In the first level of the certification you learned that a share price says nothing on its own and that the basic yardstick of expensiveness is the price-to-earnings ratio. But P/E is only the beginning. An analyst needs tools that also work for leveraged companies, loss-making growth stocks, or when comparing businesses with completely different capital structures.
In this article we walk through the level-two valuation toolkit: enterprise value and the multiples derived from it, PEG, the difference between the forward and trailing view - and above all a way of thinking that lets you break a stock's return down into its true sources.
Enterprise value: what the whole company costs
Market cap tells you what all of a company's shares cost. But if you were buying the whole company, you would take on its debts too - and in turn gain the cash sitting in its accounts. That is exactly what enterprise value captures:
EV = market cap + debt - cash
Example: a company has a market capitalization of 80 billion, debt of 30 billion and cash of 10 billion. EV = 80 + 30 - 10 = 100 billion. That is the actual price a buyer would pay for the entire business.
An important consequence follows: two companies with the same market cap can be priced completely differently. A debt-free company with a cash cushion is, at the same market capitalization, cheaper than a company leveraged up to its ears. Market cap does not see that difference, EV does - which is why differently leveraged companies are compared through EV, never through share price alone.
Debt has one more consequence valuation must not overlook: leverage at the equity level. Debt is the creditors' fixed claim - when the value of the whole company moves, the entire change lands on the shareholders. A company with an EV of 100 billion and debt of 40 billion has equity worth 60 billion; if the value of the business falls 20% to 80 billion, the debt stays at 40 and 40 billion is left for the shares - a drop of a third. The same EV move shows up amplified on leveraged equity, in both directions. A quick gauge of how bearable the debt is comes from net debt/EBITDA: how many years of EBITDA the company would need to repay its net debt. The balance sheet article covers it in depth; in valuation keep it at hand as a check on how much leverage you are buying into the multiple.
EV/EBITDA: a comparison debt cannot distort
P/E measures the price of the shares against a profit already affected by interest, depreciation and taxes. A leveraged company pays higher interest, has lower net income, and its P/E therefore looks different from its operating reality. EV/EBITDA sidesteps this problem: the numerator is the price of the whole company including debt, the denominator is operating profit before interest, taxes, depreciation and amortization.
Example: a company with an EV of 100 billion and EBITDA of 12.5 billion trades at EV/EBITDA of 8. You can fairly compare this number with a competitor financed in a completely different way.
When to reach for EV/EBITDA instead of P/E: when comparing companies with different leverage or different accounting regimes, and when considering what a strategic buyer would pay for the business. It has a blind spot, though: it ignores depreciation. In capital-intensive industries - airlines, telecoms, mining - depreciation is a real cost, because the machines and networks genuinely wear out and must be continuously replaced; depreciation there roughly approximates permanent capex. The more honest multiple for such businesses is therefore EV/EBIT, which keeps depreciation in the number: EV/EBITDA at a capital-heavy company sweeps its largest cost under the rug, and the company then looks cheaper than it is.
For loss-making growth companies without even positive EBITDA, EV/Sales remains - the ratio of company value to revenue. Even that makes sense only in context: EV/Sales of 5 can be sober for a software company growing 40% a year and absurd for a retailer with stagnating revenue. And because revenue is not profit, always read revenue multiples together with margins: a P/S of 10 at a company that can sustainably reach a 25% net margin implies a future P/E of 40; the same P/S at a business with an achievable margin of 5% implies a P/E of 200. A high revenue multiple is bearable only where high margins are realistically achievable.
P/B: the multiple for banks and insurers
A special place in the toolkit belongs to P/B - the ratio of market price to the book value of equity. For an ordinary company it says little today: a software company's value rests on brand, data and people - assets that barely appear on the balance sheet. For banks and insurers it is, on the contrary, the most useful multiple: their assets are financial - loans, bonds, investments - and are carried on the balance sheet close to market value, so book equity there genuinely approximates the value of the business. A P/B below 1 at a bank says the market trusts the capital less than the accounting does (it expects loan losses or weak returns); a P/B well above 1 has to be earned with a high return on equity.
PEG: the multiple in the context of growth
A higher P/E does not automatically mean a more expensive stock. If profit is growing fast, today's high multiple dissolves: a company at a P/E of 30 that doubles its profit in three years trades, at an unchanged price, at a future P/E of 15. The market therefore grants growth companies higher multiples for good reason - what is being paid for is future profit, not today's.
This relationship is quantified by PEG: P/E divided by the expected annual earnings growth in percent. A stock with a P/E of 20 and 10% growth has a PEG of 2. A stock with a P/E of 30 and 30% growth has a PEG of 1 - relative to its growth it is cheaper, despite the higher P/E. Just remember that PEG rests on an estimate of future growth, and estimates tend to be optimistic. Treat it as quick context, not as a verdict.
Forward, trailing, and what rerating is
Trailing P/E works with profit over the last twelve months - it is factually documented, but it looks backward. Forward P/E works with the profit estimate for next year. At a growing company the forward is naturally lower than the trailing; the size of the gap says how much growth the market already has in the price.
A share price can always be written out as profit times multiple. That is why profit can grow 20% and the stock still fall - it is enough for the market to reprice the multiple from 25 to 15. This is called multiple compression; the opposite move is expansion, and together they are referred to as rerating.
From this follows the most useful equation of the second level - return decomposition: a stock's long-term return consists of earnings growth, multiple change and dividends. Example: profit grows 50% over five years, but P/E meanwhile falls from 20 to 16, i.e. by 20%. The price rises only 20% - part of the earnings growth was eaten by multiple compression. Whoever buys a great company at an extreme multiple is betting that the rerating will not come.
A useful complement is the earnings yield, the inverted P/E: a stock with a P/E of 25 carries an earnings yield of 4%. It is precisely through this that stocks are compared with bond yields - and why rising interest rates press hardest on the multiples of growth stocks. Their profits lie far in the future, and higher rates strip them of value when converted to today.
P/FCF and the traps of cheap numbers
Profit is an accounting quantity. P/FCF - price to free cash flow - diverges from P/E wherever a company reports a nice profit but the cash disappears into high capex, or where a large part of compensation is paid in shares. When P/FCF is significantly higher than P/E, ask why; cash is harder to fake than profit.
But watch the same trap in the opposite direction: stock-based compensation (SBC) is a non-cash cost, so it does not reduce free cash flow - it is paid for by diluting shareholders, not with cash. A company that pays a large share of wages in stock can therefore show a flatteringly low P/FCF while the real bill lands on shareholders through a rising share count. At SBC-heavy companies, always read P/FCF together with the share count trend.
The second rule of cheap numbers: compute the multiple from normalized earnings. A one-off gain from selling a division or an exceptionally strong year inflates the denominator and the P/E comes out optically low - but the windfall will not repeat. Strip one-off items before computing the multiple and ask what profit the company can earn repeatably; a P/E built on an exceptional year is advertising, not valuation.
And beware the value trap. First the definition: a value stock is the stock of a mature business trading at low multiples, from which the market expects only modest growth - the counterpart of a growth stock, where a high multiple is paid for rapid expansion. A low multiple alone guarantees no value, though: a stock at a P/E of 5 looks like a gift until you realize the market prices the future. For a business with falling revenue and margins, the low multiple is often accurate - profit will be lower next year, so the true forward P/E is higher than today's number shows. A cheap number, an expensive reality.
Premiums and discounts: what else moves the multiple
Two influences on multiples that beginners overlook. The first is the predictability premium: a stable, well-forecastable profit deserves a higher multiple than an equally large profit that swings with the cycle or hangs on a single contract. The market pays for certainty - which is why a staple-foods producer trades richer than a steel mill with the same average profit.
The second is the acquisition premium: when one company buys another, it typically pays well above the market price. It is buying control - the right to run the company, fold it into its own business and harvest synergies - and it does not have to share that with the market. That is why a share price jumps toward the offer once a takeover is announced, and why the undisturbed market price is not the ceiling of what a business can be worth to the right owner.
Relatively cheap, absolutely expensive
Multiples are relative valuation: they say what you are paying for a company compared with the others. They have a blind spot, though - when an entire sector is overvalued, a stock can look fair against its peers and still be expensive in absolute terms. "Cheaper than the other tech stocks" in a bubble only means "will burst somewhat less". The counterweight is an estimate of intrinsic value: what the business will actually earn economically, regardless of what its neighbours happen to trade at. An indicative estimate of the fair price for thousands of stocks is available in the Fair Price Index.
And because every estimate of value is imprecise, the craft includes a margin of safety, as Benjamin Graham described it: buy at a significant discount to your own estimate of value, so that an estimation error does not turn into a loss. Whoever estimates value at 100 and buys at 95 is betting on their own infallibility; whoever buys at 70 survives even being wrong by a fifth.
Test yourself
Valuation is the largest area of the Senior Analyst exam - the second of four levels of the Bulios certification. Thirty questions, twelve minutes and a bar of 100% correct answers will test whether you truly command the multiples or merely know them. The earned certificate appears on your profile as proof that your analyses stand on craft.