Skip to content

P/E Ratio vs Its Own History: How to Read It

By Alexander Chepakovich · Updated

What the P/E measures

The price-to-earnings ratio is the price of a share divided by its earnings per share — how many dollars the market is paying for each dollar of annual profit. A P/E of 20 says the stock costs twenty years of current earnings; equivalently, the earnings yield — E/P, the reciprocal — is 5%, which you can compare directly with bond yields.

The ratio's strength is its simplicity and universality: almost every profitable company has one, and it is the most widely quoted valuation measure in the market. Its weakness is the same simplicity — a single number with no context tells you almost nothing.

Why a bare P/E misleads

A P/E of 25 is expensive for a regulated utility growing 3% a year and cheap for a software company compounding at 25%. The ratio only acquires meaning against three references: the business's own history, its peers, and its growth. Quoting a stock's P/E without those references is like quoting its price without the currency.

The denominator is also unstable. Reported EPS for a single year includes one-off charges, cyclical peaks and tax effects. A company at the top of a business cycle shows a high E and therefore a low P/E — the market's classic trap, where the stock screens "cheap" precisely when the earnings are about to fall.

The own-history comparison

The most underrated P/E comparison is against the stock's own past. Over fifteen years a company traces out a P/E range that reflects how the market has actually priced this specific business through different conditions — not an abstract industry multiple. Trading near the bottom of its own range suggests the market is pricing it cheaply relative to its own norm; near the top, richly.

Hypothetical example. Suppose a company's P/E has ranged between 12 and 28 for fifteen years, averaging 18. At today's price it trades at 14 — near the bottom of its own history. That placement is more informative than "14 is low": it says the market is pricing this company near its own historical floor. The signal is worth investigating, though it says nothing by itself about why — the reason might be deterioration the range does not capture.

The peer comparison

The second reference is peers — companies in the same industry with similar economics. A stock trading at half the multiple of its comparables is either a bargain or correctly priced for worse prospects; the discount is the starting point for the question, not the answer to it.

Peer comparisons fail when the comparison set is sloppy. A bank and a software company do not share a meaningful P/E, and even within an industry, business quality varies enormously. The honest peer set is a handful of companies with genuinely similar economics, not a sector-average multiple.

Normalizing the earnings

Because one year's EPS is noisy, serious P/E work uses normalized earnings — an average over a cycle, or Graham's preferred multi-year average from Security Analysis. A ten-year average earnings basis (the idea behind the cyclically adjusted P/E) flattens the cycle and gives the ratio a denominator that survives booms and busts.

Forward P/Es go the other way — dividing by analyst forecasts — and inherit all the optimism bias of the forecasts themselves. Trailing normalized earnings are less exciting and more honest.

P/E versus intrinsic value

A P/E is a relative measure: it compares price with earnings but says nothing about what the earnings themselves are worth. Two companies with identical P/Es can have wildly different intrinsic values if their growth, capital needs and risk differ — the multiple cannot see any of that.

That is why multiples work best as corroboration rather than method. When a stock's P/E sits near the bottom of its own range and its DCF-based intrinsic value is well above the price — the kind of double signal the InvestorsCraft stock pages show — the case is stronger than either measure alone. The multiple provides the context; the intrinsic estimate provides the answer.

Try it

See it on real stocks

Frequently asked questions

What is a good P/E ratio?

There is no universal good number — a P/E only means something relative to the company's growth, quality, and its own history. Graham's defensive ceiling of 15 remains a useful conservative anchor, but the right question is 'what is normal for this business?'.

Why can a low P/E be a warning?

A low P/E on peak earnings is the classic value trap: if profits are about to fall, the E will drop and the apparently cheap multiple was an illusion. Cyclicals routinely screen cheapest at the top of their earnings cycle.

What does it mean when a stock has no P/E?

Negative or near-zero earnings produce a meaningless P/E — the ratio divides by earnings, so a company losing money has no P/E to read. Other valuation methods have to take over.

This guide is educational content, not investment advice. Worked examples use hypothetical numbers unless labelled otherwise.