Introduction
Every valuation model is an argument about what a share is worth, and the dividend discount model makes the bluntest argument of the lot: a share is worth the cash it hands you. Not the cash the business generates, not the cash left over in some hypothetical unlevered world, but the dividends that actually reach a shareholder's account.
That bluntness has a consequence interviewers care about. A DCF values the whole firm and then makes you bridge to equity by subtracting net debt. A DDM skips the bridge and lands on equity value in one step, because dividends are already a shareholder-only cash flow and the cost of equity is a shareholder-only discount rate.
For most of the US market that makes the model a teaching device rather than a working tool. For banks, insurers and regulated utilities it is the primary model, because free cash flow as you learned to build it does not describe those businesses. What follows is the Gordon growth form, the multi-stage variants, where the inputs come from, a worked example, and the traps built on top of it.
What the Dividend Discount Model Values
Start from first principles. If you buy a share and never sell it, the only cash you receive is dividends. Every other reason to own it, including the hope of selling higher, rests on the next buyer's claim on that same dividend stream.
Here is value per share today, the dividend in year , and the cost of equity. Nothing in that expression references debt, EBITDA or enterprise value, which is the whole point.
Equity Value Without the Bridge
A DCF discounts unlevered free cash flow at WACC, reaches enterprise value, then subtracts net debt and other claims. Two of those steps are chances to make a mistake, which is why our guide to the enterprise value to equity value bridge exists.
The DDM has no bridge. Dividends are paid after interest, after tax, after capital expenditure and after any capital the business must retain, so they are already residual cash flows belonging to shareholders alone. Discounting them at WACC blends in cheap after-tax debt the shareholder never sees.
- Dividend Discount Model
A model that values a company's equity as the present value of all the dividends it is expected to pay, discounted at the cost of equity rather than at WACC. Because dividends belong only to common shareholders, it produces equity value per share directly, with no enterprise value step and no net debt subtraction.
How It Compares to a DCF and to Comps
Interviewers ask where the DDM sits against the two methods you already know, so hold all three on a grid.
| Dimension | Dividend discount model | Discounted cash flow | Comparable companies |
|---|---|---|---|
| What it values | Equity, directly | Enterprise value, then bridge | Either, by multiple |
| Cash flow discounted | Dividends | Unlevered free cash flow | None |
| Discount rate | Cost of equity | WACC | Inside the multiple |
| Core inputs | Next dividend, cost of equity, growth | Forecast, WACC, terminal value | Peer set, metric |
| Best suited to | Banks, insurers, utilities | Most operating companies | Sectors with clean peers |
| Main weakness | Useless without dividends | Terminal value dominates | Inherits sentiment |
The first row carries the answer: the DDM produces equity value directly, while a DCF produces a firm-level number you still have to bridge.
The Gordon Growth Form and Its Assumptions
The Gordon growth model collapses the infinite sum into one line by assuming dividends grow at a constant rate forever, which turns the series into a geometric progression with a closed-form solution.
The numerator is next year's dividend, not the one just paid. If a company paid $2.00 and grows at 3%, the numerator is $2.06. The same algebra powers the Gordon growth terminal value at the back of a DCF, with unlevered cash flow and WACC swapped in.
The Assumptions Behind the One Line
The formula is short but it commits you to three claims, and saying them out loud is what a strong answer sounds like.
- Constant growth forever. Dividends grow at exactly to infinity, which only describes a business that has finished growing.
- Growth strictly below the discount rate. If reaches the denominator hits zero and the model returns infinity, an input error rather than a valuation.
- A payout the business can fund. The dividend has to come from earnings actually produced, not from borrowing or running down capital.
The first condition is why the Gordon form is rarely applied to a whole company on its own. It is a terminal value engine bolted onto an explicit forecast.
Two-Stage and Multi-Stage Models
Real companies pass through phases. A utility midway through a capital programme grows its dividend faster than it later will; a bank rebuilding its payout takes years to settle. Multi-stage models handle that shape, and the two-stage build is the workhorse.
The first term is the present value of the explicit dividends. The second is the terminal value, built with the Gordon formula off the final explicit dividend and then discounted back years.
Building a Two-Stage Model
Forgetting that final discounting step is the most common arithmetic error in a two-stage build, and it inflates the answer badly because the terminal component carries most of the weight.
Forecast the explicit dividends
Project earnings per share and a payout ratio for each year of the high-growth phase and multiply them
Discount each one
Divide each year's dividend by one plus the cost of equity raised to that year's power
Build the terminal value
Grow the final explicit dividend by the perpetual rate and divide by the cost of equity minus that rate
Discount it back
Divide the terminal value by one plus the cost of equity raised to the power of the final year, then add the two pieces
Three-stage models insert a transition phase where growth declines gradually rather than in one step, and the H-model does the same in closed form with a linear decline. On a live deal, two stages plus an honest sensitivity table beats three stages plus false precision.
Where the Inputs Come From
A DDM has three inputs and every one is a judgment call. Naming the source of each is most of what separates a real answer from a memorized formula.
The Dividend and the Discount Rate
The cleanest source for the dividend is company guidance. Utilities and large banks commit to growth ranges over multi-year horizons, and boards signal changes early. Where guidance is absent, build it as earnings per share multiplied by a target payout ratio, which forces a view on policy rather than an extrapolated line.
The discount rate is the cost of equity, normally CAPM, and our walkthrough of the cost of equity and CAPM covers each input. Two points are DDM-specific: use the cost of equity and never WACC, and note that the tenor of the risk-free rate moves the answer. As of early September 2026 the Federal Reserve's H.15 release put the 10-year Treasury yield at 4.77% and the 20-year and 30-year at 5.25%, so the long bond adds half a point to .
Sustainable Growth as ROE Times Retention
Growth cannot simply be asserted. The consistent route is the sustainable growth rate: a company grows dividends over the long run at the rate it grows equity, and equity grows through retained earnings.
A company earning a 10% return on equity and paying out 70% retains 30%, so sustainable growth is 10% multiplied by 0.30, or 3%. Assume faster growth and you owe an explanation, normally rising returns or new equity. The return figure itself is unpacked in our guide to ROE and DuPont analysis.
- Sustainable Growth Rate
The pace at which a company can grow earnings and dividends indefinitely using only retained earnings, calculated as return on equity multiplied by the retention ratio (one minus the payout ratio). It ties the growth assumption back to profitability and payout policy, so a business distributing most of its earnings cannot be assumed to grow quickly.
Valuation questions decide more technical interviews than any other category: Work through the dividend discount model, DCF mechanics and cost of capital with full worked answers, start practicing interview questions for free and find the gaps before an interviewer does.
A Worked Two-Stage Example
Take a regulated US electric utility that paid $2.00 per share last year. Its capital programme supports dividend growth of 6% for five years, after which growth settles at 3%, consistent with a 10% allowed return on equity and a 70% payout. Its cost of equity is 9%.
The Explicit Period and the Answer
Grow the dividend at 6% for five years and discount each at 9%. The final year does double duty, since it also seeds the terminal value.
| Year | Dividend | Discount factor | Present value |
|---|---|---|---|
| 1 | 2.12 | 1.0900 | 1.94 |
| 2 | 2.25 | 1.1881 | 1.89 |
| 3 | 2.38 | 1.2950 | 1.84 |
| 4 | 2.52 | 1.4116 | 1.79 |
| 5 | 2.68 | 1.5386 | 1.74 |
Those five present values sum to $9.20. Year five's dividend is 2.00 multiplied by 1.06 to the fifth power, or $2.6765, and its discount factor is 1.09 to the fifth, or 1.5386.
Now the terminal value. Grow the year-five dividend by 3% to get 2.6765 multiplied by 1.03, or $2.7567, and divide by the spread of 0.06: the value at the end of year five is $45.95. Divide by 1.5386 to discount it back and it is $29.86 today. Add the pieces: 9.20 plus 29.86 gives $39.07 per share, of which terminal value is 29.86 divided by 39.07, or 76%.
What the Sensitivity Table Shows
No DDM output should be presented as a single number. Flexing the two assumptions that matter gives the range the answer lives in.
| Cost of equity | Growth 2.5% | Growth 3.0% | Growth 3.5% |
|---|---|---|---|
| 8.5% | 39.74 | 42.66 | 46.17 |
| 9.0% | 36.63 | 39.07 | 41.94 |
| 9.5% | 33.98 | 36.02 | 38.41 |
The corners carry the message. A 9.5% cost of equity against 2.5% growth gives $33.98; 8.5% against 3.5% gives $46.17, roughly 36% higher. Neither is unreasonable, and the gap comes from a one-point range on each of two unverifiable inputs.
When the DDM Beats a DCF, and When It Fails
The DDM is not a weaker DCF. It is the right model for one class of business and the wrong one for most others, and knowing which side a company falls on is the test.
Banks, Insurers and Regulated Utilities
For a bank, debt is raw material rather than financing, so there is no unlevered free cash flow to discount and no enterprise value to compute. What a bank can distribute is capped by regulatory capital: it holds a cushion against risk-weighted assets and only the surplus can be paid out. The DDM captures that, and our guide to valuing a bank covers the equity multiples alongside it. Insurers work the same way, with reserves as the constraint.
Regulated utilities qualify for a different reason: returns are set by regulators, payout policy is explicit, and dividends are why investors own them. Regulators even run the model in reverse. The Federal Energy Regulatory Commission sets allowed returns from a two-step discounted cash flow analysis weighted equally with CAPM, giving short-term growth 80% and long-term growth 20% inside the DCF, a weighting first adopted in Opinion No. 569-A and carried into its October 2024 order on remand.
Where the Model Breaks Down
Four situations kill a DDM outright.
- No dividend. Nothing to discount, and forecasting a first payment a decade out turns the valuation into an assumption about policy.
- Buyback-heavy distribution. A small dividend alongside a large repurchase programme leaves the raw model badly undervaluing the company.
- Growth above the discount rate. The constant-growth form cannot handle it, so you need an explicit forecast until growth falls below .
- Payout detached from capacity. Holding dividends flat while earnings collapse, or hoarding distributable cash, breaks the link the model depends on.
The Buyback Adjustment
The second of those is the common one in the US, and the fix, set out in Damodaran's chapter on modifying the model for stock buybacks, is to replace dividends with total payout, meaning dividends plus net repurchases. The scale is easy to underestimate: in the most recent full year S&P Dow Jones Indices has reported, the twelve months to September 2025, S&P 500 companies spent a record $1.020 trillion on buybacks against $664.9 billion of dividends, roughly one and a half times as much, according to S&P Dow Jones Indices.
Two cautions travel with it. Take repurchases net of share issuance, and normalize over several years because programmes are lumpy. Why companies pick one route is covered in buybacks versus dividends.
- Total Payout Model
A dividend discount model in which the cash flow discounted is dividends plus net share repurchases rather than dividends alone. It exists because many US companies return most of their cash through buybacks, which a conventional model ignores entirely.
Get the complete guide: Download our comprehensive 160-page PDF, covering valuation methodologies, DCF and DDM mechanics, and the technical questions interviewers return to every cycle.
Inverting the Model: Implied Cost of Equity and Implied Growth
Fix the market price and solve for another variable, and a valuation tool becomes a diagnostic tool. This is where the DDM earns its keep even for companies you would never value with it.
Solve for the discount rate and you get the implied cost of equity, which is the forward dividend yield plus growth.
A stock at $40 with a next-year dividend of $2.00 and consensus growth of 4% implies 2 divided by 40, which is 5%, plus 4%, or a 9% cost of equity. Rearrange for the implied growth rate the market must believe to justify today's price.
Same stock, but suppose your CAPM build says 8.5%. Implied growth is 8.5% minus the 5% yield, or 3.5%. That is a testable disagreement rather than a vague sense that something is expensive.
Interview Traps Around the Dividend Discount Model
DDM questions are rarely about the formula. They test whether you understand what it assumes, so the follow-ups cluster around a few predictable traps.
Why Not a DDM for a Technology Company
The obvious layer is that most technology companies pay nothing, so there is no cash flow to discount. The better layer is that even when they do pay, the dividend bears no relationship to distributable capacity: a company reinvesting at high returns should be retaining cash, and a token dividend says nothing about what it can afford.
Then add the arithmetic. A company growing dividends at 15% against a 10% cost of equity has a negative denominator in the constant-growth form. You would need an explicit forecast running until growth falls below the discount rate, by which point you are building a DCF.
When Growth Approaches the Discount Rate
The denominator approaches zero and value approaches infinity, which is the model reporting an impossible input rather than a priceless company. No business grows faster than the economy forever.
The working constraint is that perpetual growth should sit at or below long-run nominal GDP growth, which keeps it in the low single digits for a US company. If a model needs 7% perpetual growth to justify today's price, the price embeds a longer high-growth phase instead.
Equity Value, and the Link to P/E
Candidates who have memorized the DCF sometimes subtract net debt from a DDM output. That double-counts leverage: the model has already discounted a cash flow only shareholders receive, at a rate only shareholders demand.
The cleanest way to prove you understand this is to derive the justified P/E in one step. Divide both sides of the Gordon formula by next year's earnings, and because equals earnings multiplied by the payout ratio:
Run the utility through it: a 70% payout, a 9% cost of equity and 3% growth give 0.70 divided by 0.06, or a forward P/E of about 11.7 times. That explains why high-payout, low-growth companies trade on modest multiples, and shows the multiple and the model are one statement.
Key Takeaways
- The dividend discount model discounts expected dividends at the cost of equity, giving equity value per share with no enterprise value step and no net debt bridge.
- The Gordon growth form is next year's dividend over the cost of equity minus perpetual growth, and it assumes constant growth, growth below the discount rate, and a funded payout.
- Value is driven by the spread between r and g: half a point on either input moves the answer by roughly six to nine percent.
- Two-stage models forecast dividends through the high-growth phase, apply the Gordon form to the stable phase, and discount that terminal value back.
- Inputs come from guidance or a target payout ratio, CAPM for the discount rate, and return on equity times retention for growth.
- It is the standard model for banks, insurers and regulated utilities, where free cash flow is not constructible and payouts are capped by capital rules.
- It fails for non-payers, buyback-heavy names and high-growth companies; the fix for the second is a total payout model.
- Inverting the formula yields an implied cost of equity or growth rate, and the justified P/E equals the payout ratio over the spread.
Conclusion
The dividend discount model looks like the simplest thing in valuation and works as one of the better filters an interviewer has. Far fewer candidates than you would expect can explain why the output is equity value, why the denominator matters more than the numerator, and why a bank has no alternative while a software company cannot use it at all.
Prepare it the way you would prepare a DCF. Know the Gordon growth form and the two-stage build cold, and be able to source every input, including growth, because return on equity times the retention ratio is the answer that shows you did not pick a number.
Above all, be able to invert it. The moment you can look at a share price and say what growth rate the market is assuming, the model stops being a formula and becomes a way of arguing about what a company is worth.
