Stocks
Stock Valuation Basics: P/E, P/S, P/B, EV/EBITDA and PEG
One company, five multiples, the same numbers running through all of them, and the specific thing each ratio cannot see.
Company A trades at $80.00 with 130 million shares outstanding. Market capitalization is 130m x $80 = $10.4bn. Everything below comes out of the table underneath, so the five ratios stay comparable to each other, and most explanations of this demonstrate each ratio on a different company, which is how they hide the awkward parts.
| Item | Value | Per share |
|---|---|---|
| Share price | $80.00 | |
| Shares outstanding | 130m | |
| Market capitalization | $10.4bn | |
| Revenue (last 12 months) | $6.5bn | $50.00 |
| Net income | $520m | $4.00 |
| EBITDA | $1.18bn | $9.08 |
| Shareholders equity (book value) | $3.25bn | $25.00 |
| Total debt | $2.0bn | |
| Cash | $0.6bn | |
| Expected earnings growth | 15% a year |
A multiple is a price quoted in units of something the business produces. Each one answers a different question. Each one is blind to something specific, and the blind spot is the part worth memorizing.
P/E
P/E = share price / earnings per share. For Company A, $80.00 / $4.00 = 20.0.
You are paying twenty times one year of current profit. Invert it and you get the earnings yield: $4.00 / $80.00 = 5.0%. That is the profit the business throws off for every dollar you put in at this price, and the inversion is why multiples compress when bond yields rise. A Treasury yield competes with that 5% without the business risk. How interest rates affect stocks traces the rest of that chain.
What it misses is debt. Two companies with identical earnings and identical P/E ratios are not identical if one carries $2bn of borrowings and the other carries none. The ratio prints the same number for both. It also stops working when earnings are near zero or negative. The number goes enormous or undefined.
P/S
P/S = market cap / revenue. For Company A, $10.4bn / $6.5bn = 1.6. Price to sales works where there are no profits to divide by: early-stage growth companies, and cyclicals in the trough of their cycle, and revenue is also harder to bend than earnings, because it sits above every judgment call about depreciation, provisions and capitalized costs.
The blind spot is margin, and it is a large one. A software company converts 30% of revenue into profit. A grocery chain converts 2%. Both have revenue. A P/S of 1.6 means something completely different for each. Using this ratio requires an assumption about the margin the business eventually reaches, and that assumption is doing all the work while the ratio takes the credit.
P/B
P/B = share price / book value per share. For Company A, $80.00 / $25.00 = 3.2. Book value is assets minus liabilities as recorded on the balance sheet. This was the original value measure and it still does its job where the balance sheet reflects the business: banks, insurers, property companies, anywhere the assets are financial and marked somewhere near their worth.
Everywhere else it has aged badly. Accounting rules force research, software development and brand building to be expensed in the year they happen, so a company whose main asset is intellectual property carries almost none of it on the balance sheet. A P/B of 3.2 on Company A tells you one thing. The assets that matter are invisible to the ratio. People still screen on it and then wonder why every result is a bank.
EV/EBITDA
Enterprise value is what it would cost to buy the whole business: market cap + debt - cash. For Company A, $10.4bn + $2.0bn - $0.6bn = $11.8bn.
EBITDA is earnings before interest, tax, depreciation and amortization. It is a rough proxy for operating cash generation. EV / EBITDA = $11.8bn / $1.18bn = 10.0.
This is the acquirer’s multiple. It compares businesses before their financing choices, which means a heavily indebted company and a debt-free one can sit in the same column. P/E cannot do that.
PEG
PEG = P/E / expected annual earnings growth rate. For Company A, 20.0 / 15 = 1.33. PEG exists because a P/E of 20 is expensive for a business growing at 3% and cheap for one growing at 40%, and a PEG near 1.0 is the conventional marker for a multiple roughly in line with growth.
The weakness is entirely in the denominator. That growth rate is a forecast. Usually an analyst consensus running three to five years out. Multi-year forecasts drift. Watch how far the answer travels: at 20% growth the same P/E gives 20 / 20 = 1.0, and at 10% it gives 2.0. Same stock, same day, two verdicts. Find out which growth rate is embedded before you quote the output, and test the compounding yourself in the CAGR calculator.
Which one to reach for
| Situation | Primary multiple | Why |
|---|---|---|
| Profitable, stable, low debt | P/E | Direct and comparable across the sector |
| Different debt loads or tax rates | EV/EBITDA | Compares the business before financing |
| No profits yet | P/S or EV to gross profit | Revenue exists where earnings do not |
| Banks, insurers, property | P/B | The balance sheet reflects the business |
| High growth, profitable | PEG plus a DCF | Anchors the multiple to the growth paying for it |
| Heavy capital spending | Price to free cash flow | Survives what EBITDA ignores |
Cyclicals invert the usual reading, and this catches people every cycle. A steelmaker or a homebuilder shows its lowest P/E at the top, when earnings are at maximum and about to fall, and its highest P/E in the trough. Screening a cyclical for a low P/E is a mechanism for buying peaks. Compare it against its own ten-year average multiple instead.
Where multiples stop working
A multiple compresses an entire business into one number, and everything it left out is still there: two companies on the same P/E can have opposite debt maturities, opposite margin trends, and one of them can have a customer worth a third of revenue about to walk.
Multiples also move together for reasons unrelated to any of the companies. When the discount rate rises, every long-duration multiple falls at once, and the cheapest stock in a de-rating market still falls with it. That is the rotation described in growth vs value stocks. No amount of stock picking exempts you from it.
The alternative is building the cash flows yourself. A discounted cash flow model projects free cash flow forward and discounts it back at a rate reflecting risk, which forces every assumption out into the open where you can argue with it. It is also sensitive enough to the terminal growth rate and the discount rate that small changes swing the answer by a third, so treat the output as a range and never as a price target. The DCF calculator shows you that sensitivity directly.
Run these ratios on live names with the stock screener, get the input numbers right by reading the source in how to read an earnings report, and put valuation back in its proper place among the other checks with how to analyze a stock. The arithmetic on the earnings-based ones is in the P/E ratio calculator.
Frequently asked questions
What is a good P/E ratio for a stock?
There is no universal number. A utility with flat earnings and a software company growing 30% a year deserve different multiples, and the same company deserves a different multiple when interest rates change. The useful comparison is against the company's own history and its direct competitors, with the growth rate attached.
Why use EV/EBITDA instead of P/E?
Enterprise value includes debt and subtracts cash, and EBITDA sits above interest and tax, so the pair compares the whole business before financing decisions. That makes it better for comparing companies with very different debt loads or tax positions. It ignores the real cost of capital spending, which flatters asset-heavy businesses.
What does a PEG ratio of 1 mean?
It means the P/E ratio equals the expected earnings growth rate, so a company on a P/E of 20 growing earnings at 20% a year has a PEG of 1.0. Peter Lynch popularized 1.0 as a rough fair value marker. The number is only as good as the growth forecast inside it, and forecasts several years out are frequently wrong.
Can a stock with no earnings be valued?
Yes, with measures that sit higher up the income statement. Price to sales works when revenue exists and profits do not, and enterprise value to gross profit is often better because it accounts for the quality of that revenue. Both require an assumption about the margin the business will eventually earn.
Do valuation multiples predict returns?
Over long horizons, starting valuation has historically had a relationship with subsequent returns across a broad market, and the relationship is loose. Over a single year, sentiment and rates dominate. Treat a multiple as a description of what is being paid rather than as a forecast.
What is the difference between trailing and forward P/E?
Trailing P/E uses the last twelve months of reported earnings, which are facts. Forward P/E uses the next twelve months of estimated earnings, which are forecasts. Forward P/E is almost always the lower number because analysts generally expect growth, and it becomes unreliable at the exact moment earnings estimates start falling.