The Gordon Growth Model estimates what a dividend-paying stock is worth today by treating every future dividend as a stream that grows at a constant rate forever and discounting that stream back to present value. The formula is P = D₁ / (r − g), where P is the intrinsic value per share, D₁ is next year’s expected dividend, r is your required rate of return, and g is the assumed constant growth rate of the dividend. Compare the P you calculate to the stock’s current market price, and the model tells you whether the market is pricing the dividends generously or cheaply.
The Formula
Written out: P = D₁ / (r − g).
- P is the intrinsic value of one share, in dollars.
- D₁ is the dividend per share you expect over the next twelve months.
- r is your required rate of return, acting as the discount rate.
- g is the dividend’s constant annual growth rate, assumed to hold forever.
The denominator does most of the work. When r and g are close together, the fraction gets large and the stock looks valuable, because the dividend is growing nearly as fast as the return you demand. When they are far apart, the stock looks cheaper, because the growth doesn’t compensate you enough for waiting.
One hard rule: g must be strictly less than r. If growth equals or exceeds the discount rate, the math produces an infinite value, which is the model telling you the assumptions are broken. No stock is worth infinity. When you see a nonsensical result, the growth rate is almost always too high.
Finding Each Input
The formula is simple. The inputs are where the work lives, and the output is only as good as the weakest number you feed it.
Next Year’s Dividend (D₁)
Take the most recent annual dividend per share (D₀) and multiply by (1 + g). A company that paid $3.00 last year with 5% expected growth gives you a D₁ of $3.15. The current dividend appears in the company’s annual report filed with the SEC, in the section covering equity securities, dividends, and stock repurchases.1U.S. Securities and Exchange Commission. How to Read a 10-K Quarterly filings show interim dividends if the company has recently changed its payout.
Required Rate of Return (r)
Most analysts build r from the Capital Asset Pricing Model: r = risk-free rate + (beta × equity risk premium). The risk-free rate comes from the 10-year U.S. Treasury yield, which sat near 4.34% in late April 2026.2Federal Reserve Economic Data (FRED). Market Yield on U.S. Treasury Securities at 10-Year Constant Maturity, Quoted on an Investment Basis Beta measures how much the stock moves relative to the overall market; a beta of 1.0 tracks the index, 0.7 is calmer, 1.3 is jumpier. The equity risk premium is the extra return investors demand for holding stocks instead of risk-free bonds, and the implied premium for U.S. equities at the start of 2026 was approximately 4.23%.
For a stock with a beta of 0.8: r = 4.34% + (0.8 × 4.23%) = 7.72%. A riskier stock with a higher beta pushes r up, widens the denominator, and lowers the calculated price. That’s the model penalizing you for taking on more risk.
Growth Rate (g)
Growth is the most subjective input and causes the most damage when it’s off. Two common approaches:
The sustainable growth method uses the company’s own books. Multiply the retention ratio by the return on equity. The retention ratio is the share of net income the company keeps rather than paying out, calculated as (net income − dividends) / net income from the income statement. A company that retains 40% of earnings and earns 15% on equity has a sustainable growth rate of 6%.
The analyst estimate method uses consensus figures from equity research reports, which incorporate industry conditions and competitive dynamics that a backward-looking formula misses.
Either way, cap g at the long-term nominal GDP growth rate, roughly 4% to 5% for the U.S. economy. No company can grow its dividends faster than the whole economy forever without eventually becoming the whole economy. If your model is producing an eye-watering valuation, the growth rate is usually why.
A Worked Example
You’re looking at a utility that paid $2.00 per share in dividends last year. You estimate dividends will grow at 4% annually, and your CAPM-derived required return is 10%.
Compute next year’s dividend: D₁ = $2.00 × 1.04 = $2.08.
Plug in: P = $2.08 / (0.10 − 0.04) = $2.08 / 0.06 = $34.67.
If the stock trades at $30, the model says it’s undervalued by about $4.67 per share, roughly 13%. If it trades at $42, the model flags it as overpriced relative to its dividend stream.
Now change one input. Raise g from 4% to 5%, and the valuation jumps to $42.00, a 21% increase from one percentage point. Drop g to 3%, and the value falls to $29.43. Lower r from 10% to 9% while keeping g at 4%, and the value climbs to $41.60. These aren’t rounding differences. They’re the reason every assumption deserves scrutiny.
Why Small Input Changes Produce Big Swings
The denominator (r − g) is the fulcrum. At r = 10% and g = 4%, the denominator is 0.06. Nudge g to 5% and the denominator shrinks to 0.05, a 17% reduction that inflates the valuation by the same proportion. As g approaches r, the denominator approaches zero and the calculated price runs toward infinity.
This sensitivity forces honesty. Two analysts who agree on the dividend and the discount rate but differ by one percentage point on growth can produce valuations 20% apart. That’s why experienced practitioners rarely report a single intrinsic value. They run the model across a range of growth and discount rates and produce a valuation band.
Do the same thing yourself. Run the model three times: once with your best estimate of g, once pessimistic, once optimistic. If all three scenarios show the stock as undervalued, the signal is far more credible than a single calculation that only works under the rosiest assumption.
Comparing the Result to Market Price
Once you have P, set it next to the current trading price. If P is higher than the market price, the model considers the stock undervalued. If P is lower, it’s overvalued relative to its dividend stream.
Value investors don’t buy the moment a model shows any undervaluation. They want a margin of safety, a buffer against errors in the inputs. For a stable blue-chip with a rock-solid dividend history, a 10% discount to intrinsic value might be enough. For a company whose growth estimate is shakier, the cushion needs to be larger, sometimes 30% to 50%. There is no universal correct figure; the point is to acknowledge that your inputs are estimates and leave room to be wrong.
Treat the comparison as a screening tool, not a verdict. A model saying a stock is 25% undervalued is a reason to dig into the balance sheet, the earnings calls, and whether the dividend is actually sustainable. It is not a reason to buy on the number alone. The model tells you what the math says the dividends are worth. Whether the company will actually pay them is a separate question.
Where the Model Fits
The model works for mature, financially stable companies with long records of paying and growing dividends at a steady rate. Point it at a fast-growing company that reinvests every dollar of profit and you get nothing useful, because there are no dividends to discount.
Regulated utilities are the textbook case. Revenue is constrained by service territories and regulatory approval, so growth tends to track population and economic expansion. Large consumer staples companies with decades of consecutive dividend increases fit the same mold.
REITs are also natural candidates because federal tax law requires them to distribute at least 90% of their taxable income as dividends to maintain their tax-advantaged status.3Office of the Law Revision Counsel. 26 USC 857 – Taxation of Real Estate Investment Trusts and Their Beneficiaries The mandatory payout creates the kind of predictable stream the model needs. For REITs, the growth rate should reflect property income growth and rent escalation rather than broader earnings metrics.
Preferred stock is the cleanest application of all. It pays a fixed dividend that never changes, which is the Gordon Growth Model with g = 0. The formula collapses to P = D / r. A preferred share paying $7.50 annually with a required return of 8% is worth $7.50 / 0.08 = $93.75. The biggest source of estimation error, the growth rate, drops out entirely.
When the Model Breaks Down
- Non-dividend payers. If a company doesn’t distribute cash, there is nothing to discount, and the model doesn’t apply.
- Cyclical businesses. Commodities, construction, and automotive earnings swing with the economic cycle, and a constant-growth assumption falls apart when the dividend could be cut in the next downturn.
- High-growth companies. A firm growing dividends at 15% today won’t sustain that pace forever, and the formula assumes g holds forever. Plugging in a high growth rate produces dangerously inflated valuations.
- Companies with unstable payout ratios. If dividend policy swings year to year between generous payouts and heavy retention, the constant-growth assumption isn’t reliable.
When a company has a period of fast growth that will eventually settle down, analysts move to a multi-stage dividend discount model that splits the future into a high-growth phase and a stable-growth phase. The Gordon Growth Model usually appears inside those multi-stage models as the terminal-value calculation at the point where growth stabilizes.
Pre-Tax Versus After-Tax
The model calculates intrinsic value from pre-tax dividends, but what you keep depends on tax treatment. Qualified dividends, which include most regular dividends from domestic corporations held for a minimum period, are taxed at 0%, 15%, or 20% depending on taxable income.4Internal Revenue Service. Revenue Procedure 2025-32 For 2026, single filers pay 0% on qualified dividends up to $49,450 in taxable income, 15% up to $545,500, and 20% above that. Joint filers reach the 15% bracket at $98,900 and the 20% bracket at $613,700.
High-income investors face an additional 3.8% net investment income tax when modified adjusted gross income exceeds $200,000 for single filers or $250,000 for joint filers.5Office of the Law Revision Counsel. 26 USC 1411 – Imposition of Tax That pushes the effective top rate on qualified dividends to 23.8%. If you’re comparing two dividend stocks held in the same account, pre-tax valuation is fine. If you’re comparing a dividend stock to a tax-free alternative like a municipal bond, adjust r to reflect after-tax yields so the comparison is apples to apples.