valuation
Terminal value is the present-value stand-in for all cash flows after the explicit forecast. In a typical DCF it is most of enterprise value, computed either as a growing perpetuity of FCFF or as an exit multiple on a mature earnings measure.
After you have forecast five to ten years of free cash flow, you still have a going concern. Terminal value (TV, continuing value, horizon value) replaces the infinite tail with one amount at year n, which is then discounted like any other cash flow in the DCF.
Two professional constructions:
Gordon / perpetuity growth (on FCFF):
TV_n = FCFF_n × (1 + g) / (WACC − g)
Exit multiple:
TV_n = Metric_n × Exit multiple (usually EBITDA_n or NTM EBITDA after n × a normalised EBITDA multiple)
g must be less than WACC and, for a mature firm in perpetuity, not above long-run nominal growth of the economy you are in. Exit multiple must match a mature peer, not the target's current growth multiple.
If WACC is 9% and g is 2%, a large share of PV lives beyond year 8. That is mathematics, not a trick. It becomes a trick when year-n margins are peak-cycle, reinvestment is too low for g, or ROIC in perpetuity exceeds WACC without fade. Damodaran's fade and reinvestment identities exist to stop that trick:
In stable growth, g = reinvestment rate × ROIC. If you set g = 2.5% and ROIC = 25% forever, you have implied a tiny reinvestment rate that may contradict your capex forecast.
Invented Harbourline, year-8 FCFF €5.5m, year-8 EBITDA €8.4m, WACC 9.5%, g 2.0%. Illustration only.
| Method | TV at year 8 | Comment |
|---|---|---|
| Gordon | 5.5 × 1.02 / (0.095 − 0.02) = €74.8m | Implies EV/EBITDA at t=8 of 74.8/8.4 = 8.9x |
| Exit 8.0x EBITDA | €67.2m | More conservative; implies a lower g or higher reinvestment |
| Exit 10.5x EBITDA | €88.2m | Needs a story why mature multiples stay this high |
Showing both methods is the control. If Gordon implies 14x mature EBITDA, g or margin is wrong.
ICs always ask "what percentage of value is terminal?" High is normal; unexplained is not. LBO models replace Gordon with an exit multiple in the hold year—the same economic object with a sponsor timetable. Fairness DCFs that hide TV in a footnote are not fairness.
TV assumes a going concern in a stable regime. It is the wrong tool for finite concessions, mines, or firms that will be liquidated. It cannot rescue a nonsense explicit period; it will amplify it.
Dazur always prints TV as a share of EV, the implied exit multiple, and a g/WACC grid. That is how we keep a discounted cash flow from becoming a perpetual-growth marketing document. If the implied exit multiple cannot live on a comps page, the terminal value cannot live in the IC memo.