Valuation multiples

A multiple compares a company's value with a measure of what it earns or owns. Multiples are the quickest way to value a business, and they are everywhere: in pitch books, research notes, deal announcements and the financial press. If a headline says a company was bought for 10 times EBITDA, you should know exactly what that means and whether it sounds cheap or expensive.

The idea

If similar companies trade at similar multiples, you can value one company by looking at the others. This is called relative valuation, or trading comparables (comps). It answers "what is the market paying for businesses like this?" rather than "what is this business worth in theory?", which is the question a DCF tries to answer.

A multiple packs a lot of assumptions into one number. A company on a high multiple is usually expected to grow faster, earn higher margins or carry less risk than one on a low multiple. When you compare multiples, you are really comparing those expectations.

The consistency rule

The most important rule is that the value on top must match the figure underneath. Ask who the figure belongs to.

• Figures before interest, such as revenue, EBITDA and EBIT, belong to all capital providers, so they pair with enterprise value
• Figures after interest, such as net income and earnings per share, belong to shareholders only, so they pair with equity value or the share price

So EV/EBITDA and P/E make sense, but equity value/EBITDA does not. It would make a company look cheaper simply because it has more debt, since debt reduces equity value without changing EBITDA.

The common multiples

EV/EBITDA is the workhorse because it ignores differences in capital structure, tax and depreciation policy. P/E is the one most investors quote, and its inverse, earnings per share divided by share price, is the earnings yield. A P/E of 20 means an earnings yield of 5%.

Banks are different. Debt is their raw material rather than a way of financing operations, so enterprise value is not meaningful for them. Analysts use P/E and price to book, or price to tangible book, instead.

Trailing and forward

A multiple can use last year's figures, known as LTM (last twelve months) or trailing, or a forecast, known as NTM (next twelve months) or forward. Forward multiples are lower for a growing company, because next year's earnings are higher. Always say which one you mean, and never compare a trailing multiple for one company with a forward multiple for another.

Valuing a company from its peers

• Choose peers that are genuinely similar: same sector, similar size, growth, margins and geography
• Gather each peer's enterprise value, equity value and the relevant earnings figures
• Clean the figures: remove one-off items and make sure every company is treated consistently
• Work out each peer's multiples and look at the range, especially the median
• Apply the chosen multiple to your company's figure to get an implied value

The median is usually more useful than the average, because one unusual company can drag an average a long way. In practice you show a range, say between the 25th and 75th percentile, rather than a single number.

Precedent transactions work the same way, but use the multiples paid in past acquisitions of similar companies. They reflect control premiums and the market conditions at the time of each deal, so check the dates and whether the buyers were strategic or financial.

IFRS 16 and other traps

Since IFRS 16 put most leases on the balance sheet, lease costs sit below EBITDA as depreciation and interest. That raises EBITDA, particularly for retailers, airlines and restaurant chains with lots of property leases. If you use post-IFRS 16 EBITDA, include lease liabilities in net debt so both sides of the multiple are consistent. Some analysts prefer to strip leases out of both instead. Either works if you are consistent.

Other traps include comparing companies with different year ends (you calendarise them to the same period, covered in the superday lesson on comps), using a P/E for a company with tiny or negative earnings, where the multiple becomes meaningless, and forgetting that a takeover usually includes a control premium, so multiples paid in past deals are normally higher than trading multiples.