Divide the current share price by earnings per share. With a $100 share price and $5 EPS, P/E = $100 / $5 = 20x, meaning investors pay $20 today for every $1 of annual earnings. Use trailing twelve-month EPS for trailing P/E and analyst-estimated forward EPS for forward P/E.

P/E is a raw valuation multiple. PEG divides P/E by the expected EPS growth rate to factor in growth: a 20x P/E on a company growing 20% has a PEG of 1.0, while the same 20x on a 5% grower has a PEG of 4.0. PEG below 1.0 is often viewed as attractive, though it depends on whether growth estimates are realistic.

Trailing P/E uses the last four reported quarters of EPS (actual, known data). Forward P/E uses analyst-estimated EPS for the next twelve months. Forward P/E is more predictive but depends on the accuracy of estimates, which often skew optimistic. Most professional screens display both side by side.

P/E breaks down for companies with negative or near-zero earnings (the ratio is undefined or astronomical), for cyclicals at peak or trough earnings (the multiple looks low at peak and high at trough—the opposite of value), and for companies with large one-time gains or write-offs. In these cases, prefer EV/EBITDA, P/S, or normalized earnings.

Shiller's Cyclically Adjusted P/E (CAPE) divides price by the ten-year average of inflation-adjusted earnings. By smoothing across a full business cycle, CAPE filters out the peak/trough distortions that affect trailing P/E. Historically, high CAPE readings have correlated with weaker forward ten-year market returns.

Higher P/Es reward expected growth, durable margins, low capital intensity, and recurring revenue. Technology and healthcare typically trade at 25–40x. Financials and energy sit lower at 8–15x because of cyclicality and capital intensity. Utilities cluster at 15–20x with bond-like cash flows. Always compare a stock to its own sector before drawing conclusions.

The S&P 500's long-run median trailing P/E sits around 15–16x, with extended periods between 12x (cheap) and 25x+ (expensive). Multi-year readings well above the median have historically preceded lower future returns, while readings below have preceded higher returns—useful context, not a market-timing signal.