Intrinsic value

A company is worth the cash it will hand its owners over its life, valued in today's money. Intrinsic value is an attempt to write that down: forecast the cash, decide what a dollar in five years is worth today, and add it up.

The output looks precise and is not. Every figure in it depends on three assumptions about the future, and small changes to them move the answer enormously. Its value is not the number — it is that it forces the assumptions into the open, where they can be argued with.

Free cash flow of $170m growing at 10% a year, discounted at 9%, with 2.5% growth assumed forever after, plus $200m of cash less $600m of debt across 25m shares.

1. 1Five projected years: $187m, $206m, $226m, $249m, $274m.
2. 2Discounted back to today, those are worth $874m in total.
3. 3The terminal value is $4,317m, worth $2,806m today.
4. 4874 + 2,806 + 200 - 600 = $3,280m of equity, over 25m shares.

Every example in this glossary describes the same imaginary company, so the figures join up as you move between metrics.

• Look at how much of the answer is terminal value. Here it is 76% of the total, which means three quarters of the valuation rests on a guess about the years nobody forecast.
• Change one assumption at a time and watch the answer. If a plausible range of inputs spans half to double the price, the model is telling you it cannot decide.
• Use it to work out what the current price implies, which is often more useful than arguing about what the company is worth.

• It is extraordinarily sensitive to the discount and terminal rates, which are the two figures nobody can observe.
• It needs positive and reasonably predictable cash flow, so it says nothing useful about early-stage or cyclical businesses at the wrong point in the cycle.
• Five years of growth at one rate is a convenient fiction. No business grows smoothly.

The Valuation tab runs exactly this model over five years. The starting free cash flow is the latest fiscal year's operating cash flow less capital expenditure, and where that is not positive no automatic valuation is produced at all — a labelled manual override is required instead, rather than the model quietly inventing a number. The growth rate opens at the company's own three-year free-cash-flow growth, clamped to a sane range, and all three assumptions are editable. The result is compared against the price with a fair band either side, so a small difference reads as fair rather than as a signal.

The full methodology covers where the figures come from, how the statements are normalised, and what is deliberately left unadjusted.