Income investing · Dividend valuation · Gordon Growth Model
Dividend Discount Model: Formula, Assumptions & When It Works
The dividend discount model is the most transparent valuation method available to income investors. It says a stock is worth the present value of every dividend it will ever pay. For a utility or consumer staples company with 30 years of uninterrupted dividend growth, that is a tractable estimate. This guide walks through the Gordon Growth Model formula, the two-stage extension for faster growers, how to estimate each input, and the model's key failure modes.
DDM values a stock as the present value of all future dividends. Gordon Growth Model: P = D1 / (r - g), where D1 is next year's dividend, r is the required return, and g is the perpetual growth rate.The dividend discount model (DDM) values a stock as the present value of all future dividends. It is the foundational model for income-stock valuation — if a company pays dividends forever, the DDM says exactly what it is worth today.
What Is the Dividend Discount Model?
The dividend discount model starts from a simple premise: a stock is worth the present value of everything it will ever pay you. For a company that returns cash exclusively through dividends, that sum of discounted future payments is the theoretical fair value of the stock today. If you can predict those dividends and you know your required rate of return, you can price the stock independently of whatever the market happens to be doing.
DDM works best for companies where dividends are genuinely predictable. Utilities, consumer staples, and REITs fit this profile: regulated cash flows, modest growth, high payout ratios, and decades of uninterrupted dividend history. When you model Realty Income or Procter & Gamble with DDM, you are anchoring to a stream of cash payments that management has committed to maintaining and growing for generations. The model is not speculative in that context — it is arithmetic.
For growth stocks that reinvest everything — Amazon, Berkshire Hathaway, most technology companies — DDM is the wrong tool. You cannot discount dividends that do not exist. For those companies, a discounted cash flow model values the full free cash flow stream, regardless of whether it is returned as dividends, buybacks, or retained for reinvestment.
The Gordon Growth Model Formula
The Gordon Growth Model is the single-stage DDM. It assumes dividends grow at a constant rate forever, which is unrealistic in the strict sense but a reasonable approximation for mature dividend payers over a long horizon.
- D₁ (Next year's expected dividend)
- Current annual dividend multiplied by (1 + g). If a stock pays $3.00 today and you expect 4% growth, D₁ = $3.12. Use the declared annual rate from the company's investor relations page or any financial data provider.
- r (Required rate of return)
- The minimum annual return you require to hold the stock — your cost of equity. Typically derived from CAPM: r = risk-free rate + beta × equity risk premium. For a utility with beta near 0.6 and a 4.5% risk-free rate plus 5% ERP, r = 4.5% + 0.6 × 5% = 7.5%. See the CAPM and cost of equity guide and WACC guide for the full calculation.
- g (Dividend growth rate)
- The expected perpetual dividend growth rate. For most mature dividend payers this is close to nominal GDP growth (2-4%). Never let g approach r — as the spread (r − g) narrows toward zero, the model value approaches infinity, which is a signal that your inputs are unrealistic, not that the stock is worth an unlimited sum.
- Constraint
- r must exceed g. If the assumed growth rate equals or exceeds the required return, the formula produces a negative or infinite result. The model breaks down: either your growth assumption is too high or your discount rate is too low.
Worked Example: Realty Income (O)
Realty Income is a net-lease REIT that has paid monthly dividends since 1994 and raised them annually. It is a classic DDM candidate.
- Current annual dividend (D₀)
- $3.17 per share
- Expected growth rate (g)
- 4% (in line with 10-year dividend CAGR)
- Next year's dividend (D₁ = $3.17 × 1.04)
- $3.30
- Required return (r)
- 9% (CAPM-derived for a moderate-beta REIT)
- Intrinsic value = $3.30 / (0.09 − 0.04)
- $66.00
If Realty Income currently trades at $55, the model implies a 17% margin of safety — the stock is priced below its DDM-derived intrinsic value. A higher required return (say 10%) produces a lower intrinsic value ($55), confirming the market price is consistent with investors demanding a 10% return rather than 9%.
Two-Stage DDM for Faster-Growing Dividend Payers
The Gordon Growth Model assumes dividends grow at a single constant rate forever. For a company currently growing its dividend at 8-10% per year — faster than any company can sustain in perpetuity — the constant-growth assumption produces an overestimate of intrinsic value. The two-stage DDM solves this by separating the valuation into two distinct phases.
Stage 1 covers the high-growth period (typically 5-7 years). You discount each year's dividend individually using the near-term growth rate. The present value of each payment is: PV = Dt / (1 + r)^t.
Stage 2 applies the Gordon Growth Model at the end of Stage 1. Take the final year's dividend (D₇ in a 7-year high-growth model), apply the long-run sustainable growth rate, and calculate a terminal value: TV = D₇ × (1 + g₂) / (r − g₂). Then discount that terminal value back to today: PV(TV) = TV / (1 + r)^7. Total intrinsic value = sum of Stage 1 PVs + PV(TV).
Worked Example: Johnson & Johnson
Johnson & Johnson has raised its dividend every year for over 60 consecutive years. Suppose current DPS = $5.00, Stage 1 growth = 6% for 5 years (consistent with recent 5-year CAGR), Stage 2 (terminal) growth = 4%, and required return = 9%.
- Year 1: $5.00 × 1.06 = $5.30 → PV
- $5.30 / 1.09 = $4.86
- Year 2: $5.30 × 1.06 = $5.62 → PV
- $5.62 / 1.09² = $4.73
- Year 3: $5.62 × 1.06 = $5.96 → PV
- $5.96 / 1.09³ = $4.60
- Year 4: $5.96 × 1.06 = $6.31 → PV
- $6.31 / 1.09&sup4; = $4.47
- Year 5: $6.31 × 1.06 = $6.70 → PV
- $6.70 / 1.09&sup5; = $4.35
- Sum of Stage 1 PVs
- $23.01
- Terminal value = $6.70 × 1.04 / (0.09 − 0.04)
- $139.36 → PV = $139.36 / 1.09&sup5; = $90.56
- Intrinsic value = $23.01 + $90.56
- $113.57
The two-stage model produces a more conservative estimate than applying 6% growth in perpetuity (which would yield $138.90). The terminal value dominates — $90.56 of the $113.57 total, or 80% — which is typical for DDM. This underscores why the assumed long-run growth rate matters far more than the near-term Stage 1 assumptions.
Inputs and How to Estimate Them
DDM has only three inputs, but each requires careful estimation. Garbage in, garbage out.
- Current dividend (D₀)
- Use the declared annual dividend from the company's investor relations page, SEC filings, or a financial data provider. For quarterly payers, multiply the most recent quarterly dividend by 4. Use the actual declared rate, not analyst estimates.
- Dividend growth rate (g)
- Use the 5-year dividend CAGR as a starting point, then sanity-check it against the payout ratio, free cash flow coverage, and analyst consensus. Sustainable growth ceilings by sector: utilities 2-4%, consumer staples 4-6%, high-growth dividend growers 7-10%. The growth rate must be below the required return or the model breaks. For long-run terminal growth, 2-4% (near nominal GDP) is a defensible ceiling for most businesses.
- Required return (r)
- Derive from CAPM: r = Rf + β × ERP. Use the current 10-year Treasury yield as Rf and a forward equity risk premium of 5-5.5% (Damodaran publishes this annually). For a low-beta utility (β = 0.5) with Rf = 4.5%: r = 4.5% + 0.5 × 5% = 7.0%. See the CAPM guide and WACC guide for the full derivation.
When DDM Works and When It Doesn't
DDM works well for companies with long, stable, predictable dividend histories and mature business models with limited reinvestment needs. The ideal DDM candidates are:
- REITs (Realty Income, Prologis, Agree Realty) — legally required to distribute 90% of taxable income; dividend growth is closely tied to funds from operations growth.
- Regulated utilities (NextEra Energy, Consolidated Edison, Duke Energy) — predictable cash flows from regulated rate bases; growth tied to rate cases and capex programs.
- Consumer staples blue chips (Coca-Cola, Procter & Gamble, Colgate) — Dividend Kings and Aristocrats with 25-60+ years of consecutive dividend growth; pricing power supports consistent growth rates.
DDM breaks down in four situations. First, for non-dividend payers like Amazon or Berkshire Hathaway: DDM produces zero because there are no dividends to discount. Second, for volatile or recently-cut payers: a company that slashed its dividend in 2020 provides no stable growth baseline. Third, for high-growth companies that reinvest most free cash flow: the dividend is a small residual and does not reflect the full earnings power of the business. Fourth, for companies where buybacks dwarf dividends: Apple returns far more capital via repurchases than dividends; DDM ignores that value entirely.
For non-payers and growth companies, use a DCF model (which values total free cash flow) or an EV/EBITDA multiple for a comparables-based cross-check.
Sensitivity: Why Small Changes in g Destroy the Model
The (r − g) spread in the denominator is the model's greatest source of instability. With D₁ = $3.00 and r = 9%, here is what happens as g changes:
| Growth rate (g) | (r − g) spread | Intrinsic value |
|---|---|---|
| 2% | 7% | $42.86 |
| 4% | 5% | $60.00 |
| 5% | 4% | $75.00 |
| 6% | 3% | $100.00 |
| 8% | 1% | $300.00 |
Moving g from 2% to 6% more than doubles the intrinsic value ($42.86 to $100). Moving from 6% to 8% triples it again ($100 to $300). This hypersensitivity is not a flaw in the math — it is a feature. It tells you that your margin of safety must be wide enough to survive being wrong about g. If the model only works at an exact growth rate, you have no cushion. The appropriate response is to run the model at conservative, base, and optimistic growth assumptions, and only buy at a price that works in the conservative scenario.
DDM vs. DCF vs. Comparable Company Analysis
DDM is the most transparent valuation method for dividend payers because every assumption is explicit. You declare a growth rate, a required return, and you get a price. The weakness is that it only captures dividends — it misses buybacks, debt paydown, and retained earnings growth entirely. For companies where management returns capital through repurchases rather than dividends (Apple is the textbook example), DDM will systematically undervalue the business. See the discounted cash flow guide for a full comparison of DDM and DCF mechanics.
DCF is more flexible — it values total free cash flow regardless of how it is distributed — but adds complexity. FCF projections, terminal growth, and WACC each introduce estimation error. DDM's simplicity is a genuine advantage for mature payers where dividends are the primary value mechanism. Comparable company analysis (comps) anchors to current market multiples: EV/EBITDA, P/E, or dividend yield of similar companies. Comps tell you whether a stock is cheap relative to peers but not whether peers themselves are fairly valued. Use all three in combination — DDM as a first-principles anchor, DCF as a cross-check, and comps as a sanity check against market consensus. See the DCF vs. comparable company analysis guide for a side-by-side comparison.