What Dividend Growth Rate Should You Assume?

Every dividend projection rests on one number that nobody knows: the rate at which the dividend will grow. It is not a detail. Change it and the answer changes by more than half.

And it is where projections go wrong most often — not through arithmetic errors, but through an assumption that was never examined.

Take $100,000 in an ETF yielding 3.2%. It pays about $3,200 this year.

Now ask what it pays in ten years.

Same fund, same starting point, and the answer ranges from $4,300 to $8,300. The growth assumption moves the result almost twice as much as the starting yield does.

Assumptions of 10% or higher circulate widely, and they are not invented. Several large dividend ETFs have genuinely grown their distributions at roughly that rate over specific historical stretches, and some have sustained it for a decade.

But when these figures get repeated, two things tend to get lost: the period they were measured over, and the fact that the period was unusually favourable.

A widely-read argument on this played out publicly. An investor asked what $280,000 in a dividend ETF would pay over ten years. The projection returned assumed 10% growth and arrived at $24,500 a year by year ten. The response was blunt — the assumption was called wildly optimistic, and a scenario using 5% produced $18,000 instead.

That is a 36% difference in outcome, driven entirely by one unexamined input.

Rather than argue about the right number, it is more useful to look at what the largest dividend ETFs have actually paid, measured across complete calendar years.

These are the median year-over-year change across every complete year with data — not a start-to-end average, because a single one-off distribution can distort those badly.

Three observations from that list. First, the spread is enormous: from -10% to +10.9%. Second, the highest figures cluster closer to 8% than to 10%. Third, not every fund grows at all.

RDVY is the instructive case. It is marketed as a dividend growth fund, and over this window its distributions declined. It pays ordinary dividends plus capital gains distributions that vary wildly year to year, so its headline growth number is genuinely negative. That is not a data problem; it is what the fund did.

Even a historically accurate growth rate is a statement about the past. Three reasons it may not persist.

The measurement window matters. Almost all available dividend history starts in 2020. A window that begins in a pandemic, runs through a strong equity market and high inflation, and ends today is not a neutral sample. A different six-year stretch would give different answers.

Payout ratios have limits. A company can only raise its dividend as fast as its earnings grow, unless it increases the share of earnings it pays out — and that cannot rise indefinitely.

Index composition changes. The funds rebalance. A fund that grew its dividend at 9% under one set of constituents is not guaranteed to do so under the next.

The temptation is to pick the growth rate that produces the number you want. The useful approach is the opposite: run several and look at the shape of the outcome.

If your plan only works at 10%, it does not work — it depends on a number at the top of the historical range persisting for decades. If it works at 5%, you have something that survives a middling outcome, and the better outcome becomes upside.

The Dividend ETF Calculatoris built around this. It shows every projection at 3%, 5%, 8% and 10% side by side, marks where the fund’s own measured history falls, and does not pick a number for you.

Dividend growth is not the only variable, and it is not usually the most important one. A projection that assumes the share price never moves is a model of the dividend, not of your return.

Total return — price change plus distributions — is what you actually earn. A fund with a 3% yield and 6% price growth beats a fund with a 6% yield and no price growth, even though the second pays more cash. Income is a way of receiving a return, not a substitute for one.