Dividend Discount Model (DDM): Formula, Types, and Examples
When to Use a DDM
The DDM works best for mature, stable companies that pay consistent and predictable dividends — think utilities, REITs, large-cap banks, and consumer staples. It’s less useful for growth companies that don’t pay dividends or companies with erratic payout policies.
Compared to a DCF model based on free cash flow, the DDM is simpler but more restrictive. The DCF values the total cash-generating ability of the firm; the DDM only values cash actually distributed to shareholders. For companies where dividends track free cash flow closely, the two approaches converge.
Types of Dividend Discount Models
| Model Type | Assumption | Best For |
|---|---|---|
| Gordon Growth Model (GGM) | Dividends grow at a constant rate forever | Mature, stable dividend payers |
| Two-Stage DDM | High growth period, then constant growth | Companies transitioning to maturity |
| Three-Stage DDM | High growth → declining growth → stable growth | Companies with clear growth lifecycle |
| H-Model | Growth declines linearly to a stable rate | Smoother transition than two-stage |
The Gordon Growth Model
The Gordon Growth Model (GGM) is the simplest form of the DDM. It assumes dividends grow at a constant rate forever. The formula is elegant and intuitive:
Where D₁ is next year’s expected dividend, r is the required rate of return (typically the cost of equity), and g is the constant dividend growth rate. The model requires that r > g — otherwise the formula produces nonsensical results.
For example, if a stock pays a $2.00 dividend next year, your required return is 10%, and dividends grow at 4% annually, the stock is worth $2.00 ÷ (0.10 − 0.04) = $33.33.
Two-Stage DDM
The two-stage model is more realistic for companies that are currently growing dividends faster than their long-term sustainable rate. You project dividends explicitly for a high-growth phase (typically 5–10 years), then apply the Gordon Growth Model for the terminal value.
The first term sums the present value of dividends during the high-growth phase. The second term is the terminal value — the Gordon Growth Model applied at the end of the explicit forecast period, discounted back to today.
H-Model
The H-Model assumes that the growth rate doesn’t jump abruptly from high to low — instead, it declines linearly over a transition period. This is often more realistic than a two-stage model because growth rarely shifts overnight.
Where gS is the short-term (high) growth rate, gL is the long-term (stable) growth rate, and H is half the length of the transition period in years.
Key Inputs and How to Estimate Them
| Input | How to Estimate | Typical Range |
|---|---|---|
| Required Return (r) | CAPM: Risk-free rate + β × Equity Risk Premium | 8–12% for most equities |
| Growth Rate (g) | Retention ratio × ROE, or historical dividend growth | 2–6% for stable companies |
| Payout Ratio | Dividends per share ÷ EPS | 30–70% for dividend payers |
| Current Dividend (D₀) | Most recent annual dividend per share | Company-specific |
DDM vs. DCF: When to Use Each
| Dimension | DDM | DCF |
|---|---|---|
| Cash Flow Basis | Dividends only | Free cash flow to firm or equity |
| Best For | Stable dividend payers | Any cash-generating company |
| Complexity | Low to moderate | Moderate to high |
| Limitation | Ignores retained cash flow | Requires more assumptions |
| Convergence | Both give similar results when payout ratio is stable and predictable | |
Key Takeaways
- The DDM values a stock as the present value of all expected future dividends.
- The Gordon Growth Model (D₁ ÷ (r − g)) is the simplest DDM — use it for stable, mature dividend payers.
- Two-stage and H-models handle companies transitioning from high to stable growth.
- The model is highly sensitive to the r − g spread — always run sensitivity analysis.
- DDM and DCF converge when dividends track free cash flow consistently.
Frequently Asked Questions
Can I use the DDM for a company that doesn’t pay dividends?
Not directly. The DDM requires actual dividend payments to value. For non-dividend payers, use a DCF model based on free cash flow, or estimate what dividends could be if the company started paying out. Some analysts use a theoretical payout approach, but it adds assumptions that weaken the model.
What growth rate should I use in the Gordon Growth Model?
The terminal growth rate should not exceed the long-term nominal GDP growth rate (typically 2–4% in developed markets). You can also estimate it as retention ratio × ROE. Using a growth rate above the cost of equity makes the formula mathematically invalid.
Why does a small change in growth rate cause such a big change in value?
Because the Gordon Growth Model divides by (r − g), and when that spread is small, the denominator is small — making the result very large. Moving g from 3% to 4% when r = 9% changes the denominator from 6% to 5%, increasing the valuation by 20%. This is why sensitivity analysis is essential.
How do I calculate the cost of equity for the DDM?
Use the Capital Asset Pricing Model (CAPM): Cost of Equity = Risk-Free Rate + Beta × Equity Risk Premium. The risk-free rate is typically the 10-year Treasury yield. Beta measures the stock’s volatility relative to the market. The equity risk premium is the excess return investors demand for holding stocks over risk-free bonds.
Is the DDM still relevant for modern valuation?
Yes — especially for income-oriented investors and for valuing regulated utilities, REITs, banks, and other high-payout sectors. It’s also a key topic on the CFA exam. While it’s less popular for growth stocks, the underlying principle — that a stock’s value equals the present value of its future cash distributions — remains foundational to all valuation theory.