Price is a fact. It is what somebody paid, at a moment, and it is published.
Value is an estimate. It is what an asset is worth on some set of assumptions, and it exists only in the model that produced it.
Treating the second as though it had the observable character of the first is a persistent source of confusion.
Where value estimates come from
The standard approach values an asset as the present value of the cash it is expected to produce.
Which requires forecasting future cash flows, and choosing a rate at which to discount them.
Both inputs are judgements. Small changes in either produce large changes in the output, particularly for assets whose cash flows are expected far in the future.
Which is why growth companies have such volatile valuations. Not because their businesses change quickly, but because the discount applied to distant cash flows is enormously sensitive to rates.
The multiple shortcut
Most practical valuation uses multiples — price relative to earnings, sales, book value or cash flow — rather than explicit forecasting.
A multiple is a compression of the full calculation into one number, and it hides the assumptions rather than removing them.
Which makes multiples useful for comparison and misleading when used absolutely.
Two companies with the same multiple can have quite different prospects, and the multiple says nothing about which.
What a multiple actually encodes
A higher multiple implies some combination of higher expected growth, lower expected risk, or higher return on the capital employed.
Which means describing a high multiple as expensive is incomplete. It is expensive relative to current earnings, and whether that is justified depends entirely on the expectations embedded in it.
The useful question is not whether a multiple is high. It is what growth rate would have to be achieved to justify it, and whether that rate is plausible.
Working backwards from the price to the implied expectation is more informative than working forwards from assumptions to a target.
Book value and its limits
Book value measures assets less liabilities as recorded in the accounts.
For a bank or an insurer, whose assets are largely financial and marked in some fashion, this carries real information.
For a software company whose principal assets are people, code and customer relationships, none of which appear on a balance sheet, it carries almost none.
Which is why price-to-book has become less useful as economies have shifted toward businesses whose assets are intangible.
Accounting standards treat internally developed intangibles quite differently from acquired ones, which makes comparison across companies genuinely difficult.
Earnings are an accounting output
Worth stating plainly. Reported earnings depend on accounting policy choices within permitted ranges.
Revenue recognition timing, depreciation schedules, provisioning, capitalisation of development costs — each involves judgement.
Which means earnings comparisons between companies with different policies are not clean, and the notes explaining the policies are where the useful information is.
Cash flow is harder to adjust and is correspondingly more informative, which is why analysts look at it.
The market as an aggregator
Price reflects the aggregate of participants' views, weighted by the money behind them.
Which makes it informative — it summarises what many people who have studied a company collectively believe.
And it does not make it correct, since collective belief has been wrong on a large scale repeatedly.
The reasonable position is that price is a strong prior, not a conclusion, and that beating it requires knowing something the aggregate does not.
Most people, most of the time, do not.
Why this is worth thinking about
Because most commentary states valuations as facts.
Once you know that every valuation is a model output with assumptions inside it, the correct response to a stated target price is to ask what was assumed, and the answer is generally not provided.
That single habit filters out a large amount of noise.
None of this is advice about any investment. It is a description of how these estimates are constructed and where they are fragile.
Margin of safety
The concept that follows directly from valuation being uncertain.
If an estimate of value carries a wide error band, then acting only when price sits well below the central estimate provides room for the estimate to be wrong.
Which is a response to model uncertainty rather than a prediction, and it is a different discipline from trying to estimate more precisely.
The practical difficulty is that large discounts to a plausible estimate of value usually exist for reasons, and distinguishing a mispricing from a correct assessment of a deteriorating business is the entire problem.
Reflexivity
A complication worth naming.
Price does not only reflect value; it can affect it.
A company whose share price falls may face higher borrowing costs, lose customer confidence and find raising capital dilutive, all of which reduce the value the price was supposedly estimating.
Which means the separation between price and value is cleaner in theory than in practice, particularly for businesses dependent on continuous access to funding.