Paying for IBM's Software, Getting Quantum for Free
The market prices IBM off a single line item. It is missing the machine underneath.
Update — August 1, 2026. I wrote the piece below in early July, before IBM fell 25 percent on July 14, its worst single-day drop on record. The trigger: IBM disclosed that some enterprise customers had delayed large capital-spending projects, and CEO Arvind Krishna issued an unusually blunt letter admitting the company had been slow to adapt. My read is that this strengthens the case rather than breaking it. The market treated a timing problem as a demand problem. A semiconductor price spike pushed clients to redirect budget toward physical servers, which hit the roughly one-fifth of IBM’s software sold upfront, not the roughly four-fifths that is recurring subscription. Demand was deferred, not lost, and Krishna has since said about a third of the delayed deals have already returned. Then on July 30 he told Jim Cramer that quantum would have a measurable impact on IBM’s revenue and earnings by 2028 or 2029, and that the company is now convinced it is worth a trillion dollars of value by the end of the 2030s. At today’s lower price the market is giving IBM close to no credit for any of that. Before the drop I thought IBM was mispriced and worth holding. After it, for the same reasons, I think it is more of a buy.
IBM was once the most feared company in technology, and it has spent most of two decades being treated as a relic. The market’s entire relationship with the stock now runs through a single line item: consulting revenue. Every quarter, Wall Street checks whether consulting grew, and prices the stock off that one number.
Start with what the market is getting wrong, because it is quantifiable. In the first quarter, consulting grew about 4 percent as reported and roughly 1 percent at constant currency, and the market extrapolated that sluggishness across the whole company. But consulting is only about a third of revenue, and it is the lowest-margin third. The engine is software, which did around $7.1 billion in the quarter, grew 11 percent, is guided above 10 percent for the full year, and carries gross margins near 80 percent. As software becomes a larger share of the mix, the entire company’s margin structure re-rates upward. IBM is guiding to roughly $14 billion of free cash flow this year, its highest in a decade, which comfortably funds the dividend.
So here is what you are actually paying for. At a market multiple you get the software-and-infrastructure company, and you get the quantum franchise close to free.
On May 21, IBM and the U.S. Department of Commerce announced a letter of intent to build Anderon, described as America’s first purpose-built quantum foundry, headquartered in Albany, New York. It is backed by a proposed $1 billion CHIPS award from Commerce and $1 billion of IBM’s own cash. That award was the single largest piece of a roughly $2 billion federal quantum package spread across nine companies, the largest quantum research commitment in U.S. history, with the government taking an equity stake in each recipient. When Washington decided to plant its flag in quantum, it put half of everything it committed behind IBM.
I want to spend a moment on why quantum matters, because I do not think the market has begun to price what it is. This is a different kind of machine, built on different rules, and it can solve problems a classical computer cannot solve at all. That is the whole point. We are not talking about an upgrade. We are talking about a new category of machine that does work the old one was never capable of.
Here is how I explain the difference. Imagine you are standing inside a corn maze at ground level. At every fork you can see only the two choices directly in front of you, so you pick one, then the next, until you hit a dead end and have to walk all the way back and try again. That is how a classical computer solves a hard problem. It works one path at a time. Now imagine you are lifted above the maze and can see the whole thing at once, the entrance, the exit, and every path between them. You can work backward from the goal and forward from the start and solve in a minute what took ten from the ground. That is the leap quantum represents. Problems that would take a classical machine thousands of years can collapse to hours.
That capability is why quantum is a national-security race. Nearly all of modern encryption rests on the assumption that certain math problems are effectively unsolvable in any reasonable amount of time. Quantum breaks that assumption. The systems that protect banks, government data, and private communication were built for a world without this machine. That turns quantum leadership into an arms race, the United States against China above all, because whoever arrives first holds both the sword and the shield.
The market is beginning to catch on. IBM’s own CEO now says quantum advantage arrives this year, not in a decade. When a thesis this large is only starting to be recognized, the market’s habit of pricing the whole company off one quarterly consulting number looks less like risk and more like opportunity.
Disclosure: I own IBM, and so may Highwater Partners, the partnership I manage. This is my personal opinion, not investment advice or a recommendation to buy or sell any security. See the About page.