Price-to-Earnings Ratio (P/E)

The price-to-earnings ratio (P/E) is a company's share price divided by its earnings per share. It shows how many dollars, pounds or euros investors are paying for each unit of the company's annual profit, and it is the most widely quoted valuation multiple for listed shares. Analysts, fund managers and individual investors use it to compare a company with its own history, with rivals in the same sector and with the market as a whole.

The same figure can be calculated for the whole company as market capitalization divided by net income. Several versions are in use: the trailing P/E based on reported profit, the forward P/E based on forecasts, and the cyclically adjusted P/E (CAPE) that averages ten years of inflation-adjusted earnings. Its inverse, earnings divided by price, is called the earnings yield and is often set against bond yields.

Imagine buying a small corner shop that makes a profit of 10,000 a year. If the seller asks 100,000, you are paying ten times the yearly profit; if the seller asks 250,000, you are paying twenty-five times. The P/E ratio is that multiple, applied to a company's shares.

A high multiple usually means buyers expect the profit to grow, or see it as unusually safe. A low multiple can mean the business is out of favour, faces risks, or is simply cheap. On its own the number does not say which; it is a starting point for asking why buyers are paying what they pay.

The standard calculation is:

Here price per share is the current market price and earnings per share is usually diluted EPS for the chosen twelve-month period. Worked example: a share trades at $60 and the company earned diluted EPS of $3 over the last four quarters, so the P/E is $60 \div 3 = 20$. The earnings yield is $3 \div 60 = 5\%$, which an investor might compare with the yield on a ten-year government bond or another risk-free rate.

The choice of earnings figure changes the answer considerably:

• Trailing P/E (also written P/E TTM, for "trailing twelve months") uses the EPS reported over the last four quarters or the last financial year. It rests on audited or reviewed numbers, but it looks backwards.
• Forward P/E uses forecast EPS, normally the analysts' consensus for the next twelve months or the next financial year. It reflects expected growth, but it depends on estimates that are often revised and are frequently based on adjusted rather than reported earnings.
• The cyclically adjusted P/E, or CAPE, divides price by the average of the last ten years of earnings after adjusting them for inflation. It was developed by economists John Campbell and Robert Shiller in the late 1980s and is often called the Shiller P/E. Shiller publishes the underlying data for the US market back to 1871.

For a company whose profits are expected to rise, the forward P/E is lower than the trailing P/E. CAPE is used mainly for whole markets rather than single companies; its highest recorded reading for US stocks came at the dot-com peak of 1999–2000, and research has linked high CAPE readings with lower returns over the following decade, though it has proved a poor tool for timing exactly when a bear market will begin.

In theory a P/E reflects the present value of future profits, so anything that changes expected growth or the rate used to discount it changes the multiple:

• Expected growth. A growth stock trades on a high P/E because buyers are paying for profits years ahead; a value stock trades on a lower one.
• Interest rates. When rates fall, the discount rate applied to future profits falls and multiples tend to rise; rising rates tend to compress them.
• Risk and quality. Stable, predictable earnings usually earn a higher multiple than volatile ones.
• Sector. Software and consumer-brand companies have historically traded on higher multiples than banks, utilities or miners, so comparisons are most useful within a sector.

Index providers and data firms publish P/E ratios for benchmarks such as the S&P 500, and market commentary on a bull market often centres on whether the index multiple has risen above its long-run average. A related refinement, the PEG ratio, divides the P/E by the expected annual EPS growth rate in percent, so a P/E of 20 with 10% expected growth gives a PEG of 2.

• A company with losses has no meaningful P/E, and one with tiny profits can show an absurdly high one.
• Cyclical companies often look cheapest at the top of the cycle, when profits are at their peak, and most expensive at the bottom.
• Reported EPS can be distorted by one-off items, accounting choices and buybacks, so analysts often also use multiples based on cash flow, sales or enterprise value.
• The P/E ignores debt: two companies with the same multiple can carry very different balance sheets, which is why it is often read alongside the price-to-book ratio and the dividend yield.

Used with care, the P/E is a quick first measure of valuation and a staple of value investing screens, but it is not a measure of intrinsic value in itself.

• Investor.gov (SEC): Price-earnings (P/E) ratio
• Robert J. Shiller, Yale University: Online data (CAPE)