How to calculate earnings per share
EPS = (net income − preferred dividends) ÷ weighted average common shares. Preferred dividends come off first because that money goes to preferred shareholders before common shareholders get anything. Most companies have no preferred stock, so for them EPS is simply net income ÷ shares.
The drawing at the top shows the hypothetical example: ($540 million − $40 million) ÷ 100 million shares = $5.00. Companies report EPS for each quarter, and data sites add up the last four quarters to show trailing twelve-month (TTM) EPS.
- Weighted average shares
- The average number of common shares outstanding during the period, weighted by how long each share existed. 100 million shares for six months and 90 million for the other six give a weighted average of 95 million.
- Preferred dividends
- Dividends owed to holders of preferred stock, who rank ahead of common shareholders. They are subtracted from net income before EPS is calculated.
Basic vs diluted EPS: what is the difference?
Basic EPS divides by the shares that exist today. Diluted EPS also counts shares that could be created if employee stock options were exercised, restricted stock units (RSUs) vested or convertible securities were converted. A larger denominator gives a smaller number, so diluted EPS is the conservative figure, and the one most data sites quote.
In the drawing below, $500 million of earnings over 100 million shares gives basic EPS of $5.00. Add 5 million potential shares from options and RSUs, and the same profit over 105 million shares gives diluted EPS of $4.76. A wide or widening gap tells you how much dilution is waiting. When a company loses money, potential shares are left out, because they would make the loss per share look smaller, so basic and diluted EPS are equal.
GAAP vs adjusted EPS
US companies report GAAP EPS under the official accounting rules, and many also publish an adjusted (non-GAAP) EPS that leaves out items such as stock-based compensation, restructuring charges or the amortization of acquired intangibles. Analysts’ estimates and the “beat or miss” headline usually use the adjusted figure. The P/E ratio guide shows how the choice changes the multiple, and how to read earnings reports shows where to find the reconciliation table.
One habit covers most cases: compare GAAP with GAAP and adjusted with adjusted, and be wary when the gap between the two keeps widening year after year.
How do share buybacks increase EPS?
A buyback reduces the number of shares, so the same profit is split among fewer of them. In the hypothetical drawing below, net income is $500 million in both years. In year 1 there are 100 million shares and EPS is $5.00. The company buys back 10 million shares, and in year 2 the same $500 million over 90 million shares gives EPS of $5.56: 11% EPS growth with 0% profit growth.
That is not a trick in itself: each remaining share owns a larger slice of the business, and a buyback at a sensible price can be a good use of cash. But it is not the same as a business that sells more and earns more. Check whether net income grew too, whether the buyback was paid for with free cash flow or with new debt, and whether it mostly offsets shares handed out as stock compensation, in which case the share count barely moves.
EPS growth: how to tell if it is real
EPS growth = (this year’s EPS − last year’s) ÷ last year’s. Investors watch it because, over long periods, share prices tend to follow earnings, as the growth investing guide explains. But one growth rate hides where the growth came from.
The waterfall below splits a hypothetical rise from $4.00 to $4.60, +15%, into its sources: $0.20 from higher operating profit, $0.15 from fewer shares, $0.10 from a lower tax rate and $0.15 from a one-time gain on selling a building. Only $0.20, 5 of the 15 percentage points, came from the business itself. Check before you trust an EPS number:
- Revenue and operating income should grow too; EPS rising while sales shrink usually means cost cuts, buybacks or one-offs.
- Look for one-time gains, tax changes and share count changes in the report.
- Compare EPS with free cash flow: profits that never turn into cash deserve suspicion.
- Follow the trend over several years, not one quarter, using diluted GAAP EPS for long-term comparisons.
Is a higher EPS always better?
Not across companies. EPS depends on how many shares a company has issued, so $10 of EPS is not better than $2: a company with five times as many shares can earn the same profit and report one fifth of the EPS. A stock split cuts EPS in proportion without changing the business. Use EPS to follow one company over time and as the input of the P/E; to compare two companies, use ratios such as the P/E, margins or other valuation ratios.
Prices react to EPS against expectations, not to EPS alone: a company can grow EPS 20% and still fall if investors expected 30%. That is why earnings dates carry gap risk for traders, and why, for investors, the price paid for each dollar of EPS matters as much as the EPS itself.