A multiple is not a shortcut past valuation. It is somebody else's valuation, compressed into one number, and using it means inheriting every assumption inside it.
12 min read · Level 3
Multiples are usually presented as the practical alternative to modelling, for people who have neither the time nor the taste for a spreadsheet. That framing is backwards, and it is worth resisting. A multiple is not simpler than a discounted cash flow - it is a discounted cash flow that somebody else already performed, with the assumptions removed and the answer left behind. Using one is therefore an act of delegation, and the question it raises is whether you trust the delegate. Sometimes you should: a deep market of informed buyers pricing similar assets has considered things you will not think of. Often you should not, because the peer multiple was set in conditions that no longer hold, by people who were not buying the specific business in front of you. The discipline is not to choose between the two methods but to be conscious of which assumptions you have authored and which you have inherited, because inherited assumptions are the ones nobody ever checks.
Relative valuation asks what similar assets are changing hands for, and applies the same ratio to the one in front of you. The ratio is called a multiple: a price divided by something the business produces - earnings, cash flow, revenue, square feet, subscribers.
There are two families and mixing them is the most common error in the whole subject. Enterprise multiples put the value of the WHOLE business - what both lenders and owners have claims on - over a figure calculated before interest, such as EBITDA or operating profit. Equity multiples put the value of the OWNERS' share over a figure calculated after interest, such as net profit. The two answer different questions and are not interchangeable.
The bridge between them is net debt. Enterprise value minus net debt is equity value, and every relative valuation has to cross that bridge explicitly at some point.
Because a multiple is the fastest way to find out whether your own view is unusual. If you have concluded something is worth twenty times its earnings and every comparable business sells for nine, you have either seen something the market has not, or made an arithmetic error, and it is worth establishing which before committing money.
And because in private transactions it is frequently the only method anyone actually uses. Buyers and sellers of small businesses negotiate in multiples, so understanding what a multiple contains is not an academic exercise - it is understanding the language the price will be argued in.
Educational, not personalised financial advice. Nothing in the Investor Academy is a recommendation to buy, sell or hold anything — what is right for you depends on your own circumstances and the rules where you live.
Harbour Press is a small publishing and print business. We will use it here and in the next lessons.
EBITDA 2,000,000 Depreciation 400,000 Operating profit 1,600,000 Interest (6% on 5,000,000) 300,000 Profit before tax 1,300,000 Tax at 25% 325,000 Net profit 975,000
Borrowings 5,000,000, cash 1,000,000, so net debt is 4,000,000.
Four comparable businesses have recently changed hands at enterprise value to EBITDA multiples of 7.0, 9.5, 11.0 and 14.5 times. The mean is 42.0 divided by 4, or 10.5 times; the median is the midpoint of 9.5 and 11.0, which is 10.25 times.
Applying them to Harbour Press's 2,000,000 of EBITDA gives an enterprise value of 14,000,000 at the bottom, 20,500,000 at the median and 29,000,000 at the top. The top is 29,000,000 over 14,000,000, or 2.07 times the bottom - a range again, from a method that is usually quoted as a single figure.
Now cross the bridge. Equity value is enterprise value less the 4,000,000 of net debt: 10,000,000 at the bottom, 16,500,000 at the median, 25,000,000 at the top.
Here is the error that bridge exists to prevent. Someone who applies 10.25 times to EBITDA and reports 20,500,000 as what the owners' shares are worth has overstated it by the whole 4,000,000 of net debt. That is 20,500,000 over 16,500,000, or 24.2% too high, and the mistake is invisible because the arithmetic is correct - it is the label on the answer that is wrong.
Now why the two families matter. Consider two other businesses with identical operations, each earning 2,000,000 of operating profit, taxed at 25%.
Company X has no borrowings. Profit before tax 2,000,000, tax 500,000, net profit 1,500,000. Company Y has 10,000,000 of borrowings at 6%. Interest 600,000, profit before tax 1,400,000, tax 350,000, net profit 1,050,000.
Suppose the market values the underlying operations of each at an enterprise value of 20,000,000 - the same, because the operations are the same. X has no debt, so its equity is worth 20,000,000. Y's equity is worth 20,000,000 less 10,000,000, which is 10,000,000.
On a price-to-earnings basis, X trades at 20,000,000 over 1,500,000, which is 13.3 times. Y trades at 10,000,000 over 1,050,000, which is 9.5 times. Y looks 28.6% cheaper.
It is not cheaper at all. On enterprise value to operating profit both are at 20,000,000 over 2,000,000, which is exactly 10.0 times. The entire apparent discount is Y's borrowing, and a buyer who screened on the price-to-earnings ratio alone would have concluded the more indebted business was the better bargain.
The first mistake is the family error above: applying an enterprise multiple and reading the result as what the shares are worth. Always subtract net debt, and say out loud which side of the bridge you are standing on.
The second is calling something a comparable because it is in the same industry. A genuine comparable shares growth rate, margin, capital intensity and customer concentration. Two businesses that both print books, one of which has a single customer providing 60% of revenue, are not comparable in any sense that matters.
The third is using the mean when the sample is small. With four observations, one outlier moves the mean substantially and the median barely at all - here the mean of 10.5 and the median of 10.25 are close, but a single peer at 25 times would pull the mean to 13.4 while the median stayed at 11.0.
The fourth is forgetting that a multiple is a moment. It records what buyers were willing to pay under the conditions of that month, including the cost of borrowing and the general mood, and those conditions are not a property of the business you are valuing.
The fifth is revenue multiples used on businesses that do not yet make money. A price expressed as a multiple of revenue contains an unstated assumption about the margin the business will eventually reach, and because the assumption is unstated it is never examined.
If you are buying the whole business including its debt, use an enterprise multiple over a pre-interest figure. If you are buying shares in something whose debt stays where it is, an equity multiple is defensible - but compute the enterprise version too, because it is the one that is comparable across companies with different borrowing.
Write, for each candidate peer, the growth rate, the margin, the capital intensity and the customer concentration. Keep the ones that match on most of those and discard the ones that only match on what the business sells. A set of three genuine comparables is worth more than a set of fifteen assembled by sector code.
With a small sample the median is far more stable, and the distance between the lowest and highest peer is the real output. A tight cluster means the market agrees about this kind of business; a wide one means it does not, and your answer inherits that disagreement.
Enterprise value less net debt equals equity value. Write both, labelled, every time. This one habit prevents the single most expensive mistake in relative valuation, and it takes one line.
A business trading at 14.5 times EBITDA is being priced for growth, durability or scarcity that a business at 7.0 times is not. Write down which of those you believe applies to your asset. If you cannot name it, you are borrowing somebody else's optimism without knowing what it was about.
Transactions from a period of cheap borrowing are not evidence about a period of expensive borrowing. Date every peer and, if the dates cluster in conditions unlike today's, say so in the conclusion rather than quietly using the numbers anyway.
When you need a fast sanity check on a value you reached another way, when negotiating a private transaction where multiples are the language being used, and whenever there are genuinely similar assets with observable prices.
Avoid it as the primary method where there are no real comparables - a business with an unusual model, a one-off asset, or a market where the only recent transactions happened under conditions that no longer hold. A multiple borrowed from a poor match is not a weaker answer; it is a confident answer to the wrong question.
A shape that appears frequently in valuations produced under time pressure is a comparable set chosen after a target multiple was already in mind. The mechanism is not dishonest and is rarely conscious: peers whose multiples look wrong are examined more carefully than peers whose multiples look right, and closer examination almost always turns up a reason to exclude a business. Since no two companies are truly alike, a reason can be found for any peer, and the set that survives is the set that agreed.
Write the inclusion criteria before you look at any multiple, and apply them to every candidate with the same care. A peer excluded for a reason you would not have noticed had its number been different is a peer you excluded for its number.
In small-business transactions a multiple is usually quoted as a bare number - 'businesses like this go for four times' - with nothing attached about what was included in the earnings figure, whether the seller stayed on, what the working capital arrangement was, or how much of the price was deferred or contingent. Two deals at the same headline multiple can transfer very different amounts of money on very different terms. The headline travels between conversations; the terms do not.
Before using a quoted private multiple, ask what the denominator included and how the price was actually paid. A four times multiple with half the consideration deferred and contingent is not four times.
Create a free account to read the rest and apply it to your own investing.