A Buyback Only Helps When the Share Count Actually Falls

The line on the cash-flow statement reads “Repurchases of common stock,” with a large number in parentheses beside it. Most readers stop there. That line tells you what the company spent, which is only half of the sum, because the same company may be handing new shares to its employees in the same quarter, and every one of those shares goes straight back into the count you hoped was shrinking.

An authorization is a ceiling

A board announces that the company may repurchase up to some dollar amount. That is permission. It buys nothing. The company can use all of it, part of it or none, over a period the announcement often leaves open, and an authorization can lapse with money unspent.

Two places show what actually happened. Financing activities on the cash-flow statement give the dollars paid. The diluted weighted average share count on the income statement, or the shares outstanding on the cover of the 10-Q, give the result, and only the second one changes how much of the company’s earnings, assets and future dividends each remaining share owns. Read them side by side.

Stock pay refills the bucket

Companies that pay staff partly in stock issue new shares when restricted stock units vest and when employee options are exercised. That issuance runs whether a buyback is running or not. So the gross repurchase figure overstates the reduction. The overstatement equals the shares handed out.

The statement of shareholders’ equity in the annual report shows both movements in one table: shares issued under compensation plans on one row, shares repurchased on another. Few readers open it. It is the quickest place to see whether a year’s buyback was mostly spent standing still.

Think of a bucket with a tap running into it. Bailing faster helps. The water level is still the only thing that tells you whether it worked.

The sum on a hypothetical company

A hypothetical company has 200,000,000 shares outstanding and net income of $400,000,000, so EPS is $2.00. This year it spends $400M buying back 8,000,000 shares at $50 each. Over the same year it issues 6,000,000 shares through stock awards.

The press release would say 8,000,000 shares. The count says 2,000,000. The company spent a full year of net income, and three quarters of the shares it bought went to replace shares it had just given away, which leaves holders with a 1.0% EPS lift where a reader of the headline would have pencilled in something close to 4.2%.

The sum also holds net income flat, which flatters the buyback. Cash that has left the balance sheet no longer earns interest, so EPS after the buyback could come in a little under $2.0202. The buyback EPS calculator runs the same sum with your own company’s figures and an award line included.

The strongest objection: the buyback stopped dilution

The best case for the company runs like this. Without the buyback, the 6,000,000 award shares would have lifted the count to 206,000,000 and pulled EPS down to about $1.94. So the repurchase did something useful. It held the line.

That’s fair. It also changes what you are looking at. If $300,000,000 of the spending (6,000,000 shares at $50) exists to absorb shares paid to employees, that money is a cost of stock pay, and it belongs next to salaries when you judge the business. The cash-flow statement makes this easy to miss: stock-based compensation is added back in operating activities as a non-cash expense, while the cash that neutralizes it leaves further down, under financing, where few readers connect the two. Call it compensation. Then ask whether the business earns enough to pay it.

What to check on your own screen

• The diluted weighted average share count for each of the last several years, from the income statement.
• The repurchase line in financing activities for the same years.
• Stock-based compensation, added back in operating activities.
• The gap between them.

Dollars spent climbing while the count sits still? Then the buyback is mostly paying staff.

A rising count can hide in a single year’s figures, too. Awards tend to vest in batches, and a company can time repurchases to land in the quarter before a large vesting, so one quarter shows a fall and the next gives it back. Compare full years. The walkthrough on checking a company’s share count history shows where each figure sits in the filings, and the entry on fully diluted shares covers the options and units that have not become shares yet but will.

A falling count is the test, and the price paid is the next one

A shrinking count is necessary. It is not sufficient. A company that buys at $50 shares worth $35 on any reasonable valuation moves value from the holders who stay to the holders who sell, and the EPS lift on paper hides that transfer completely. Pay $15 too much on each of 8,000,000 shares and $120,000,000 of value leaves with the sellers. The count still drops a clean 4%. You are still worse off.

Debt is the other trap. A company that borrows to fund the buyback swaps equity for interest payments, and the interest comes out of the same net income the buyback was meant to spread over fewer shares, while the balance sheet goes into the next bad year with a thinner cushion. The broader question of when repurchases add value is worked through in are stock buybacks good for shareholders.

So the order of checks is simple. First, did the count fall, net of awards? Then, was the price sensible, and was it paid from cash the business actually generated?