Defending Terminal Value Assumptions to an Investment Committee

Committee challenges to a terminal value are predictable. In several years of sitting on both sides of the table I have heard three objections and, apart from the occasional idiosyncratic one, no others. Perpetuity growth is above long run nominal growth. Returns on invested capital never fade. The exit multiple you used came out of a DCF that used the exit multiple. That is the list.

Predictable objections should be pre-answered. The analyst who arrives with three exhibits already built spends the meeting discussing the business. The analyst who has to reason through the ROIC fade question live spends the meeting defending arithmetic, and the committee learns nothing about the company. Below is what each exhibit needs to contain.

Growth above long run nominal growth

The objection is arithmetic rather than a matter of taste. A firm compounding forever at a rate above nominal growth of its end markets converges on being those markets. Committees know this, and the challenge is usually a test of whether the analyst does.

The exhibit is a single table: your terminal growth rate, the long run nominal growth of the economies the revenue actually comes from, weighted by segment revenue, and the gap between them. If the gap is negative or zero, the objection dies in one slide. If it is positive, you need a stated reason, and there is only one class of reason that survives scrutiny, which is that the addressable market is still expanding as a share of the economy. That claim is checkable and therefore arguable, which is what you want.

What does not survive is arguing that the company is exceptional. Exceptional companies are exceptional for decades, not in perpetuity, and the perpetuity is the only thing the terminal growth rate describes. If your case rests on twenty exceptional years, model twenty explicit years and use a conservative perpetuity. That is a stronger position and it moves the argument to the forecast, where you have evidence.

Have the sensitivity ready as well. On a ten year model with a 9 percent discount rate, moving terminal growth from 2.0 percent to 3.0 percent lifts fair value by about 10 percent, and from 2.0 to 4.0 percent by about 24 percent. Knowing your own number is worth being able to say without looking it up.

The return on capital that never fades

This is the objection that catches the most models, because unfaded ROIC hides inside the terminal formula rather than sitting in a visible cell. If you take terminal cash flow as a simple margin on terminal revenue, you have implicitly assumed the company reinvests at its current returns forever, and competition never arrives.

The relationship that makes this concrete is that terminal reinvestment must equal terminal growth divided by terminal ROIC. At 2.5 percent terminal growth, a business earning 25 percent on capital needs to reinvest only 10 percent of after tax operating profit to sustain that growth, so 90 percent falls through as free cash flow. Fade the terminal return and the reinvestment burden rises immediately.

The final column assumes the terminal value is 58 percent of total present value, which is roughly where a ten year model lands at these parameters. The useful thing about this exhibit is that it reframes the argument. You are no longer defending an opinion about competitive advantage, you are showing the committee what each opinion is worth in percent of fair value, and inviting them to pick. In my experience committees accept a faded number they chose far more readily than an unfaded number you defended.

I include that screenshot because the Company Valuation Engine makes the same commitment a committee asks of you. It fuses its inputs into one verdict per company, and at the moment of capture it declined to name a single strong buy across a universe of 4,420 companies. A model that outputs a verdict is a model that can be wrong in a recorded way, which is exactly the posture a defensible terminal assumption needs.

Exit multiple circularity

The third objection is the sharpest and the one analysts most often walk into. You used an exit multiple for terminal value. Where did the multiple come from? If the answer is trading multiples of current comparables, you have valued a mature terminal year business at the multiple of today's growing businesses. If it came from a perpetuity you ran separately, you have a perpetuity wearing a multiple's clothing and the committee will find it.

The clean answer is to run both methods and reconcile them explicitly, which turns the circularity objection into a cross check.

State the exit multiple your perpetuity implies. Terminal value divided by terminal year cash flow is a multiple whether or not you called it one. At 9 percent discount and 2.5 percent terminal growth that ratio is 15.8 times. At 3.0 percent growth it is 17.2 times, and at 10 percent discount with 2.0 percent growth it is 12.8 times. Then state, separately, the multiple that mature businesses in the industry actually trade at, sourced from a peer list you fixed before running the model. If those two numbers are close, say so and move on. If your perpetuity implies 20 times and mature peers trade at 12, the committee is going to ask about it, and it is much better for you to have raised it first.

The order matters. Deriving the perpetuity, then observing that it agrees with market multiples, is a valid cross check. Choosing a multiple that produces the fair value you wanted and reverse engineering a perpetuity growth rate to match is not, and the difference is visible in the file: one has the growth rate fixed by policy and the multiple as an output, the other has it the other way around.

Building the exhibit pack once

All three exhibits are the same three exhibits for every name, which means they should be a template rather than work.

The pack is three pages. Page one is the growth reconciliation: terminal growth against revenue weighted nominal growth of the end markets, with the gap and its justification. Page two is the ROIC fade table above, populated with the company's own numbers, showing the value effect of each fade path. Page three is the multiple reconciliation: implied exit multiple from the perpetuity, mature peer multiples from a pre fixed list, and the gap.

Two disciplines make the pack credible rather than decorative. The fade paths and the peer list must be fixed before the fair value is computed, otherwise the exhibits are a defence constructed after the conclusion. And the pack must be capable of embarrassing you, meaning at least one path in it should produce a fair value that does not support the trade. A sensitivity table where every cell supports the position is not a sensitivity table, it is a marketing document, and committees recognise the difference faster than analysts expect.

When the assumption really is aggressive

Sometimes the challenge is correct and the terminal assumption cannot be defended on evidence. The instinct is to negotiate the number down until the fair value still clears the hurdle, which is the worst available response because it produces a model nobody believes and a position nobody owns.

The better response is to change what the position is. Say plainly that the case requires terminal growth at the top of the defensible range, show what the fair value looks like at the middle of that range, and propose a position size against the conservative number rather than the base case. That converts an argument about an unobservable input into a decision about sizing, which is a decision the committee is properly equipped to make.

It also leaves a record that is useful later. When the position is reviewed, the file says the terminal assumption was known to be aggressive, the sizing reflected it, and the committee agreed to that trade. Compare that with a file showing a terminal growth rate quietly revised down in the third draft, and it is obvious which one an analyst wants to be reading from.