Two halves of the same company just moved in opposite directions. International Business Machines (NYSE:IBM) saw its IBM Z mainframe line fall 42% in fiscal Q2 2026, while distributed infrastructure, which covers Power and Storage, jumped 37%. Total revenue rose only 1% to $17.2 billion, and management updated its full-year outlook to 4% to 5% constant-currency growth.
IBM sells software (Red Hat, data tools, automation), consulting, and the infrastructure that runs big companies’ systems. Software is its highest-margin segment, with a 32.2% segment profit margin. Whether the miss is a pothole or a warning depends less on the mainframe number than on what software does next, and on whether 18 times earnings already reflects it.
Margins Tell the Story
The moat shows up in the economics rather than the marketing. Software carried an 82.6% gross margin in the quarter, and its segment profit margin improved from 31.1% a year earlier. Red Hat, the open-source arm, grew 11%, and the data business grew 19%.
Even after the miss, the balance sheet showed $16,160 million of current deferred income (billed to clients, not yet counted as revenue), a sign the contract base hasn’t evaporated. For investors looking past high-flying multiples, reviewing the best 52-week low technology stocks offers a look at beaten-down alternatives.
A Pause, Not a Collapse
Zoom out and the mainframe slump looks more like a pause than a collapse. Infrastructure revenue rose 1.9% across the first half even after falling 7% in the second quarter. Power and Storage ended the quarter with an order backlog (committed but undelivered orders) of nearly $500 million.
IBM has also kept spending on what comes next. Since the report, it unveiled a dual-architecture mainframe processor, designed to let companies run operating systems and applications on both IBM and Arm platforms. It also closed its purchase of HRL Laboratories, a research institution spanning quantum computing and advanced electronics, and plans to invest more than $10 billion in quantum over five years. None of that pays off soon, but it suggests the moat is holding rather than eroding.
Is Software the Real Problem?
The tougher question for bulls is software, the business that has to carry growth. It grew 5% in fiscal Q2, slower than the 7.9% pace across the first half, and Transaction Processing, one of its lines, shrank 8%. Consulting was flat at $5.3 billion. Free cash flow of $2.5 billion slipped $0.3 billion from a year earlier, and debt totals $62.0 billion after $10.5 billion of acquisitions this year.
IBM’s finance chief pointed to revenue headwinds late in the quarter, a timing explanation the next reports need to confirm. Until software growth stops slowing, that risk stands.
Priced As Nothing Happened
At 18.00 times expected earnings, as of October 6, investors pay about $18 for every $1 of profit the market anticipates. That’s almost exactly IBM’s five-year average of 18.18, and well below the sector’s 23.77. It reads two ways. Against its own history, the stock isn’t discounted at all, as if the stumble never happened. Against the sector, the gap looks wide for a business whose software margins are climbing. Expected EPS growth for 2027 is 6.80%, respectable for a company that just posted 1% revenue growth, though not a pace that begs for a premium. Hedge fund interest ticked up, with 74 funds holding the stock in the most recent quarter, up from 59 in the prior one.
One caveat on earnings quality. Forward multiples are usually built on adjusted earnings, and IBM’s operating (non-GAAP) EPS of $2.93 compares with $2.27 on a GAAP basis. Acquisition-related adjustments account for $0.58 of that gap, mostly amortization of acquired intangibles. Quantum computing is a long-term bet for IBM, and there areother stocks exposed to the same trend. Short interest stands at 2.65% of the float, which points to relatively limited bearish positioning even after the stumble.
Patience, with Conditions
The evidence leans toward a pause rather than a break in the mainframe business, and eighteen times earnings looks like a fair price for the stumble rather than a bargain for the rebound. That setup suits patient investors comfortable with lumpy hardware quarters. Another quarter of software growth at or below 5% with consulting still flat would weaken the pause story. Delayed deals closing and the Power and Storage backlog turning into revenue would strengthen it.
READ NEXT: Top 10 AI Stocks That Will Skyrocket and 2 Stocks Trump Bought in July Are Now Down 17% to 20%. Time to Buy?