DCF valuation · Terminal value
Terminal Value: Gordon Growth Model & Exit Multiple
Terminal value captures every cash flow beyond your explicit DCF forecast — and on most stocks it drives 60–80% of the answer. Get it wrong and the whole valuation is wrong.
Terminal value estimates all the cash flows a business generates beyond the DCF forecast window — the perpetual value that sits at the end of the model. The Gordon Growth Model computes it as TV = FCFn × (1 + g) ÷ (WACC − g); the Exit Multiple method as EBITDAn × a sector multiple. Terminal value typically represents 60–80% of total DCF value, so the long-run growth rate (1.5–3% for most companies) is one of the highest-leverage assumptions in the model.
Gordon Growth Model (Perpetuity Growth)
The Gordon Growth Model — also called the perpetuity growth method — treats the business as a cash machine that grows forever at a single stable rate. You take the final forecast year's free cash flow, grow it one more year, and divide by the spread between your discount rate and that long-run growth rate.
The single most abused input is g. Because terminal value is divided by the spread (WACC − g), nudging g up by even half a percentage point can swing the result dramatically. Keep it at or below long-run GDP growth unless you can make a structural argument for permanence — optimism is not an argument.
Exit Multiple Method
The exit multiple method sidesteps the perpetuity assumption entirely. Instead of growing cash flows forever, you assume the business is sold at the end of the forecast window for a market-based multiple of its final-year earnings:
Terminal value = final-year EBITDA multiplied by a sector EV/EBITDA multiple. The multiple is anchored to where comparable companies currently trade (or have traded historically through the cycle), so the terminal value reflects real market pricing rather than a theoretical growth-in-perpetuity assumption.
Prefer the exit multiple when you are valuing a cyclical business, a private-equity-style deal with a defined hold period, or a company where a sensible perpetual growth rate is hard to defend. The catch: a multiple is itself a compressed assumption about growth and risk, so anchor it to a through-the-cycle average rather than a peak-market print, and always cross-check it against the Gordon Growth result.
Worked Example: Apple (AAPL)
Assume a terminal-year free cash flow of $115B, a long-run growth rate of 2.5%, and a 9% WACC. Plugging into the Gordon Growth Model:
- Terminal FCF (FCFn)
- $115B
- Grown one year (× 1.025)
- $117.9B
- Spread (WACC − g = 9% − 2.5%)
- ÷ 0.065
- Terminal value = 115 × 1.025 ÷ 0.065
- ≈ $1.81T
That $1.81T is the undiscounted terminal value sitting at the end of the forecast. Once discounted back to today and added to the present value of the explicit forecast years, it will routinely account for 60–80% of the total DCF value — which is exactly why a 2.5% versus 3.5% growth assumption matters far more than any single year of the forecast.
Gordon Growth vs Exit Multiple: When to Use Each
- Use Gordon Growth for stable, mature businesses. Consumer staples, utilities, and large-cap franchises with predictable, low-single-digit long-run growth are exactly what the perpetuity assumption is built for.
- Use the exit multiple for cyclicals and defined-horizon deals. When a steady perpetual growth rate is hard to defend — commodities, early-stage growth, LBO-style models with a planned exit — a through-the-cycle market multiple is more honest than a made-up g.
- Run both and triangulate. If the two methods disagree sharply, that gap is telling you your growth assumption and your implied multiple are inconsistent. Reconcile them before trusting either.
- Sanity-check the implied multiple. Always back out the EV/EBITDA your Gordon Growth terminal value implies. A perpetuity growth rate that produces a 40× implied multiple is doing too much work and should be dialed back.
Why Terminal Value Is So Sensitive to Small Assumptions
The fragility lives in the denominator. Terminal value is divided by the spread between WACC and g, and that spread is usually small — a 9% WACC against a 2.5% growth rate leaves just 6.5 percentage points. When two large numbers are subtracted to produce a small one, tiny changes in either input move the result disproportionately. Drop the spread from 6.5% to 5.5% — by nudging g up one point — and the terminal value jumps roughly 18%. Because that value dominates the total DCF, the whole fair-value estimate moves with it. This is why a DCF should never be reported as a single number: run the terminal growth rate and WACC across a range and look at the band, not the point. Our DCF calculator builds that sensitivity analysis automatically, so you can see how the valuation shifts as the terminal assumptions move.