Cost of Equity With the Dividend Growth (Gordon Growth) Model

The dividend growth model estimates cost of equity as next year’s dividend divided by the share price, plus the expected dividend growth rate. For a $50 stock paying $2.00 a year with 4% growth, next year’s dividend is $2.08, the forward yield is 4.16%, and the cost of equity is 8.16%.

You’ll also see it called the Gordon growth model or the constant growth dividend discount model. You can run it in the Dividend growth tab of the cost of equity calculator. This guide covers the formula, the most common mistake people make with it, and the cases where it gives you a bad answer.

The formula

Cost of equity = D1 / P + g
Ke = D0 x (1 + g) / P + g

• D1 is the dividend expected over the next 12 months.
• D0 is the current annual dividend (the one already paid).
• P is the current share price.
• g is the dividend growth rate, assumed constant forever.

The logic is that a shareholder’s return comes from two places: the dividend they receive and the growth in that dividend (which, in this model, also drives growth in the share price). Add the two and you get the return the market price implies.

The first piece, D1 / P, is the forward dividend yield. The second piece, g, is the capital gain.

D0 vs D1

This is where most errors happen. The formula needs next year’s dividend, D1. The dividend you see quoted on a finance site is usually the trailing or current annual dividend, D0.

To get D1, grow D0 by one year:

D1 = D0 x (1 + g)

If you skip this step and plug in D0, your cost of equity comes out slightly too low. The calculator asks for the current dividend and grows it for you, so you don’t have to remember.

Worked example

Inputs:

• Share price: $50
• Current annual dividend (D0): $2.00
• Expected growth rate: 4%

Step 1. Grow the dividend one year: $2.00 x 1.04 = $2.08.

Step 2. Divide by the share price: $2.08 / $50 = 0.0416, or 4.16%.

Step 3. Add growth: 4.16% + 4% = 8.16%.

The cost of equity is 8.16%.

If you’d used D0 by mistake, you’d get $2.00 / $50 + 4% = 8.00%. That’s 0.16 points too low. Small here, but the gap grows with higher dividends and faster growth.

How sensitive is the answer to growth?

Growth enters the formula twice: once to turn D0 into D1, and again on its own. So the answer moves almost one for one with g. Same $50 price and $2.00 current dividend:

Growth rate	D1	Forward yield	Cost of equity
2%	$2.04	4.08%	6.08%
3%	$2.06	4.12%	7.12%
4%	$2.08	4.16%	8.16%
5%	$2.10	4.20%	9.20%
6%	$2.12	4.24%	10.24%

A one-point change in your growth assumption moves cost of equity by a bit more than one point. Your growth estimate deserves as much scrutiny as any other input.

Where the growth rate comes from

There’s no single source. Analysts usually look at a few and pick a defensible number:

• Historical dividend growth over the last five to ten years.
• Analyst forecasts of long-term earnings growth.
• Sustainable growth, calculated as retention ratio x return on equity. A company that keeps 60% of earnings and earns 10% on equity can grow about 6% a year without new capital (0.6 x 10% = 6%).
• A cap tied to long-run economic growth, since no company can outgrow the economy forever.

Because the model assumes g lasts forever, a rate that looks fine for five years can be far too high as a permanent assumption.

When the model fails

The Gordon growth model works for a narrow kind of company: a mature business with a steady dividend and steady growth. Utilities, consumer staples and some banks often fit. Plenty of companies don’t.

Companies that don’t pay dividends

If D0 is zero, the formula gives you a cost of equity equal to g, which is meaningless. Many growth companies pay nothing, and many established ones return cash through buybacks instead. For these, use CAPM.

Growth close to the cost of equity

The model is the rearranged version of the share price formula:

P = D1 / (Ke - g)

As g gets close to Ke, the denominator shrinks toward zero and the implied price explodes. Using the example’s 8.16% cost of equity, a 4% growth rate gives a price of $2.08 / 0.0416 = $50. Raise growth to 7% and the same formula gives $2.14 / 0.0116 = about $184.48. If g is above Ke, the price turns negative.

So the model only makes sense when growth is well below the cost of equity. High-growth companies break it.

Irregular dividends

Special dividends, recent cuts, or a dividend that’s about to change sharply all distort D0. If the dividend isn’t a fair picture of what the company will pay going forward, the result won’t be either. A multi-stage dividend model can help, but at that point many analysts switch to CAPM.

Comparing with CAPM

The site’s CAPM example gives 11%. This DDM example gives 8.16%. Different companies and different inputs, so the numbers aren’t meant to match. When you run both on the same company, though, a big gap is worth investigating.

	CAPM	Dividend growth model
Based on	Market risk (beta)	Share price and dividends
Needs a dividend	No	Yes
Needs a share price	No (needs return history for beta)	Yes
Biggest judgment call	Equity risk premium	Growth rate
Best for	Most listed companies	Mature, steady dividend payers

Some analysts average the two for dividend-paying companies. Others use DDM as a sanity check on CAPM. Both are reasonable as long as you can explain the inputs.

Where the model came from

Myron Gordon and Eli Shapiro published the model in a 1956 paper in Management Science, “Capital Equipment Analysis: The Required Rate of Profit.” Gordon developed it further in later work, and his name stuck to the model.

Using the result

The 8.16% you get is the required rate of return implied by the market price. You can use it as the discount rate for cash flows to shareholders or as the equity piece of WACC. If you want to set the whole thing up in a spreadsheet, the Excel guide has the exact cell formulas.

For a business that doesn’t pay dividends and isn’t listed either, see the FAQ on how to calculate cost of equity without beta and the guide on cost of equity for a private company.