Dividend Discount Model Calculator

The calculation itself

A share is worth the present value of its dividends forever. If those dividends grow at a constant rate, the infinite sum collapses to one line: next year's dividend divided by the required return less the growth rate.

That is the Gordon growth model, and its simplicity is why it is taught first and misapplied often. It assumes constant growth to infinity, which almost no company delivers.

The two-stage version forecasts dividends explicitly through a high-growth phase, discounts them individually, and applies Gordon growth from the end of that phase. It is what the constant model should become as soon as growth is visibly not constant.

In practice

A $2 dividend, a 10% required return and 4% perpetual growth. Next year's dividend is $2.08, and $2.08 ÷ 0.06 gives $34.67 a share.

Now the two-stage version: five years at 15% growth, then 4% forever. The five forecast dividends are worth $11.45 in present value. The price implied at year five is $69.73, worth $43.30 discounted back. Fair value is $54.74.

The high-growth phase is worth $20.07 a share, nearly 60% more than the constant model. That splits as $11.45 from the forecast dividends themselves and $8.63 from the higher price they leave behind at year five — the dividends are the larger half, which is not what the shape of the model suggests at a glance.

The instability near r = g

The perpetuity divides by the gap between the required return and perpetual growth. As that gap narrows, the valuation explodes.

Concretely: at a 10% required return, 9% growth gives $218 a share. Raise growth to 9.5% — half a percentage point, well inside anyone's error bar — and it becomes $438. The valuation doubles on a rounding difference.

This is not a defect in the arithmetic. It is what a perpetuity growing almost as fast as it is discounted genuinely implies, and it is the reason the model should not be used on companies growing near their cost of equity.

The practical rule: if the gap is under about two points, the output is not a valuation, it is a sensitivity demonstration.

Where the figure deceives

It values dividends, so it cannot value a company that pays none. Applying it to a growth company by inventing a future dividend produces a number governed entirely by the invention.

The required return must be the cost of equity, not WACC. Dividends are paid to shareholders after debt is served, and using the blended rate understates the return required and overstates the value.

Buybacks break the model quietly. A company returning cash by repurchasing shares looks like it has a low payout, and a dividend-only model undervalues it — which is a large share of the market.

And perpetual growth above the long-run growth of the economy implies the company eventually becomes the economy. It is a common input and it is never right.

Acting on it

Check the gap between required return and perpetual growth before anything else. Under two points and the number should not be quoted.

Use the two-stage form for anything not already mature. Constant growth is a modelling convenience, rarely a description.

For companies returning cash through buybacks, value free cash flow to equity rather than dividends — otherwise the model is measuring dividend policy instead of value.