The Gordon growth model says a share of stock is worth next year's expected dividend divided by the gap between your required rate of return and the dividend's constant growth rate. A stock that just paid $2.36 per share, growing 4.2 percent a year forever against a 9.1 percent required return, is worth $50.19 — the worked example below shows every step. It is the constant-growth special case of the dividend discount model: the version you can solve with one division instead of an infinite sum.
You will meet the model in three places: intro finance courses valuing a mature dividend payer, the CFA curriculum, and the terminal-value step of a discounted cash flow. It asks for only three numbers — next year's dividend, your required return, and a growth rate — but it is unforgiving about which dividend and which growth rate you feed it.
P0 = D1 / (r − g)
P0 — Value of the stock today, per share
D1 — Dividend expected one year from now: D1 = D0 × (1 + g)
D0 — The most recent dividend, the one the company just paid
r — Required rate of return on the stock (cost of equity), as a decimal
g — Constant annual dividend growth rate, forever, as a decimal — must be less than r
D1 — next year's dividend. The formula wants the dividend the stock will pay one year from now, not the one it just paid. Problem sets usually hand you D0, the most recent dividend, so your first move is almost always to grow it one period: multiply the last dividend by one plus the growth rate. This is the single most common trap in the model. If a question says the company "just paid" $2.36, that is D0 and needs growing; if it says the company "will pay" $2.46 next year, that is already D1.
r — required rate of return. The annual return investors demand for holding this stock, also called the cost of equity. In coursework it is either given or estimated with CAPM: the risk-free rate plus the stock's beta times the equity risk premium. In the worked example below, a 4.3 percent risk-free rate, a beta of 0.80, and a 6.0 percent equity risk premium give an r of 9.1 percent. The model only works when r is larger than g — more on that below.
g — constant growth rate. The rate at which the dividend grows every single year, forever. Analysts estimate it from dividend history, from forecasts, or as the sustainable growth rate: the fraction of earnings the company retains times its return on equity. Because g lasts forever, it has to be modest. No company can permanently outgrow the economy, so long-run nominal GDP growth — roughly 4 to 5 percent in the US — is a sensible ceiling.
| Step | Calculation | Result |
|---|---|---|
| Last dividend paid (D0) | given in the problem | $2.36 |
| Growth rate (g) | given: constant 4.2% per year | 4.2% |
| Next year's dividend (D1) | $2.36 × 1.042 | $2.4591 |
| Required return (r) | CAPM: 4.3% + (0.80 × 6.0%) | 9.1% |
| Denominator (r − g) | 9.1% − 4.2% | 4.9%, or 0.049 |
| Value per share (P0) | $2.4591 ÷ 0.049 | $50.19 |
Under these assumptions the stock is worth $50.19 per share. If it trades below that, the model calls it undervalued; above, overvalued — but only relative to the g and r you chose. Before trusting the number, see how hard it leans on the growth estimate.
| Growth rate (g) | Next year's dividend (D1) | r − g | Value per share (P0) |
|---|---|---|---|
| 3.2% | $2.36 × 1.032 = $2.4355 | 5.9% | $41.28 |
| 4.2% | $2.36 × 1.042 = $2.4591 | 4.9% | $50.19 |
| 5.2% | $2.36 × 1.052 = $2.4827 | 3.9% | $63.66 |
When g is at or above r. The denominator is the gap between required return and growth. If g equals r, that gap is zero and the model divides by zero — the price comes out infinite. If g exceeds r, the price comes out negative. Neither is a valuation; both are the model telling you it does not apply. Growth that genuinely exceeds the return investors require cannot last forever, so if your inputs collide like this, the growth estimate is the thing to question.
When the company pays no dividend. With no dividend, D1 is zero and the model values the stock at zero — plainly wrong for a company that reinvests every dollar of earnings. Value non-payers with free-cash-flow models or market multiples instead.
When g outruns the economy. Because g is a forever rate, a company growing its dividend 8 percent annually in a 4 to 5 percent nominal-GDP world would eventually become the entire economy. If your estimate sits above long-run GDP growth, cap it — or switch to a two-stage dividend model that lets growth start high and decay to a sustainable rate.
The model also strains when dividends are erratic or newly initiated. "Constant growth forever" does not have to be literally true, but it has to be roughly believable — which is why the model fits mature utilities and consumer staples far better than young growth firms.
Using D0 instead of D1. Plugging the just-paid $2.36 into the numerator gives $2.36 ÷ 0.049 = $48.16 instead of $50.19 — an answer that is exactly one year of growth too low. On an exam, the wrong answer choice built from D0 is almost always sitting there waiting for you. Grow the dividend first.
Choosing a g that exceeds r. A 10 percent growth assumption against a 9.1 percent required return produces a negative denominator and a meaningless negative price. Students sometimes report the absolute value as if it were the answer; it is not. The assumption is broken, not the sign.
Mixing percentages and decimals. The denominator must be a decimal: 9.1 percent minus 4.2 percent is 0.049, not 4.9. Dividing $2.4591 by 4.9 gives $0.50, a hundredth of the right answer — a wrong result that should fail your own sanity check against the current share price.
Treating the answer as precise. $50.19 looks exact, but the sensitivity table shows a single percentage point of g moves the value by $9 to $13. Professional analysts report a range across growth assumptions; on homework, at least note which input the answer is most sensitive to.
Value the stock as next year's dividend divided by the gap between required return and constant growth — and get next year's dividend by growing the last one first. Keep g below r and below long-run economy growth, and treat the output as an estimate: one percentage point of g moves the price by roughly 18 to 27 percent.
It calculates the intrinsic value of a dividend-paying stock today, assuming the dividend grows at one constant rate forever. It answers the question: given the return I require, what is this stream of steadily growing dividends worth right now?
Three common ways: the historical growth rate of the company's dividends, analyst forecasts, or the sustainable growth rate — the retention ratio times return on equity. Whichever you use, keep it below long-run economy growth, since no company can grow faster than the economy forever.
The denominator becomes zero or negative, so the model returns an infinite or negative price. That is not a valuation — it is a signal the constant-growth assumption fails. Use a multi-stage dividend model that lets growth start high and settle to a sustainable rate.
It is the simplest special case of it. The general dividend discount model discounts each future dividend separately and lets them change year by year; the Gordon growth model collapses that infinite sum into one division by assuming a single constant growth rate forever.
Not directly. With no dividend the numerator is zero and the model prices the stock at zero, which is wrong for companies that reinvest their earnings. Value non-dividend payers with free-cash-flow models or comparable-company multiples instead.
After forecasting cash flows explicitly for 5 to 10 years, analysts apply the same constant-growth logic to everything beyond the forecast: grow the final-year cash flow one period, then divide by the discount rate minus a modest terminal growth rate. That terminal value is then discounted back to today.
By the FinanceBrain Team · Last verified July 10, 2026 · How we produce and verify articles