A half-century streak that screens like a disaster
Fortis is about as close to dividend royalty as Canada produces: a regulated utility serving millions of customers across North America, with a streak of consecutive annual dividend increases now running past fifty years. Through oil shocks, double-digit interest rates, the financial crisis, and a pandemic, the raise arrived every year. If dividend investing has a hall of fame, Fortis is in the first room.
Now run it through a standard dividend-safety screener. Free cash flow: negative, year after year — the company routinely spends more on capital projects than its operations bring in. Debt: heavy, a balance sheet that would look alarming on a retailer or a miner. Verdict, by the standard rules: this company cannot afford its dividend, let alone raise it.
The screener says danger. Fifty years of raises say otherwise. When a tool and half a century of evidence disagree this completely, the interesting question isn't who's right — the evidence settled that — it's what the tool is measuring instead of the thing you asked about. And the answer generalizes far past Fortis, because the same wrong lens misreads nearly every regulated utility and pipeline on the TSX, in both directions.
Why the free-cash-flow rule exists — and where it breaks
The FCF test is a good rule in its home territory. For an ordinary company, free cash flow — operating cash minus capital spending — approximates the money genuinely available to shareholders. If dividends exceed it persistently, the payout is being financed by borrowing or dilution, and that arithmetic has a deadline. The rule catches real trouble at real companies constantly.
The rule's hidden assumption is that capex is a cost of staying in business — maintaining the stores, replacing the trucks. Money that must be spent before anyone can honestly call the rest "free." That assumption fails at a regulated utility, because a utility's capital spending is something categorically different: it's the growth engine, running through a mechanism most investors never have reason to learn.
Here's the machine in one paragraph. A regulated utility earns a return set by its regulator — a sanctioned percentage on its rate base, essentially the invested capital serving customers: the wires, pipes, substations. The way a utility grows earnings is by growing that rate base, which means building things — and every approved dollar of construction becomes a dollar the utility is entitled to earn a return on, recovered through customer rates for decades. So when Fortis pours billions a year into grid projects, it isn't leaking cash to survive. It's buying regulator-guaranteed future earnings, on purpose, with the regulator's sign-off on recovery.
That inverts the meaning of the screener's key input. Negative free cash flow at a retailer means the business consumes more than it makes. Negative free cash flow at Fortis means management found more approved projects to fund — which is, precisely, the source of the next decade of dividend raises. Same number. Opposite verdicts.
Then what actually funds the dividend?
Fair question, because "the capex is good, actually" doesn't by itself pay anyone.
The dividend is funded from operating earnings — the regulated returns already flowing from the existing rate base. On that measure, the one that actually corresponds to the machine, Fortis's payout has long sat in comfortable territory as a share of earnings, with room to spare. The growth capex is deliberately funded the other way: with debt and periodic equity issuance, because each borrowed dollar goes into an asset earning a sanctioned return above its cost. That's also why the debt load is heavy by design — a utility's balance sheet is structured around the regulator's approved capital mix, and leverage against government-sanctioned cash flows is a different animal than leverage against fashion retail sales.
The honest checks for a utility's dividend are therefore different ones: the payout ratio against earnings and its direction; the regulatory environment (allowed returns get reviewed, and they can go down); interest-rate exposure, since debt-heavy models feel every refinancing cycle; and whether management's dividend-growth guidance is backed by an approved capital plan rather than hope. Fortis publishes multi-year guidance for mid-single-digit annual increases tied to exactly such a plan — a discipline available to utilities because the earnings are regulated, and one almost no ordinary company could responsibly offer. None of this makes the stock riskless. Rising rates genuinely hurt, regulators genuinely tighten, and a streak is never by itself proof of safety. But those are the real risks, visible through the right lens. "Can't afford the dividend" isn't on the list.
The pattern, because this keeps happening
If this argument feels familiar, it's because we made its twin about a different sector: REIT payout ratios look insane until you swap accounting earnings for FFO, at which point they turn ordinary. Same disease, different symptom — a universal formula fed a sector where its key input means something else entirely. Screeners misread REITs because depreciation and fair-value swings poison the earnings line; they misread utilities because growth capex poisons the free-cash-flow line. Banks and insurers break the standard toolkit in their own ways too — FCF is close to undefined for a lender.
The general lesson is worth keeping even if you never buy a utility: a metric computed through the wrong sector lens isn't a rough approximation — it's confident noise. A wrong-lens number will fail a Fortis and pass a genuinely fragile payer with equal conviction, and the failure carries no warning label. The number renders. It has decimals. It's simply about something else.
Where our own engine had to learn this
House policy is to say this part plainly. When DivSight first expanded from funds into individual dividend stocks, our scoring applied the standard playbook — free-cash-flow coverage, conventional debt metrics — across the board. Utilities and pipelines promptly scored the way this post opens: multi-decade raisers graded as if their payouts were in danger, because our engine was reading rate-base investment as cash bleed. The fix was sector-aware scoring: recognizing what kind of business a company is and grading it against the metrics that carry meaning there, versioned like every methodology change we make. The subtle discovery along the way: a wrong-lens score is worse than no score, because confidently wrong beats visibly missing at misleading people. We'd rather show a neutral gap than a false alarm.
That's the whole moral, and it scales past screeners to every tool in this business, ours included: before trusting any number about a dividend, ask whether the number was computed for this kind of company. Fortis passed the only test that matters for fifty consecutive years while failing the test most tools actually run. You can see how we grade it now — the utility-aware version — and the rest of our coverage, where the same question gets asked of every sector before any score gets published. The streak deserved a tool that could see it.