In a discounted-cash-flow (DCF) model you can only forecast individual cash flows for a finite window — typically 5 to 10 years — yet a going concern is presumed to live indefinitely. Terminal value (TV) is the lump-sum estimate of all cash flows beyond the explicit forecast horizon, valued as of the final forecast year and then discounted back to the present. Its core tension is structural: because the bulk of a long-lived firm's cash generation lies in the distant future, TV routinely makes up the majority of the computed enterprise value — which makes the whole valuation acutely sensitive to a small number of long-run assumptions that no one can observe. The two mainstream methods — the perpetuity (Gordon) growth model and the exit multiple method — embody two different philosophies for taming that uncertainty.

How it's calculated / formed

Perpetuity growth (Gordon growth) method. Treats the terminal year's free cash flow as a growing perpetuity:

> TV = FCF₍ₙ₎ × (1 + g) / (WACC − g)

where FCF₍ₙ₎ is the final explicit-year free cash flow, g is the perpetual growth rate, and WACC is the weighted average cost of capital (use cost of equity if discounting equity free cash flow). The result sits at year n and must then be discounted to present value alongside the explicit cash flows. The hard constraint: g must be below WACC (or the formula explodes/goes negative) and, more importantly, below the long-run nominal growth rate of the economy. Damodaran's standing rule is that the perpetual growth rate cannot exceed the risk-free rate, which proxies for nominal GDP growth — no firm can outgrow the economy forever. In practice analysts use roughly 2%–4%, with ~2.5%–3% common for stable U.S. companies (Wall Street Prep, Valuation Master Class).

Exit multiple method. Applies a market-observed valuation multiple to a terminal-year metric:

> TV = Terminal-year metric × Exit multiple

Most commonly EV/EBITDA (or EV/EBIT) for an unlevered DCF, since the output is enterprise value; a levered DCF must use an equity multiple such as P/E to stay internally consistent (Wall Street Prep). The multiple is typically drawn from current trading comparables or transaction comps.

Mid-year convention (an optional refinement) assumes cash flows arrive mid-period rather than year-end, shifting discount exponents by half a year and modestly raising both explicit and terminal PV.

How it's used in practice

The discipline that separates a credible model from a number-fishing exercise is running both methods and reconciling them. Each cross-checks the other: solve the perpetuity formula for the g implied by your chosen exit multiple, and check the EV/EBITDA implied by your chosen g. Large divergence is a signal that one input is unreasonable — for example, a "conservative" 6x EBITDA exit multiple may secretly embed negative perpetual growth, or a 4% g may imply a multiple far above where comparable firms actually trade. Appraisers treat material divergence between the two methods as a flag that inputs need review (Valuation Master Class, ValuAdder).

Practitioners often prefer the exit multiple for client-facing work because "10x EBITDA based on comps" is easier to defend to an investment committee than "2.7% perpetual growth" (Sofer Advisors). Academics and Damodaran lean the other way (see below). The mature analyst uses the perpetuity model as the theoretical anchor and the exit multiple as a market sanity check — not the reverse.

Adoption, debate & evidence

Both methods are universal in banking, equity research, and appraisal — they are not contested as techniques; the debate is over which is more sound and how much to trust the result.

The central controversy is the "terminal value ate my DCF" critique: because TV commonly accounts for ~60%–80% of total enterprise value (and a value above ~85% is widely treated as a red flag warranting a longer forecast period — Damodaran Online, Wall Street Prep), critics claim the DCF is "really" just a guess about the far future dressed up as rigor. Damodaran rebuts this directly: a high TV proportion is normal and expected for any long-lived going concern, because the bulk of equity returns historically comes from price appreciation, not near-term dividends. His sharper point is that a high TV share should make you scrutinize the high-growth-period assumptions more, not less — and that the dominant driver of terminal value is not the growth rate itself but the spread between return on invested capital and cost of capital. When ROIC equals the cost of capital, raising perpetual growth from 2.5% to 5% adds almost no value, because growth that earns only its cost of capital creates nothing.

On the exit multiple, the principled objection is contamination by relative valuation: an exit multiple imports today's market pricing (and any market mis-pricing) into what is supposed to be an intrinsic-value model, and can be circular when the comps include the very firm being valued. There is no robust empirical "edge" claim attached to either method — both are accounting/finance identities, not predictive signals; their accuracy is entirely a function of input quality.

Strengths & limitations

The perpetuity model is theoretically clean and intrinsic, and forces an explicit, falsifiable long-run growth assumption — its weakness is extreme sensitivity to the (WACC − g) denominator, where small changes swing TV enormously. The exit multiple is grounded in observable market data and intuitive, but inherits market sentiment and is internally inconsistent if the multiple type doesn't match the cash-flow stream (a frequent error). The #1 misuse across both is failing to reconcile them — picking whichever method (and input) produces the desired answer, which turns the DCF into reverse-engineered advocacy. A close second is the mechanical error of leaving TV undiscounted, or discounting it with the wrong number of periods.

Sources

• Aswath Damodaran, "Myth 5.5: The Terminal Value ate my DCF!" — Musings on Markets (2016) and Stern Session 9 / Growth Rates and Terminal Value slides (primary author; growth ≤ risk-free-rate constraint, ROIC-vs-cost-of-capital point, TV-proportion rebuttal).
• Wall Street Prep, "Common Errors in DCF Models" (2%–4% terminal growth, levered vs. unlevered multiple matching, 85% red-flag threshold).
• Valuation Master Class, "Terminal Value Formula: Gordon Growth vs Exit Multiple."
• Sofer Advisors and ValuAdder, "Terminal Value" / "Perpetuity Growth vs Exit Multiple" (practitioner preference, reconciliation practice).
• Wall Street Oasis; IB Interview Questions — TV formula and method comparison.

Flag: the "60%–80% of EV" and "85% red-flag" figures are widely cited by practitioner sources (Damodaran Online, Wall Street Prep) but are conventions/rules of thumb, not statistically estimated thresholds. The 1928–2015 return-decomposition figures are Damodaran's; exact percentages vary by data window.