Financial modelling6 min read
DCF Terminal Value: Gordon Growth or Exit Multiple
The terminal value in a DCF often makes up more than two thirds of enterprise value. Here are the two ways to calculate it, the Gordon growth model and the exit multiple, a worked example and the checks that keep it from being inflated.

Why DCF terminal value weighs so much
A DCF (discounted cash flows) projects a company's cash flows over five to ten years, then brings them back to today's value. But a company does not stop at the last year of the plan. The DCF terminal value sums up, in a single figure, every cash flow that follows the explicit forecast period, out to infinity.
That figure dominates the result. In the example below, it accounts for 73% of enterprise value, a common proportion for a growing company. An error of a few tenths of a point on a terminal value assumption therefore moves the valuation far more than an error on year-two revenue. Two methods coexist: the perpetuity growth method, known as the Gordon growth model (or Gordon-Shapiro in French-speaking markets), and the exit multiple. Investment banks calculate both and compare them.
Method 1: the Gordon growth model (perpetuity growth)
The method assumes that after the last year of the plan, free cash flow grows at a constant rate g forever. The sum of those discounted flows has a closed formula:
Terminal value = FCF(n+1) ÷ (WACC − g)
- FCF(n+1) is the free cash flow (what is left after tax, capital expenditure and the change in working capital) in the first year after the plan, that is FCF(n) × (1 + g).
- WACC is the weighted average cost of capital, the rate at which all the flows are discounted. Its calculation is covered in our article on how to calculate WACC.
- g is the perpetual growth rate.
The value you obtain sits at the end of year n. You then need to discount it over n years, just like the year-n cash flow.
Choosing g takes discipline. No company can grow faster than the economy forever, or it would eventually outgrow it. So you pick a rate close to long-term nominal growth: in the eurozone, most practitioners stay between 1.5% and 2.5%, around the ECB's 2% inflation target. A g of 4% or 5% almost always betrays a plan that is too short, one that stops before growth has normalised.
Method 2: the exit multiple
The second method assumes the company is sold at the end of the plan, at the price the market pays for comparable companies:
Terminal value = EBITDA(n) × EV/EBITDA multiple
EBITDA is operating profit before depreciation and amortisation; the EV/EBITDA multiple relates enterprise value to that EBITDA. You take it from listed comparable companies or recent transactions in the sector. Our article on the EBITDA multiple by industry explains how to choose it.
This method appeals to buyers because it rests on observed prices. It has one flaw of principle: it imports a dated market price into a DCF that is meant to measure intrinsic value. If sector multiples are high at the time of the calculation, the terminal value is high too.
Worked example: an industrial company
Take a fictional industrial company with a five-year plan, a WACC of 9% and net debt of €4.0m.
| Year | 1 | 2 | 3 | 4 | 5 |
|---|---|---|---|---|---|
| Free cash flow (€m) | 1.6 | 1.7 | 1.8 | 1.9 | 2.0 |
| Discount factor at 9% | 0.917 | 0.842 | 0.772 | 0.708 | 0.650 |
| Present value (€m) | 1.47 | 1.43 | 1.39 | 1.35 | 1.30 |
The sum of the discounted flows over the five years is €6.93m. Year-5 EBITDA is €3.2m.
With the Gordon growth model (g = 2%):
- year-6 cash flow: 2.0 × 1.02 = €2.04m;
- terminal value at the end of year 5: 2.04 ÷ (9% − 2%) = €29.14m;
- present value: 29.14 ÷ 1.09⁵ = €18.94m;
- enterprise value: 6.93 + 18.94 = €25.88m, of which 73% comes from the terminal value.
With an exit multiple of 7.0x:
- terminal value: 3.2 × 7.0 = €22.4m;
- present value: 22.4 ÷ 1.09⁵ = €14.56m;
- enterprise value: 6.93 + 14.56 = €21.49m.
The two methods are €4.4m apart on the same company. Once net debt is deducted, equity value goes from €21.9m to €17.5m (the bridge between the two is set out in our article on enterprise value vs equity value). Neither is wrong a priori. The gap signals that the assumptions do not tell the same story, and that is where cross-checks earn their keep.
The cross-check: implied multiple and implied growth
Each method hides an assumption from the other. Two calculations bring it to light.
The implied multiple of the Gordon growth model. Divide the terminal value by year-5 EBITDA: 29.14 ÷ 3.2 = 9.1x. If comparable companies trade at 6x to 8x, growth of 2% implies an exit valuation above the market. Either the WACC is too low or the year-6 cash flow is too optimistic.
The implied growth of the exit multiple. Solve for the g that gives the same terminal value: 22.4 = 2.0 × (1 + g) ÷ (0.09 − g), so g ≈ 0.1%. A 7x multiple amounts to assuming the company will barely grow, in current euros, after year 5. For a company growing 5% to 6% a year during the plan, that is prudent, perhaps too much so.
In this example, the truth probably lies between the two: a g of 1.5% or a multiple of 8x. A credible DCF shows both methods with their implied values, and explains the gap.
Sensitivity: what one point of WACC and half a point of g do
The table shows enterprise value (in €m) using the Gordon growth model, for the same cash flows:
| WACC \ g | 1.5% | 2.0% | 2.5% |
|---|---|---|---|
| 8% | 28.4 | 30.3 | 32.5 |
| 9% | 24.5 | 25.9 | 27.4 |
| 10% | 21.6 | 22.6 | 23.7 |
One point of WACC moves the value by €3.3m to €4.4m, or 13% to 17%. Half a point of growth moves it by about €1.5m. The gap between the lowest and highest corners reaches 50%. Presenting a DCF as a single number therefore makes no sense: you show a range, often on a football field chart that compares it with the other methods.
The four most common mistakes
- Discounting over n + 1 years. The Gordon growth formula gives a value at the end of year n. Discounting it over six years instead of five brings the terminal value here from €18.94m to €17.38m and the enterprise value from €25.88m to €24.31m, 6% less for no reason.
- Extending a cash flow that is not normalised. If year 5 includes €0.5m of capital expenditure against €0.7m of depreciation, the company is consuming its asset base. In perpetuity, capital expenditure must at least equal depreciation. Normalising the cash flow to €1.8m brings the discounted terminal value from €18.94m to €17.05m. Our article on the business plan behind a DCF shows how to build that normalised year.
- Picking a g above or close to the WACC. As g approaches the WACC, the denominator tends to zero and the value to infinity. A gap of less than 4 or 5 points between the two deserves a written justification.
- Mixing conventions. A nominal WACC requires a nominal g, inflation included. A multiple drawn from past transactions often embeds a control premium, which you must not add a second time.
In an Excel model
In a spreadsheet, isolate the terminal value assumptions (WACC, g, multiple) in dedicated cells, calculate both methods side by side, and add three check lines: the share of terminal value in enterprise value, the implied multiple and the implied growth. If the share exceeds 80%, the plan is probably too short. A WACC × g sensitivity table, built with Excel's Data Table function, completes the set: our guide to the sensitivity table in Excel explains how to build it and what to avoid. To review the rest of the model, from the cash flow to the discounting, follow our five checks for a DCF model review.
Our Excel DCF model calculates terminal value using both methods, shows the implied multiple and implied growth, and includes the WACC × growth sensitivity table. It is a calculation tool: the result remains an estimate that depends on your assumptions, and is neither a certified valuation nor investment advice.


