The P/E ratio, explained — and five ways it misleads

The price-to-earnings ratio is the most quoted number in investing and one of the most misunderstood. It is easy to calculate, easy to compare, and easy to misread. This guide explains what a P/E actually measures, shows the full range of P/Es across the companies we cover, and walks through five ways the ratio leads people astray — each with a real company whose numbers you can check on our comparison pages.

What a P/E measures

A P/E divides the share price by earnings per share over the past twelve months. Coca-Cola closed at $91.78 on 14 October 2026; over the previous twelve months it earned $3.18 per share; so its P/E is 28.86. Put plainly, investors were paying about $29 for each dollar of Coca-Cola's annual profit.

Turned upside down, the same number becomes an earnings yield: a P/E of 20 means the company earns 0.05 of its price — 5% — each year. That framing helps when comparing a share with a bond, and it makes one thing obvious: a P/E is a price, not a quality score. It tells you what the market charges for current profits, not whether those profits are good, growing or durable.

All of our P/Es are trailing: they use reported earnings from filings, not analysts' forecasts. We explain why we never print a forward P/E in the numbers we refuse to print.

The range across the companies we cover

At the 14 October 2026 close, P/Es ran from about 11 to about 403:

A spread that wide is not a ranking from expensive to cheap. Almost every company on the list is "mispriced" by P/E alone for a different reason. Here are the five that matter most.

Trap one: ignoring growth

Nvidia and Coca-Cola traded on almost the same P/E — 27.98 and 28.86. Yet Nvidia's earnings per share grew 206.61% a year over its last three fiscal years, while Coca-Cola's grew 11.55%. Paying the same multiple for both cannot mean the same thing.

The PEG ratio tries to fix this by dividing the P/E by the growth rate. Nvidia's PEG is 0.14; Coca-Cola's is 2.5. A PEG has its own problems — it assumes past growth continues, which for Nvidia's pace is a heroic assumption — but it is a useful reminder that a P/E without a growth rate beside it is half a sentence. Our Nvidia vs AMD comparison shows growth and valuation side by side.

Trap two: the cyclical peak

Micron's P/E of 12.3 is one of the lowest on the list, and it comes at a moment when its operating margin is 74.59% and its revenue has grown 104.65% a year for three years. That combination is the classic setting for the cyclical trap. Memory chips are a cyclical business: when supply is short, prices and margins soar, and earnings look enormous relative to the share price. When the cycle turns, earnings can fall faster than the price, and the "cheap" P/E turns out to have been the most expensive point to buy. Micron itself lost money in its 2023 fiscal year. With cyclical companies, a low P/E often marks peak earnings rather than a bargain. See how Micron compares with Nvidia in Nvidia vs Micron.

Trap three: depressed or inflated earnings

A P/E is only as good as the "E", and earnings can be temporarily squashed or swollen:

• Depressed. Pfizer traded at 38.43 times earnings, a multiple usually associated with fast growers. But its earnings per share fell 37.12% a year over three years, and the same shares cost only 2.61 times annual revenue. The high P/E says that earnings shrank, not that Pfizer became expensive. General Motors shows how quickly the "E" can move: at the same 14 October price, its P/E fell to 18.17 once a weak quarter dropped out of the trailing twelve months.
• Inflated. Alphabet's P/E of 17.44 looks modest for a company growing revenue 12.51% a year. But its net income of $244.21bn over the past twelve months exceeded its operating income of $147.63bn: a large slice of profit came from outside the operating business. Value the operations alone and the multiple is much higher, which is why its EV/EBITDA of 25.01 tells a different story.

The fix is to look at a second, steadier measure alongside the P/E. Our guide to EV/EBITDA explains the most useful one.

Trap four: comparing across industries

Banks almost always trade on lower P/Es than technology companies, because their earnings depend on credit cycles and interest rates, and because they need to retain capital. JPMorgan at 14.37 and Bank of America at 11.87 are not "cheaper" than Microsoft at 29.73 in any useful sense; they are different kinds of business. Compare a P/E with peers that earn money the same way. For banks, price-to-book and return on equity do most of the work — see how to compare bank stocks.

Trap five: no earnings at all

When a company loses money, it has no meaningful P/E. Ford, Intel, Rivian and Lucid all reported losses over the past twelve months, and our pages give no P/E for them rather than a negative one. That is not a gap in the data — it is a sign that a different set of questions applies: how much cash is left, how fast losses are shrinking, and what revenue is worth. For loss-makers, price-to-sales and cash burn matter more than any earnings multiple.

The 403-times question

Tesla is the extreme case. At 403.47 times earnings per share, a buyer was paying about 403 years of current profit for each share. On whole-company figures the multiple is higher still — $1,451.3bn of market value against $3.22bn of net income over the past twelve months — because market value uses the latest share count, which is larger than the average count behind earnings per share. Tesla's operating margin was 3.35%. No reasonable investor expects those earnings to stay flat; the price reflects a belief that Tesla's future profits will be far larger than its present ones, or will come from businesses that barely show up in today's accounts. The P/E cannot tell you whether that belief is right. It can only tell you how much of the price depends on it — and at 403 times, the answer is almost all of it. Compare it with a traditional carmaker in Tesla vs General Motors.

How to use a P/E well

A P/E is a good first question and a bad last answer. Use it this way:

1. Compare it with close peers, not with the market as a whole.
2. Put the growth rate next to it. A higher P/E can be cheaper if growth is much faster and durable.
3. Check the quality of the earnings. If net income is far above or below operating income, find out why before trusting the multiple.
4. Ask where the cycle is. For cyclical businesses, a low P/E near peak margins is a warning.
5. Confirm with cash. A company's free cash flow yield is harder to flatter than its earnings.

Every comparison page puts the P/E beside these other measures for exactly this reason — and the six-factor score deliberately does not let trailing P/E decide the valuation verdict on its own.