The P/E Ratio

Introduction

If there is one number investors reach for first when sizing up a stock, it is the price-to-earnings ratio — the P/E. It is the most quoted, most misunderstood measure in all of investing: a single figure that compresses "how expensive is this share relative to what the company earns?" into one comparable number. Used well, the P/E is a powerful first lens for valuation and comparison. Used naively — as a simple "high means expensive, low means cheap" rule — it misleads more often than it helps. The difference lies in understanding what a P/E actually represents: not a verdict, but a statement of the market's expectations.

This lesson builds on the income-statement lesson (which introduced earnings per share) and the fundamental-analysis overview. It explains what the P/E means, the trailing-versus-forward distinction, why high and low P/Es reflect growth expectations rather than simple cheapness, the useful trick of the earnings yield, and the traps that catch the unwary.

Quick Definition

The price-to-earnings (P/E) ratio is a company's share price divided by its earnings per share (EPS). It tells you how much you pay for each dollar of the company's annual earnings. A P/E of 20 means you are paying $20 for every $1 of earnings.

P/E = Share price ÷ Earnings per share.

Another way to read it: the P/E is, very roughly, the number of years of current earnings it would take to "pay back" the price you paid — a P/E of 20 implies 20 years of today's earnings. That framing instantly shows why context matters: 20 years is a lot to pay for a stagnant company, but a bargain for one whose earnings will multiply.

A Simple Calculation

The arithmetic is trivial; the interpretation is everything. A stock trading at $40 with earnings per share of $2 has a P/E of $40 ÷ $2 = 20. A stock at $30 with EPS of $3 has a P/E of 10. On the surface the second looks "cheaper" — but as we'll see, that comparison is meaningless without knowing why the market assigns each its multiple.

Trailing Versus Forward

There are two flavours of P/E, and confusing them causes real errors:

• Trailing P/E uses the past 12 months' actual, reported earnings. It is factual and reliable — these earnings really happened — but backward-looking. For a company whose fortunes are changing fast, last year's earnings may be a poor guide.
• Forward P/E uses analysts' forecast of the next year's earnings. It is forward-looking and often more relevant — you are buying the future, not the past — but it relies on estimates that can be wrong, sometimes badly. A low forward P/E built on optimistic forecasts is no bargain if those forecasts don't materialise.

Sensible investors look at both: the trailing P/E for what is solid and known, the forward P/E for where the business is heading — while treating the forecast with appropriate scepticism.

The Heart Of It: P/E Reflects Expectations

Here is the insight that separates a useful understanding of the P/E from a dangerous one. A P/E is not a measure of cheapness — it is a measure of expectations. The market assigns a high P/E to companies it expects to grow earnings strongly, and a low P/E to those it expects to stagnate or shrink. The multiple is the price of anticipated growth.

This is why comparing raw P/Es across different companies is so treacherous. A software company at a P/E of 35 and a utility at a P/E of 12 are not "expensive" and "cheap" — they reflect utterly different growth prospects. A meaningful comparison is always relative: a company versus its own history, versus its direct competitors, or versus its industry average. A bank trading at a P/E of 8 when its peers trade at 12 is genuinely worth investigating; a bank at 8 versus a tech firm at 30 tells you almost nothing.

The Earnings Yield

A useful trick is to flip the P/E upside down. The earnings yield is EPS ÷ price — the inverse of the P/E — expressing the company's earnings as a percentage return on the price you pay. A P/E of 20 is an earnings yield of 1 ÷ 20 = 5%; a P/E of 10 is a 10% earnings yield. This reframing is powerful because it puts a stock on the same footing as other investments: you can compare a 5% earnings yield directly with a government bond yield, asking whether the stock's earnings return (plus its growth potential) justifies its extra risk. When bond yields are low, a 5% earnings yield looks attractive; when bonds yield 6%, that same stock looks dear. The earnings yield turns the abstract P/E into a comparable rate of return.

The Traps

The P/E's simplicity hides several pitfalls that catch beginners:

• The value trap. A very low P/E is not automatically a bargain. Often the market has marked the stock down because it expects earnings to fall — and a P/E of 6 on a declining business can become a P/E of 20 once earnings shrink, with the price dropping further. Cheap-looking can mean rotting. Always ask why a P/E is low.
• Negative or tiny earnings. A company with no profits has no meaningful P/E (you cannot divide by a loss), and one with temporarily depressed earnings can show an absurdly high P/E that says nothing. The ratio breaks down precisely when earnings are unusual.
• Cyclical distortion. For cyclical companies (miners, carmakers, banks), the P/E is treacherous: earnings peak at the top of the cycle, making the P/E look low just when the stock is most dangerous, and collapse at the bottom, making the P/E look high just when it is cheapest. For cyclicals, a low P/E can be a sell signal.
• Accounting quality. Earnings can be massaged. A P/E built on flattered, low-quality earnings (recall the cash flow lesson) is unreliable — which is why P/E should never be used alone.

Common Misconceptions

• "High P/E means overvalued, low P/E means cheap." The P/E reflects expectations. A high P/E can be justified by growth; a low one can signal decline. Neither is a verdict on its own.
• "I can compare any two companies' P/Es." Only relative comparisons — versus peers, the industry, or the company's own history — are meaningful. Across different industries, raw P/Es mislead.
• "A loss-making company has a low P/E." It has no meaningful P/E — you cannot value earnings that don't exist. The ratio simply doesn't apply.
• "The P/E tells me everything I need." It is one lens among many. Used alone — especially without checking earnings quality and growth — it is dangerously incomplete.

Real-World Application

An investor screens for "cheap" stocks and finds two at a P/E of 9, far below the market average. Rather than buy both as bargains, they ask the essential question — why is each cheap? The first is a steady, profitable consumer-goods company whose earnings have grown slowly but reliably for a decade; the low P/E reflects modest growth expectations, and at a 11% earnings yield against low bond yields, it looks genuinely attractive. The second is a retailer whose sales are shrinking as customers desert it; its low P/E is the market correctly pricing in falling earnings — a value trap, where the "9" will balloon as profits evaporate. Same headline number, opposite realities. By treating the P/E as a statement of expectations to be interrogated, not a verdict to be obeyed, the investor buys the first and avoids the second — using the most popular ratio in investing the way it is meant to be used: as the start of a question, never the end of one.

Key Takeaways

• The P/E ratio = share price ÷ EPS — how much you pay per dollar of annual earnings (a P/E of 20 means $20 per $1 of earnings, roughly 20 years of current earnings).
• Trailing P/E uses actual past-year earnings (factual but backward-looking); forward P/E uses forecast earnings (relevant but uncertain). Use both.
• A P/E reflects expectations, not cheapness: high P/Es price in growth, low P/Es price in stagnation or decline — so compare only relative to peers, industry or history.
• The earnings yield (1 ÷ P/E) reframes the P/E as a percentage return, letting you compare a stock to bonds and other investments.
• Beware the traps: the value trap (cheap for a reason), negative/erratic earnings, cyclical distortion, and low-quality earnings — the P/E is a first lens, never the whole analysis.