IBM Has Fallen 33% From Its High and Yields 3%. Here's What That Dividend Actually Costs the Company.

International Business Machines (NYSE: IBM) trades at about $224 as of this writing, roughly 33% below its 52-week high of $332.46. The slide has pushed the technology veteran's dividend yield to about 3% -- a level that tends to attract income investors' attention.

A yield is only half of a dividend's story, though. The other half is what the payments actually cost the company, and whether the underlying business produces enough cash to keep writing the checks. So, what does IBM's dividend cost?

About $6.4 billion a year. The company paid $3.2 billion of dividends in the first half of 2026 alone. At $1.69 per quarter, or $6.76 per year, the payout consumes about 60% of IBM's trailing earnings per share of a bit over $11.

Image source: Getty Images.

The cash behind the checks

A 60% payout ratio on earnings sounds tight, and after a quarter in which the company cut its own growth outlook, it deserves a hard look. However, the coverage is sturdier than it first appears.

IBM generated $14.7 billion of free cash flow in 2025, up $2 billion from the prior year. And with its second-quarter report on July 22, management said it still expects free cash flow to increase by about $1 billion in 2026, which would put the full-year figure near $15.7 billion. Against that, a $6.4 billion dividend consumes only about 40% of the cash the business is expected to produce this year.

Of course, the year is only half over. First-half free cash flow of $4.8 billion was flat year over year, and the second quarter's $2.5 billion was down about 11%, so the roughly $1 billion of growth management projects has to arrive in the second half. It helps that IBM is not a heavy spender on plants and equipment -- net capital expenditures were just $359 million in the second quarter -- so its operating cash largely flows through to free cash flow instead of into construction.

The company has also raised its payout for 31 consecutive years, a streak management has protected through far worse stretches than this one.

However, April's increase was a single penny, from $1.68 per quarter to $1.69. IBM guards the streak without lavishing it.

Why the stock fell anyway

If the dividend is affordable, the 33% drawdown needs another explanation, and the second quarter offered one. Revenue rose just 1% year over year to $17.2 billion. Alongside the report, management trimmed its full-year outlook to 4% to 5% revenue growth in constant currency, down from the more than 5% it had forecast entering the year.

The segment detail explains the softness. Software, the segment that carries the growth case, delivered, with revenue up 5% year over year and Red Hat revenue up 11%. Even there, though, growth slowed: The same segment grew 11% in the first quarter. But consulting revenue was flat year over year, and infrastructure revenue fell 7%, dragged down by a 42% decline in IBM Z, the company's mainframe computer line. Mainframe revenue tends to run in cycles tied to product launches, so that decline is likely more cyclical than permanent -- but it is a drag all the same.

So the market hasn't repriced the dividend. It has repriced the growth.

At about 17 times forward earnings estimates, the stock is now valued like a company that grows slowly, because in the second quarter it was one.

Could the growth case reassemble itself? It could. A new mainframe cycle eventually resets the infrastructure comparison, and CEO Arvind Krishna said with the second-quarter release that IBM is in the "early innings" of a structural shift for business. If software growth reaccelerates, today's price may end up looking conservative.

As dividend stocks go, the case here is respectable as it stands. The dividend is covered about 2.5 times by this year's expected free cash flow, and the payout has grown for 31 straight years. After all, at the stock's 52-week high, the same $6.76 payout yielded just 2%. Today's fatter yield came from the falling stock price, not from faster dividend growth.

I'd still hold off, personally. A dividend yielding 3% and growing by a penny a year isn't enough on its own, and the growth that has to carry the rest of the case decelerated last quarter. I'd want to see software reaccelerate before buying this drawdown. The dividend is safe. That alone doesn't get me there.

Daniel Sparks and his clients do not have positions in any of the stocks mentioned. The Motley Fool has positions in and recommends International Business Machines. The Motley Fool has a disclosure policy.