What's a "good" dividend growth rate?

It depends entirely on sector and maturity. For mature US dividend-growth stocks — the Dividend Aristocrats and Kings universe, plus dividend-growth ETFs like SCHD, VIG, and DGRO — 6% to 10% trailing 5-year DGR is the typical band. Consumer staples (PG, KO, PEP) usually sit at the lower end, around 4% to 7%, because their underlying earnings grow slowly but reliably. Information technology dividend payers (MSFT, AVGO, TXN) often print double-digit DGR because their earnings base expanded faster over the last decade. Regulated utilities (SO, DUK, NEE) cluster around 2% to 5% — the regulator effectively caps earnings growth, which caps dividend growth. REITs trail equities by structure: a 0% to 4% DGR is normal for a healthy equity REIT, because most cash flow already exits as the current distribution. A 15%+ DGR from a small-cap that initiated three years ago is real but not extrapolatable — it reflects starting from a tiny base, not a sustainable trajectory.

Why is the DGR I compute different from what the company reports in its investor materials?

Three common reasons. First, fiscal-year versus calendar-year mismatch — PG's fiscal year ends in June, so the "FY2024 dividend" the company reports includes payments from July 2023 through June 2024, which won't match a calendar-year sum from Yahoo Finance. Second, announced-rate timing — when a company announces "our 67th consecutive annual increase," it usually counts from the declaration date or the first paid increase, which can fall in a different calendar year than where the cash actually landed in your account. Third, ex-date versus pay-date — some data vendors group dividends by ex-date (when the price drops), others by pay-date (when cash arrives), and at year-end boundaries these can fall in different years. None of these is a calculation error. For consistency, pick one convention (calendar-year, ex-date) and stick with it across every ticker you compare. The absolute numbers will differ slightly from corporate IR pages; the relative ordering of DGRs across stocks will not.

Does DGR account for share buybacks?

No. DGR is strictly the growth rate of dividends-per-share, and dividends-per-share already incorporates the share count — when a company buys back stock, the same total cash dividend is divided among fewer shares, mechanically lifting DPS. So DGR captures the per-share effect of buybacks indirectly, but it does not measure the buyback itself as cash returned to shareholders. The metric that combines both is "total shareholder yield" — (dividends paid + net buybacks) ÷ market cap — which is a separate ratio computed at the company level, not per share. For a company like AAPL that returns far more cash via buybacks than dividends, DGR alone understates the income story by a wide margin. For a pure dividend grower like PG, which buys back relatively little, DGR is a near-complete picture. When you screen on DGR alone, you bias toward dividend-heavy capital-return policies and against buyback-heavy ones — a deliberate choice, but worth being conscious of.

How does DGR interact with the dividend payout ratio?

The payout ratio (dividends ÷ earnings) sets a ceiling on how long the current DGR can persist. A company paying out 40% of earnings has room to grow its dividend faster than earnings for several years, lifting the ratio toward 60% or 70%, before the math breaks. A company already paying out 90% can only grow its dividend as fast as earnings grow — there's no payout-ratio runway left. This is why the highest-yielding stocks usually have the lowest DGRs: a 6% yield often implies a 70%+ payout ratio, which means the next raise has to come from earnings expansion, not from raising the payout share. Conversely, a 1.5%-yielder with a 25% payout ratio (think the early years of MSFT's dividend) can compound DGR at 10%+ for a decade by lifting payout and growing earnings simultaneously. Always read DGR and payout ratio together — they tell you whether the historical growth was earnings-driven (sustainable) or payout-driven (one-time runway).

Why doesn't DGR work for REITs, option-income ETFs, or Treasury ETFs?

DGR assumes the dividend is funded by earnings that grow with the underlying business. Three asset classes break that assumption. REITs are required to distribute 90%+ of taxable income, so there's no retained earnings to fund accelerating growth — their dividend tracks rental cash flow, which grows slowly. Option-income ETFs (JEPI, MSTY, QYLD) pay out option premium that scales with volatility, not earnings; their distributions can swing 40% month-over-month with no underlying business change, so a CAGR over those distributions is statistical noise, not a growth rate. Treasury ETFs (SGOV, BIL, TLT) pass through Fed-set interest, which has no growth trajectory — only a rate cycle. For these instruments, compounding DGR forward produces dangerously misleading projections, which is why divcalc shows N/A on the DGR field for SGOV and similar rate-sensitive funds. See [Why DGR doesn't apply to T-bill ETFs](/guide/dgr-rate-sensitive-instruments) for the long-form explanation.

How to Calculate Dividend Growth Rate

By divcalc Editorial · Last reviewed May 26, 2026 · Methodology

Dividend growth rate is the second number every long-horizon income investor learns, after yield. Yield tells you what you collect today; DGR tells you what that payment becomes in twenty years. For Path A investors compounding into retirement, DGR is often the more important of the two — a starting yield of 2% compounding at 8% per year ends up paying more cash, on the original cost basis, than a 5% yield compounding at 2%. The arithmetic is one CAGR formula. The judgment is which window to measure, when DGR is meaningful, and when the historical number is an artifact rather than a forecast.

What DGR measures, and why Path A investors must understand it

Dividend growth rate is the compound annual growth rate of a company's dividend per share, measured over a defined window — typically three, five, or ten years. It is not the growth rate of total dividends paid, or of yield, or of earnings. It is strictly the rate at which the cash payment per share has compounded.

For a traditional dividend-growth investor — the Path A audience that screens on Dividend Aristocrats, holds SCHD as a core position, and projects income twenty or twenty-five years forward — DGR is the engine of yield-on-cost. Yield-on-cost is the dividend you receive divided by the price you originally paid, not today's price. A position bought at $50 yielding 3% with 8% DGR pays 6.5% on cost after ten years and 14% on cost after twenty. The starting yield is the seed; DGR is the compounding rate. If you do not measure DGR correctly, every long-horizon projection compounds the error.

The shorthand most retail screens use — "5-year DGR" without a methodology footnote — usually means the trailing 5-year CAGR of dividends per share. That is the right default. But the same input data can be measured several different ways, and the differences matter when comparing tickers across vendors.

Two algorithms: trailing CAGR vs year-over-year

There are two common ways to compute DGR, and they answer different questions.

Trailing CAGR smooths the entire window into a single annualized rate. For a 5-year window, it asks: at what constant rate would the dividend have had to grow each year to get from the year-0 value to the year-5 value? This is the right metric for projections, because projections assume constant compounding. It is the wrong metric for spotting trend breaks, because a single steep raise and four flat years produces the same CAGR as five identical raises.

Year-over-year (Y/Y) point estimates compute one growth rate per year and report them as a series — "8%, 9%, 7%, 11%, 6%". This is the right metric for diagnosing the trend: is the company accelerating, decelerating, or stable? Did one year have an outlier raise? It is the wrong metric for projection, because you still have to summarize five numbers into one to compound forward.

For mature dividend payers — PG, KO, JNJ, the Aristocrats universe — trailing 5-year CAGR is the standard. For early-stage payers (a company that initiated three years ago, or an ETF launched in 2020) Y/Y is more honest, because a CAGR over a too-short window with a non-stationary base produces misleading triple-digit numbers. The divcalc engine defaults to trailing 5-year CAGR for projections and exposes Y/Y in the chart layer for diagnosis.

The 5-year CAGR formula in detail

The formula is the standard compound annual growth rate:

DGR = (DPS_end / DPS_start)^(1/n) − 1

Where DPS_end is the dividend per share in the period-end year, DPS_start is the dividend per share in the period-start year, and n is the number of years between them.

The trap is n. For a 5-year DGR you want to measure five years of growth, which requires six annual data points: the year-0 value (the starting period-end) and the year-5 value (the ending period-end). If you take the dividends from years 2021, 2022, 2023, 2024, 2025, and divide the 2025 sum by the 2021 sum, you have measured four years of growth, not five — and dividing by 5 in the exponent will undercount the rate.

The convention that avoids this: always measure period-end to period-end. To compute a trailing 5-year DGR as of calendar-year 2025, use the 2020 calendar-year DPS as the starting value, the 2025 calendar-year DPS as the ending value, and n = 5. The starting year is included as the baseline, not as one of the growth years. Most data vendors get this right, but cross-check by computing the same DGR two ways and confirming they match. A 1% discrepancy is usually rounding; a 20% discrepancy is usually an off-by-one error in n.

Worked example 1 — SCHD, the canonical dividend-growth ETF

SCHD tracks the Dow Jones US Dividend 100 Index, a screened universe of high-quality US dividend payers that pass a fundamental quality filter. It is the default core position for Path A dividend-growth investors and the most common reference ticker for what "good DGR" looks like in a diversified ETF wrapper.

Recent share price $32.50 with starting yield 3.25% and trailing 5-year DGR of 9.15%/yr. → Open the SCHD calculator

What that DGR means in plain terms: the SCHD distribution per share has compounded at roughly that rate annually over the past five years. Project that forward at the same rate, hold the share-price growth assumption constant, and a position held for twenty-five years sees its yield-on-cost rise from the starting figure to a multiple of it. That is the entire investment thesis for the ticker. The accuracy of the projection rests on whether the underlying screen continues to produce a portfolio whose aggregate dividend can compound at a similar pace — historically a reasonable assumption for a quality-screened US dividend basket, but not a guarantee.

Worked example 2 — Procter & Gamble (PG), 67+ years of consecutive raises

PG is a Dividend King — the small subset of companies with more than 50 years of consecutive annual dividend increases. It pays quarterly, raises every April, and has done so without interruption since 1957. The growth rate is necessarily lower than a faster-growing dividend stock, because the underlying business — consumer staples — grows in the low single digits.

Recent share price $143.56 with starting yield 2.97% and trailing 5-year DGR of 6.02%/yr. → Open the PG calculator

A low-single-digit DGR sounds disappointing relative to SCHD, but the durability of the raise is the product. PG has compounded its dividend through every recession, every macro regime, and every consumer-spending downturn since the Eisenhower administration. The investment case is not maximum DGR; it is minimum DGR variance. For a retiree whose income plan cannot tolerate a single cut, that consistency is worth a lower headline growth rate. Pair PG's DGR with its payout ratio and you can estimate how much runway the current trajectory has: a payout ratio in the low 60% range leaves room to compound for another decade even if earnings growth slows.

Worked example 3 — MSTY, when DGR doesn't apply

MSTY is the YieldMax MSTR Option Income Strategy ETF. It does not own MSTR; it owns a synthetic exposure plus short calls. The distribution is option-premium income, paid weekly, scaled to whatever the volatility surface on MSTR options happens to be selling. There are no underlying earnings, no payout ratio, no dividend policy.

Recent share price $22.29 with starting yield 96.86% paid weekly. The DGR field on the divcalc page is intentionally muted or treated separately because the metric does not measure what it measures for SCHD or PG. → Open the MSTY calculator

Why DGR breaks here: the distribution is not funded by a growing earnings base. It is funded by option premium that scales with implied volatility. When MSTR is volatile, premium is rich and distributions are large. When MSTR is calm, premium collapses and distributions shrink. Computing a CAGR over a window that includes both regimes captures volatility-cycle artifacts, not a growth trajectory you can project forward twenty years. The same problem applies to JEPI, QYLD, and the broader option-income category. See Why DGR doesn't apply to rate-sensitive instruments for the parallel argument on Treasury ETFs.

Using DGR in a 25-year projection — the yield-on-cost crossover

The single most useful application of DGR is comparing a low-yield-high-growth holding against a high-yield-low-growth holding over a long horizon. The crossover math is what separates a yield-chaser from a Path A income compounder.

Take two stylized positions, each bought for $10,000 today:

Position A: 2% starting yield, 8% DGR (a dividend-growth stock or ETF profile)
Position B: 5% starting yield, 2% DGR (a mature high-yielder or utility profile)

In year 1, Position B pays $500 in cash; Position A pays $200. B looks like the obvious income choice. But A's payment compounds at 8% per year; B's at 2%. In year 10, A pays about $432 and B pays $610 — B is still ahead. In year 15, A pays about $635 and B pays $673 — the gap has nearly closed. Somewhere around year 16 to 17, A's payment crosses B's and pulls away. By year 25, A pays roughly $1,370 and B pays about $820. On cumulative cash received, A has overtaken B somewhere around year 20.

The crossover year shifts with the starting numbers. A 3% yield and 7% DGR crosses a 6% yield and 1% DGR somewhere around year 12. The general rule: when the DGR differential exceeds the yield differential by more than about 4 percentage points, the lower-yielder wins on long-horizon cash income. Path A investors who hold for 20+ years should weight DGR more than current yield. Investors who need cash in the next 5 years should do the opposite. The growth calculator at divcalc back-solves these crossovers for arbitrary inputs.

Common pitfalls

Three errors distort DGR calculations across vendor data and DIY spreadsheets.

Including special dividends. A one-time supplemental distribution lifts the year's DPS sum without changing the underlying dividend policy. Including a special in the year-5 endpoint of your CAGR window mechanically inflates DGR. The convention — match what you do in yield calculations — is to use only regular dividends. Most data vendors strip specials by default; check the label on any DPS series before computing DGR.

Ignoring share splits. When a company splits its stock, the per-share dividend halves (for a 2-for-1) on the ex-split date. Pre-split DPS values must be split-adjusted to be comparable to post-split values. Failing to do this either erases an apparent DGR or inverts it. Most vendors publish split-adjusted historical DPS, but always check by spot-comparing a single year against a primary source.

Comparing windows that span a cut. A 5-year CAGR over a window that includes a dividend cut is technically valid but tells you nothing about the company's current trajectory. AT&T's trailing 5-year DGR computed across the 2022 WarnerMedia spin shows a deeply negative number that has no bearing on T's post-spin dividend policy. For any company that has cut, restart the DGR window at the year of the cut and measure forward from there — or use Y/Y point estimates and note the discontinuity explicitly.

Try the Dividend Growth Calculator → to project income forward with custom DGR assumptions, or revisit the basics with the Dividend Yield guide →.