
Simply Wall St also publishes dividend safety analysis, but the methodology differs in two material ways:
- They weight dividend yield as a primary positive factor. invest-like ignores yield entirely. Yield is a market output (price-derived). High yield from a dropping price is a warning signal, not a quality signal.
- They include forward-looking analyst estimates. invest-like uses only trailing twelve months of actual cash flow data. Analyst estimates have systematic optimism bias and add noise to safety calculations.
The result: Simply Wall St tends to rate high-yield stocks higher than we do. We tend to rate established compounders with low yields (1-2%) higher than they do.
Neither methodology is "right" - they answer slightly different questions. Simply Wall St answers "is this dividend going to keep paying its current yield?" invest-like answers "is this dividend structurally safe through realistic downside?"
For income investors trying to maximize current yield while accepting some cut risk, the SWS framing is useful. For total-return value investors who can't absorb a 30% drawdown from a dividend cut, our framing is more conservative and harder to fool.
See /vs/simply-wall-st for the full methodology comparison.
The 3 warning signs of a coming dividend cut
In addition to the metric thresholds above, three behavioral warning signs precede most dividend cuts:
1. The CFO becomes elusive about the dividend on earnings calls
A confident CFO will reaffirm the dividend explicitly when asked, even in tough quarters. A CFO who hedges ("we'll continue to evaluate our capital allocation") is laying groundwork. Three quarters of hedging language is usually followed by a cut.
Reference: our guide to earnings transcript analysis walks through the specific tonal patterns to watch for.
2. Debt refinancing at materially higher rates
When the company has to refinance debt at 200+ bps higher than its existing weighted-average cost, interest expense surges. If the leverage was already 3×+, the new interest expense often pushes recurring coverage below 1.0×. This is a math problem; the dividend cut becomes arithmetic.
3. Asset sales to fund the dividend
When a company sells profitable subsidiaries or product lines and uses the proceeds to maintain the dividend, you're in the late stages of a story that's about to end. The dividend is funded by liquidation. Look for "strategic portfolio review" language in 10-Ks.
A 5-minute dividend safety check on any ticker
Combining everything above, here's the workflow:
- Pull the FCF payout ratio: dividends paid TTM / free cash flow TTM. Should be < 80%.
- Pull net debt to EBITDA: total debt minus cash, divided by EBITDA TTM. Should be < 3× for cyclical industries, < 4× for utilities/staples.
- Check dividend history: how many consecutive years of increases? Any cuts in the last 10 years?
- Eyeball the business model: is this consumer staples / healthcare / utilities (stable), or cyclical?
- Read the most recent earnings call Q&A: how is management talking about capital allocation?
If 1-4 all check out and 5 shows confident, specific dividend commentary - the dividend is in the A-B range. If 1 or 2 is failing, drop to C-D. If 3 shows a recent cut, drop further.
For a precomputed grade with the underlying metric scores visible, use invest-like's dividend safety on any stock page.
What to do with the grade
The grade is an input to your decision, not the decision itself. A few useful applications:
- Filter your income portfolio to A and B grades only. This is the highest-leverage rule for income investors. C and D grades have a meaningful probability of cuts within 5 years.
- For total-return investors, the grade matters less directly - you're not relying on the income. But the underlying signal (recurring cash coverage, leverage health) is the same signal you want for any value-investing position.
- In a rising-rate environment, dividend safety gets harder to maintain because rolling debt at higher costs eats into recurring coverage. Re-check the grade quarterly for any portfolio name where leverage is > 2×.
The most expensive single mistake an income-focused investor can make is paying attention to yield and ignoring safety. The five-metric framework above flips that priority - safety first, yield is whatever it is.