Valuation ratios put a price in relation to something about the business. Each measures a different thing, and each fails in a different situation. Used without understanding that, they are worse than useless — they look precise.
Price-to-earnings (P/E)
The share price divided by earnings per share, or the market value divided by net income.
What it says: how much the market is paying per unit of accounting profit.
When it misleads:
- Earnings can be negative, making the ratio meaningless.
- Earnings are an accounting figure. Depreciation policy, one-off items, and revenue recognition all change it without changing the business.
- The trailing and forward versions differ. Trailing uses actual past earnings; forward uses estimates, which may be wrong.
- Comparisons across industries fail. A stable utility and a cyclical miner are not comparable at the same multiple.
Price-to-book (P/B)
Market value divided by accounting net assets — roughly, what the market pays relative to the recorded worth of the assets.
What it says: something about how the market values the assets in place versus the returns expected from them.
When it misleads: increasingly often, as economies shift toward businesses whose main assets — software, brands, trained staff — are not on the balance sheet at all. A high P/B may reflect an asset-light model rather than an expensive stock.
Dividend yield
Annual dividend per share divided by share price.
What it says: the cash return paid out to shareholders.
When it misleads:
- A high yield can be a warning. If the price has fallen because the dividend is at risk, the yield looks attractive precisely when it is least reliable.
- It ignores buybacks, which are another route for returning cash.
- A company that pays nothing is not returning nothing — it may be reinvesting at a higher rate of return, which can be better for a taxable investor.
Enterprise value multiples
EV/EBITDA and its relatives relate the value of the whole business — equity plus debt, minus cash — to its operating profits.
What it says: useful when comparing companies with different capital structures, because it sits above the debt.
When it misleads: EBITDA is a figure before interest, tax, depreciation and amortisation. For a capital-intensive business, depreciation is a real economic cost, and ignoring it flatters exactly the companies that need the most capital to operate.
The traps in using them
- Comparing across sectors. Different business models have different structural multiples. The comparison is only meaningful within a comparable group.
- Treating a low multiple as cheap. A low multiple can reflect a deteriorating outlook, a cyclical peak in earnings, or an accounting issue. It is a question, not an answer.
- Ignoring the denominator's quality. A ratio is only as sound as the number beneath it.
- Single-year snapshots. Cyclical businesses look cheapest at the peak of the cycle, when earnings are highest, and most expensive at the trough — the opposite of what the ratio appears to say.
What ratios are actually for
A valuation ratio is a starting point for a question, not a conclusion:
- Why is this multiple what it is? What does the market appear to believe?
- What would have to be true for that belief to be wrong?
- How does it compare with the company's own history, and is the business the same as it was then?
None of these are answered by the ratio itself. Eonvera's research combines quantitative screening with fundamental analysis for this reason: a screen finds candidates; only the work behind the number decides anything.