A record that stood since Black Monday, broken in a single morning

International Business Machines has been a publicly traded company since 1916, and reliable daily trading records for the stock stretch back to 1968. In all that time, nothing had ever hit IBM's share price as hard as Tuesday, July 14, 2026. Shares closed down more than 25%, at $217.07, marking the steepest single-day decline in the company's recorded trading history — surpassing even the 23.7% single-day drop IBM suffered on October 19, 1987, the day of the broader market crash known as Black Monday.

What makes this collapse unusual isn't just its size. It was entirely self-inflicted, and it arrived a full week ahead of schedule. IBM was set to report its official second-quarter results on July 22. Instead, CEO Arvind Krishna sent an unscheduled letter to investors before markets opened Tuesday, disclosing preliminary revenue and earnings that fell meaningfully short of Wall Street's expectations — a voluntary early disclosure that erased roughly $68.8 billion of the company's market value within a single trading session.

The numbers behind the warning

IBM's preliminary figures showed revenue of $17.2 billion for the quarter, against a FactSet consensus estimate of $17.86 billion — a shortfall of roughly $660 million, or about 3.7% below expectations. Adjusted earnings per share came in at $2.93, versus consensus estimates of $3.01 to $3.02 depending on the data provider. Notably, revenue still grew 1% year over year, and operating earnings per share still increased 5%, with operating pre-tax margin actually expanding 30 basis points to 19.2%. That's an important detail buried beneath the panic: IBM didn't report an outright earnings collapse. It reported a miss against inflated expectations, on numbers that were still modestly growing in absolute terms.

Krishna didn't shy away from taking direct responsibility in his letter to investors, writing simply, "this quarter we faltered." He attributed the bulk of the shortfall to two intertwined factors: a batch of large enterprise deals that failed to close within the expected timeframe, and a broader, more structurally concerning shift in how corporate clients allocated their technology budgets during the final weeks of the quarter.

Where the money actually went instead

The more consequential explanation Krishna offered centered on a scramble happening across the broader technology sector. "In the last few weeks of June, we saw clients shift their quarterly capex spend toward servers, storage, and memory purchases to secure supply-constrained infrastructure ahead of expected price increases," Krishna wrote. "This dynamic impacted client buying patterns." In plainer terms: IBM's enterprise customers, facing a tightening global memory chip shortage and anticipating further price increases, redirected budgeted spending away from IBM's software and mainframe products and toward securing physical hardware — servers, storage, and memory chips — before prices climbed further or supply tightened more severely.

That shortage isn't hypothetical or IBM-specific. Memory chip giants Samsung, SK Hynix, and Micron have all redirected substantial manufacturing capacity toward the specialized high-bandwidth memory chips used in AI data centers, leaving the broader supply of conventional memory used in everyday servers, PCs, and phones increasingly squeezed. Micron's own CEO has said publicly that he expects "tight conditions to persist beyond calendar 2027" given AI-driven demand across essentially every market segment, layered on top of structural supply constraints that predate the current AI buildout. IBM's warning is, in effect, the first major earnings casualty of that broader squeeze showing up in a software and services business rather than a hardware manufacturer.

The z17 mainframe cycle, and where it actually broke

The specific business segment hit hardest was IBM's Z mainframe line, particularly its newest generation, the z17, alongside the associated Transaction Processing software that runs on it. Krishna had told investors back in April that the company expected a low-single-digit revenue decline in its infrastructure segment for the year, as the z17 program wrapped its initial launch cycle — a normal, anticipated pattern for mainframe product cycles, where an initial surge of sales inevitably tapers as the immediate launch wave of customer upgrades completes. "What played out was worse than our expectations," Krishna acknowledged, "driven by a shortfall in our Z performance and the associated software stack, primarily in Transaction Processing."

What's notable, and what several analysts flagged as a reason for skepticism toward the more dire read of this selloff, is that the underlying z17 program itself doesn't appear structurally broken. According to figures IBM disclosed, z17 placements are running at nearly 130% of the equivalent point in the prior z16 cycle — IBM's strongest mainframe launch on record by that measure — and roughly 85% of the installed base is holding steady or growing its capacity. That's not the profile of a product line losing customers. It's the profile of a strong underlying business that experienced a specific, quarter-end timing disruption in when large contracts actually closed.

A genuine debate: temporary scramble, or structural shift

The central question Wall Street is now wrestling with is whether this quarter represents a one-off timing anomaly or the first visible sign of a durable shift in enterprise technology spending. Read one way, this was a temporary scramble: corporate budget teams rushed to lock in hardware purchases ahead of anticipated price increases, some large software and mainframe deals slipped from June into July, and normal ordering patterns should reassert themselves once component supply stabilizes. Under that reading, the revenue IBM "lost" this quarter simply moved into the next one, and the 25% stock decline reflects panic disproportionate to the actual underlying business disruption.

Read the other way, this quarter is an early warning that AI infrastructure spending is durably capturing a larger share of overall enterprise technology budgets — permanently, not temporarily — at the direct expense of the software, mainframe, and services revenue that still generates the majority of IBM's profit. Under that reading, IBM's core business model faces a genuine structural headwind that a supply chain easing won't necessarily reverse, because the shift reflects where corporate priorities are actually moving, not merely where component scarcity happens to be pushing near-term purchasing.

What actually resolves this, and when

IBM's own disclosure offers some evidence for the more optimistic reading. Red Hat, the open-source software business IBM acquired in 2019, saw its growth rate accelerate sequentially to 11% during the quarter. IBM's Distributed Infrastructure segment — the servers-and-storage business arguably best positioned to benefit from exactly the AI hardware shift that hurt the Z mainframe line — posted its strongest performance on record, up 37%, and exited the quarter carrying roughly $500 million in backlog. Year-to-date free cash flow stood at $4.8 billion. None of that is the profile of a company in broad decline; it's the profile of a company where growth is simply redistributing across business lines faster than some of those lines can absorb the shift smoothly.

The market's forward-looking valuation of IBM shifted accordingly — its forward price-to-earnings multiple compressed from roughly 23 times to about 17 times in a single session, reflecting genuine uncertainty about whether current earnings estimates need to come down, not just a knee-jerk reaction to a single data point. That uncertainty won't be resolved by commentary or analyst speculation. It gets resolved on July 22, when IBM delivers its full, official second-quarter results and, critically, updates its full-year guidance. Investors will be watching specifically for whether the delayed contracts Krishna described have actually closed since quarter-end, at what values, and whether IBM's own outlook treats this quarter as an isolated timing disruption or acknowledges a more durable reallocation of enterprise technology spending toward AI infrastructure — the distinction that will determine whether Tuesday's historic drop marks a buying opportunity or the first data point in a longer reassessment of IBM's core business.

*This article was researched using publicly available reporting from CNBC, Yahoo Finance, Business Insider, EBC Financial Group, FX Leaders, and BestStocks coverage of IBM's preliminary second-quarter results and market reaction. It is intended for informational purposes and does not constitute financial advice.*