How Share Buyback Programs Affect Your Analysis of Per-Share Metrics

Share buybacks are now the main way big US companies return cash to shareholders. Among the largest names, repurchases have exceeded dividends for years, and the annual totals run into the hundreds of billions of dollars. At that scale, buybacks quietly reshape nearly every per-share number you look at. If you read EPS growth at face value without checking what the share count did, you can badly misjudge a business.

The mechanics are simple. A company that retires 5 percent of its shares reports roughly 5 percent higher earnings per share even if total profit never moved. Nothing about the business improved. The share count fell, so the same earnings get divided across fewer shares. Do that for a few years in a row and the stock screens like a steady grower while the underlying company treads water. The good news is that unwinding the effect takes maybe twenty minutes with the filings, and this post walks through exactly how to do it.

How the arithmetic works

Earnings per share is net income divided by weighted average shares outstanding. Buy back shares and the denominator falls, so EPS rises even when net income is flat.

Say a company earns $1 billion with 500 million shares outstanding. EPS is $2.00. It buys back 50 million shares, a tenth of the float. Earnings are still $1 billion, but now they're spread over 450 million shares, so EPS comes out to $2.22. The company just reported 11 percent EPS growth with zero growth in actual profit.

The effect compounds. Retire 4 or 5 percent of the shares every year for five years and cumulative EPS growth starts to look genuinely impressive, even if net income never moves. Screeners, quant models, and plenty of human analysts will read that as a growth story, but on a whole-company basis there's no growth to find.

Check total growth against per-share growth

The fix is a simple habit. Whenever you look at a per-share metric, put the total right next to it. Compare revenue growth to revenue per share growth, and net income growth to EPS growth. The gap between them is the buyback's contribution, and it tells you how much of the improvement is business expansion versus share count arithmetic.

Consider two hypothetical companies that each earn $1 billion today. Company A grows earnings 8 percent to $1.08 billion and leaves its share count alone. Company B grows earnings 3 percent to $1.03 billion and retires 5 percent of its shares. Both report EPS growth in the same neighborhood, but Company A built a meaningfully bigger business while Company B mostly rearranged its capital structure. Five years out those are very different companies, and per-share metrics alone won't tell you which one you own.

The data is easy to find. The income statement discloses basic and diluted weighted average shares for every period, and the cover page of any 10-K or 10-Q states shares outstanding as of a recent date. Pull three or four years of filings from EDGAR and you can chart the share count trend in a spreadsheet in a few minutes.

The price paid decides whether value was created

Whether a buyback actually helped you depends almost entirely on the price the company paid. Repurchasing stock below intrinsic value raises the value of every remaining share, because the company effectively bought a dollar for less than a dollar. Repurchasing above intrinsic value transfers wealth from the shareholders who stay to the ones who sell.

You don't have to guess at the prices. Every 10-Q includes a table, usually in Part II under Issuer Purchases of Equity Securities, showing shares repurchased by month, the average price paid, and how much remains under the authorization. Compare that average price to the current quote and to your own estimate of value. If the company paid $50 a share and the stock now trades at $80, the program worked out well for holders. If it paid $100 and the stock sits at $60, management spent shareholder cash retiring expensive stock.

There's also a well-documented behavioral pattern worth knowing. Companies tend to repurchase the most stock when times are good and prices are high, because that's when cash flow is abundant, and they cut buybacks when prices collapse, because that's when cash gets scarce. Buying high and stopping low is exactly backwards from an owner's point of view. A management team that kept repurchasing through a downturn deserves real credit, because that kind of discipline is uncommon.

What heavy buybacks do to your other ratios

EPS gets the attention, but buybacks bend other numbers too. Repurchased stock reduces shareholders' equity on the balance sheet, so return on equity rises even when the returns earned on the actual assets haven't improved. Push this far enough and book equity can go below zero. Starbucks and McDonald's have both reported negative shareholders' equity in recent years, largely because years of buybacks and dividends returned more cash than the businesses retained. Both stayed solidly profitable the whole time, which is exactly why their ROE and debt-to-equity ratios stopped meaning anything.

So before you compare ratios across companies, glance at the share count trend. A rising ROE at a serial repurchaser may reflect a shrinking denominator rather than a better business. Return on invested capital or return on assets gives you a cleaner read in those cases, since neither is distorted by the equity shrinkage.

Buybacks versus dividends

Both return cash, but they behave differently in practice. Dividends are sticky. Once a company establishes one, a cut gets read as distress, so managements protect the dividend even in bad years. Buybacks are discretionary and can be paused or quietly abandoned with no announcement at all. That flexibility is genuinely useful for the company, and it also means a buyback commitment is softer than it looks.

An authorization is just a ceiling on what the company may repurchase, and plenty of boards approve big programs that get executed slowly or never finished. If a company announced a $5 billion authorization two years ago and the cash flow statements show only $1 billion of actual repurchases since, the headline overstated the capital return by a factor of five. Actual cash spent on repurchases appears every quarter in the financing section of the cash flow statement, so checking execution against authorization takes one line item.

One recent wrinkle worth knowing about is the excise tax. Since 2023, US companies pay a 1 percent tax on net buybacks under the Inflation Reduction Act. It's small, but it tilts the math slightly toward dividends and you'll sometimes see it discussed in the equity footnotes.

The stock compensation offset

Here's the detail that catches even experienced investors. A big share of buyback spending never reduces the share count at all, because it just offsets new shares issued through stock compensation.

Say a company issues 10 million shares a year through option exercises and RSU vesting, and buys back 12 million. The net reduction is 2 million shares. Management can truthfully say it spent billions on repurchases, and the press release will lead with that number, but continuing shareholders only benefit from the 2 million net reduction. For technology companies with heavy stock-based compensation, the offset can eat most of the buyback budget.

The test is to ignore the gross spending and compute the net share count change over the period, ending shares minus beginning shares. If the company spent $3 billion on buybacks and the diluted count fell 1 percent, most of that money went to mopping up dilution. You can cross-check against the statement of stockholders' equity, which breaks out shares issued under compensation plans, and against the stock-based compensation line on the cash flow statement. The proxy statement fills in the rest of the picture on how much equity is being granted and to whom.

Watch for buybacks funded with debt

Some companies borrow to buy back stock. On the balance sheet this swaps equity for debt, and when rates are low it can even boost EPS, since the after-tax cost of the new debt can run below the earnings yield on the retired shares. The EPS accretion is real, but so is the added leverage. If earnings fall, the interest expense doesn't, and a balance sheet that looked fine at the top of the cycle can get uncomfortable quickly.

The check here is straightforward. Look at total debt at the start and end of the buyback period. If debt grew by roughly the amount spent on repurchases, the program was funded by borrowing rather than by free cash flow. A quicker version of the same test is to compare free cash flow to the sum of dividends and buybacks over a few years. A company that consistently pays out more than it generates is drawing the difference from the balance sheet, and you should understand where that leads before you own it.

A working checklist

Pulling this together, here's what I actually do when a company's EPS growth looks better than its revenue growth suggests it should:

1. Chart total revenue, net income, and free cash flow next to their per-share versions over at least five years. The gap between the growth rates is the buyback effect.
2. Pull the repurchase tables from the last several 10-Qs and note the average prices paid. Compare them to today's price and to your own estimate of value.
3. Compute the net share count change and set it against gross buyback spending to see how much went to offsetting stock compensation.
4. Check whether total debt rose alongside the program, and whether free cash flow covered the full capital return.
5. Compare executed buybacks to announced authorizations. Big announcements with thin follow-through tell you how seriously to take management's other announcements.
6. If the share count has fallen sharply, sanity-check ROE against return on invested capital before trusting it.

None of this makes buybacks good or bad on their own. A company that repurchases stock at sensible prices, funds it from cash the business actually generates, and shrinks the share count in net terms is doing right by its owners. A company that borrows to retire expensive shares while operating earnings go sideways is manufacturing the appearance of growth, and that appearance eventually runs out. The filings give you everything you need to tell the two apart, and the whole exercise takes less time than reading a single sell-side note.