Fair Value: What It Means, How It Is Measured and Where Judgement Enters

Fair value is the price a willing, informed buyer and a willing, informed seller would agree for something, with neither forced to deal. Fair value differs from market price, the number the last trade actually happened at, and from book value, the number the thing cost less what has already been written off. When a market exists, fair value leans on it; when none does, it is estimated, and judgement enters.

Here is the idea underneath. A thing does not carry one number. The same thing carries at least three numbers: what it cost, what somebody last paid for one like it, and what a fair deal struck today would settle at. The three numbers can sit very far apart, and almost every argument about money is really an argument about which of the three somebody has quietly picked. An ordinary object, taken before any business appears, shows all three at once.

Fair value can be defined without using the word value at all, and the definition rests on four conditions, a three-step ladder of measurement, and a small number of choices that decide how wide the answer comes out.

What is fair value, in words a first-time reader can hold?

A three-year-old scooter sits parked in a lane. The scooter cost Rs 85,000 new. Three doors down, a neighbour moving cities on a fixed date sold an identical scooter last month for Rs 32,000. The buyer knew perfectly well that the seller had four days left. Any mechanic on that street, asked what the scooter should fetch when neither side is under time pressure, gives a range: somewhere around Rs 44,000 to Rs 54,000, depending on the tyres and the service record. Three numbers, one scooter. Only the third answers the question a person actually wants answered, and that question is what a sensible deal today would settle at.

Fair value is the price two people would agree on when both of them want the deal, both of them know what they are trading, and neither of them has to do it. Each of those clauses does a great deal of work. Wanting the deal keeps out the seller who is being forced. Knowing what is being traded keeps out the buyer who has not looked under the seat. Not having to do it keeps out the deadline. With any one of the three stripped away there is still a price, and people will still shake hands on it, but that price stops describing the thing and starts describing the situation of whoever was under pressure.

The third bar is a band rather than a line, and that shape is worth holding on to. Fair value is almost never a single rupee figure that falls out of a formula. Fair value is the zone inside which two reasonable, unhurried, well-informed people would settle. When somebody quotes fair value as one exact number, they have either found a live market for the identical thing, or they have made a set of choices and reported only the answer.

Which four conditions have to hold before a price counts as fair value?

The definition hides four separate conditions inside two short phrases, and pulling them apart is worth the two minutes it takes, because in practice a disputed number almost always turns out to have exactly one of them broken. There must be a willing buyerSomeone who wants to acquire the thing on its merits, rather than someone under an obligation, a deadline or an instruction to buy something by a certain date. who wants the thing on its merits. There must be a willing seller who wants to sell rather than needing to. Both sides must be informed, meaning each knows what is being handed over and what it earns. And the deal must be an orderly transactionA deal with the usual amount of time and exposure to the market before it closes, so a normal set of possible counterparties gets a chance to look. The opposite of a distress sale or a three-day auction., given the normal amount of time and exposure that deals of that kind usually get.

Break any single one of the four and the price stops describing the thing and starts describing somebody's circumstances, so all four conditions have to hold at once. A household selling gold in the week a hospital bill lands is not testing the gold's fair value; it is testing how fast it needs cash. A shop clearing stock the night before a lease expires is not testing the stock. A buyer who has not been shown the order book is not informed, and the price agreed reflects what was hidden. Each of the four failures below produces a real, agreed, perfectly binding price. None of them produces a fair value.

How is fair value different from the market price?

A market priceThe amount at which the most recent completed transaction in a thing actually took place, as reported by whoever records those deals. It is one observed fact about the past, not a statement about worth. is a fact about the past. Somewhere there is a place where many buyers and many sellers meet, agree deals, and have those deals recorded, and the last agreed number gets displayed. The last printed number is enormously useful for one reason: it is not anybody's opinion. It happened. Somebody actually parted with money at that level. Most numbers written about value cannot say that much for themselves.

No single fact says more about fair value than the market price. The market price is still not the same thing, and the reason is that any particular trade can be forced, thin or stale. Forced means one side had to deal, so the price carries their deadline. Thin means the trade was tiny, so it shows what one small buyer would pay and nothing about what a serious quantity would fetch. Stale means the last trade happened four months ago and the world has moved since. The plot below sets out the trades. Most of them cluster inside the band where unhurried, informed people were settling. Two do not, and each of the two fails a different condition. A rule of always taking the last printed number would, on the day of the forced sale, have called Rs 100 the value of a thing that had been changing hands around Rs 124 all month.

So the honest relationship runs one way. A market price is evidence about fair value, usually the strongest evidence available, and when the market is deep and the trading is regular the two numbers sit so close together that separating them is a pedantic exercise. When the market is thin, panicked or quiet, they come apart, and the discipline is to ask what conditions produced that particular trade before treating it as the answer.

How is fair value different from book value, what the thing cost?

Sohan Ply and Boards Private Limited, an invented maker of plywood and laminates working one plant, carries the rest of the argument. Sohan Ply sells Rs 1,80,00,00,000 of board a year and earns Rs 21,00,00,000 of operating profit before depreciation. Sohan Malhotra is the owner and Managing Director; Ritu Chandran runs the finance side. Deodar Growth Partners, an invented investor, has offered to buy 20 per cent of the business.

Book valueThe amount at which something is carried in a set of accounts: normally what was paid for it, less the part of that cost already charged against profits. A record of spending, kept in the accounts. is the number the accounts carry. For a machine, book value is what the machine cost, less the portion of that cost already charged against profits year by year. The name for that year-by-year charge is depreciationThe spreading of an asset's cost across the years it is used, so each year's profit carries a share of it. It reduces the amount at which the asset is carried, and it is not an estimate of resale value.. For the business as a whole, book value usually means equityWhat is left when everything a business is recorded as having is added up and everything it owes is taken away. Here it means the book equity in the accounts, not the shares themselves.: add up everything the business is recorded as having, take away everything it has to repay, and what is left is the equity. At Sohan Ply that figure is Rs 62,00,00,000, and it is built as the diagram shows.

Book value says what was spent, fair value says what a deal today would settle at, and no rule anywhere makes the two agree. Sohan Ply's book equity of Rs 62,00,00,000 records historical spending on a plant and a working stock of timber and glue, less two loans. The book equity does not know that the plant runs at high utilisation, that the laminate line has a two-month order backlog, or that a buyer would pay for the earnings rather than for the machines. Deodar's offer values the whole business at Rs 1,10,00,00,000. The offer stands Rs 48,00,00,000 above book, about 1.8 times it, and that gap is not an error in the accounts. The accounts were never trying to answer the question.

The gap does not always run upward, and which way it runs is worth watching. A business whose plant is worth more as scrap than as a working line carries a book value above any price a buyer would pay, and when that happens the accounts eventually have to be written down through a step called impairmentReducing the amount at which something is carried in the accounts when the amount it could realistically bring in has fallen below that carrying figure.. Book above fair value and book below fair value are both ordinary. Book value is never the answer by itself.

How is fair value measured when a market exists, and when none does?

Measurement follows a ladder, and the rule is to stand on the highest rung actually within reach. The top rung is a price quoted right now, in an active market, for the identical thing. Nobody has had to adjust a price quoted right now, so it is the strongest possible evidence: the number is what people are paying for exactly this, today. Very few things sit on that rung. A listed share does. A tonne of a standard grade of steel roughly does. A single privately held plywood business does not, and never will, because there is only one of it.

The second rung uses observable inputsNumbers that come from outside the person doing the valuation and can be checked by someone else: quoted prices, published deal terms, rents actually agreed on comparable premises. about similar things, adjusted for the differences. If three plywood makers of a comparable size changed hands in the last two years and the terms are known, those deals say a great deal about Sohan Ply, so long as the adjustment for what differs is made honestly: they were bigger, or they had two plants, or they sold into a different set of customers. Each adjustment is a decision somebody made.

The third rung builds an estimate from what the thing itself earns, and by the time that rung is reached every input is a choice, so judgement has grown from almost none to a great deal. Which year's earnings serve as the starting point? Are the one-off items stripped out? What multiple do the earnings deserve? What discount applies to a stake that cannot decide anything? None of those questions has a checkable outside answer. Each of those questions has a defensible answer, and a defensible answer is a different thing from a checkable one. The honest practice is to state each choice next to the number it produced, so a reader can disagree with one input without throwing away the whole exercise.

The ladder does not say that a step three number is unreliable, or that only step one counts. The ladder says that the further down the answer comes from, the more of it is somebody's reasoning rather than somebody's transaction, and therefore the more of the reasoning has to be shown. A step three number with its assumptions written out beside it is far more useful than a step three number quoted bare.

Where does judgement enter, and how much of the answer is it?

Three levers do most of the work at Sohan Ply, and each one is a place where two honest, competent people can pull in different directions without either of them lying. The first is which earnings figure to start from. Sohan Ply's operating profit before depreciation this year is Rs 21,00,00,000, but that year contains a Rs 2,20,00,000 insurance recovery for a fire in the previous year and a Rs 90,00,000 one-time cost of relocating a finishing line. Take the recovery out because it will not repeat, and add the relocation cost back for the same reason, and the recurring figure is Rs 19,70,00,000. A buyer paying for future earnings says the Rs 19,70,00,000 is the honest starting point. A seller says the fire was real bad luck and the relocation was real investment, so this year's Rs 21,00,00,000 is what the plant genuinely earns.

The second lever is the multiple. If comparable private plywood businesses have been changing hands at somewhere between five and a half and six and a half times operating profit before depreciation, that is a range, not a number, and picking the point inside it is judgement about how good this particular business is. The third lever is what a minority stakeA holding too small to decide anything on its own: it cannot appoint the managers, set the dividend or force a sale. Buyers usually pay less per share for one than for a controlling holding. is worth. Deodar Growth Partners would hold 20 per cent. A twenty per cent holder cannot set the dividend, cannot replace the Managing Director and cannot force a sale, so Deodar argues that its shares are worth less per share than shares that carry control.

Two honest people applying the same three levers to the same audited figures can land almost thirty per cent apart, and that spread is the measurement, not a failure of it. The idea that a spread can itself be the answer is harder to accept than anything else in valuation, and it repays sitting with. Judgement does not mean anything goes. Every lever above has a defensible range and a silly range, and a valuation that starts from a year with a fire recovery in it, applies nine times, and gives no discount for a stake that cannot control is not a difference of opinion. Such a valuation is wrong on three counts at once, and any competent reader can say which.

What is a fair value for 20 per cent of Sohan Ply?

Both sides, built properly, put the whole subject into one worked case. Deodar Growth Partners starts from the recurring Rs 19,70,00,000, applies 6.2 times, which sits inside the comparable range and slightly above its middle, and reaches Rs 1,22,14,00,000. Deodar then takes off Rs 12,14,00,000, just under ten per cent, for a stake that cannot control anything, and lands at Rs 1,10,00,00,000 for the whole business. Twenty per cent of that is Rs 22,00,00,000, and that is the offer on the table.

Sohan Malhotra builds it differently and says so openly. He starts from the reported Rs 21,00,00,000, applies 6.5 times because he thinks the laminate line makes his business better than the average comparable, and reaches Rs 1,36,50,00,000. He then adds Rs 3,50,00,000 for a strip of surplus land beside the plant that the business does not use and could sell tomorrow, giving Rs 1,40,00,00,000 for the whole and Rs 28,00,00,000 for the fifth Deodar wants. Both builds are made of the same audited figures and the same publicly known comparable deals, and they differ by Rs 30,00,00,000, which is about twenty seven per cent of the buyer's own number.

So where is the fair value? Take the comparable range on its own terms. At five and a half to six and a half times the reported Rs 21,00,00,000, the range runs from Rs 1,15,50,00,000 to Rs 1,36,50,00,000, which puts 20 per cent between Rs 23,10,00,000 and Rs 27,30,00,000. Both parties sit just outside that band, in opposite directions and by roughly similar distances: the buyer Rs 5,50,00,000 below the bottom, the seller Rs 3,50,00,000 above the top. The symmetry is not a coincidence and it is not dishonesty. Opening a negotiation is exactly what both sides are doing, and the band between them is the honest reading of fair value.

How do a lender, an investor and a household actually use fair value?

A lender uses it on the collateral, not on the business. Sohan Ply's working capital line of Rs 15,00,00,000 is secured on stock and on the dues its customers owe. The books carry those two at Rs 27,00,00,000 and Rs 30,00,00,000, a total of Rs 57,00,00,000. The bank does not lend against that. The bank asks what those items would fetch in a sale it did not choose the timing of, applies a haircutThe percentage a lender knocks off the recorded amount of security before deciding how much it will lend against it, to allow for what the security would actually fetch in a hurried sale. to each, and works with what is left. On illustrative haircuts of a quarter off the customer dues and sixty per cent off the stock, the lendable base falls to Rs 33,30,00,000, and the Rs 15,00,00,000 drawn is 45 per cent of that. The whole gap between Rs 57,00,00,000 and Rs 33,30,00,000 is one institution's estimate of fair value under conditions that are anything but orderly.

An investor considering a stake uses it the way Deodar Growth Partners just did, and then uses it again every year afterwards. Once the stake is bought, somebody has to say each year what the stake is now worth. The people whose money Deodar invested want a statement. Nothing traded, no price exists, and yet a number must be produced and defended. The annual re-estimate is fair value measurement in its purest form, and it is why the wording written next to the number always matters more than the number.

A household meets fair value more often than it realises, and Sohan Ply's own accounts contain one instance. The Rs 2,20,00,000 insurance recovery came from an assessor deciding what the burnt assets were worth on the day of the fire, not what they had cost when new and not what they were carried at in the books. The same logic runs a flat sale, where a broker quotes a range from what similar flats in the same block have fetched, and a claim on a stolen chain, and the value a court puts on damaged goods. In every one of those, somebody stands on the ladder, reaches for the highest rung available, and writes down the assumptions. Practitioners almost never ask what a thing is worth; they ask what it would fetch, from whom, under what conditions, and on what evidence.

Sohan Ply and Boards Private Limited, Sohan Malhotra, Ritu Chandran and Deodar Growth Partners are invented.
Educational material. Not advice on any investment, tax, budget or market position.