Following IBM's warning on July 14, the stock experienced a one-day drop of around 25%, and then officially lowered its full-year revenue guidance in its Q2 earnings report on July 22. What triggered the sell-off was not a small percentage decrease in revenue, but the company's acknowledgment that in the final weeks of June, customers temporarily redirected capital expenditures towards servers, storage, and memory required for AI.
This statement is highly significant to investors. Over the past few years, the market has been willing to value companies like IBM based on stable cash flows because mainframes, transaction processing software, and enterprise services do not experience drastic fluctuations like consumer electronics. However, this time the market discovered that even critical systems in banks, airlines, and more might be temporarily sidelined by customers during a tight AI infrastructure supply situation.
IBM's CEO, Arvind Krishna, admitted in an investor letter on July 14 that the company did not anticipate the extent of the customer capital spending reshuffle at the end of June and described it as a misexecution for the quarter. During the earnings call, he provided a reverse explanation, stating that a significant portion of the capital spending decline, about one-third, had been completed in the first three weeks of the third quarter and characterized many demands as 'deferrals, not disappearances'.
What the market is now trying to assess is the distance between these two statements. Is this a quarter of supply chain panic amplification out of sync, or an early signal of AI capital expenditure starting to compress traditional enterprise IT's annuity revenue?
The Plummeting Pricing Involves Budgetary Zero-Sum
IBM's total revenue in Q2 was around $17.2 billion, a 1% year-over-year increase. Software revenue was approximately $7.8 billion, up 5%. Infrastructure revenue was about $3.8 billion, down 7%, with IBM Z mainframe revenue declining by 42% year-over-year. The company adjusted its full-year constant currency revenue growth expectations for 2026 from the previous "above 5%" to 4%-5%, with the free cash flow expectation still set to increase by about $1 billion year-over-year.
Merely looking at the guidance cut alone does not seem sufficient to explain a roughly 25% single-day drop. What truly triggered the reassessment was the structural shift. Large systems are inherently a cyclical business, and the market can tolerate volatility as long as it does not harm the higher-quality software cash flow.
The issue this time is that customers are not simply buying fewer machines; instead, they are reallocating their budgets to grab servers, storage, and memory needed for AI. The so-called capital expenditure reprioritization refers to the money that enterprises originally planned to use for system upgrades or software purchases. Due to concerns about AI hardware shortages and price increases, they have temporarily rushed to secure infrastructure capacity.
This has transformed IBM's decline from a financial reporting error to a valuation anchor disturbance. If AI infrastructure becomes a top priority in a company's budget, traditional IT firms, once seen as stable revenue streams, may need to be revalued.
Mainframe Decline Will Impact Software Revenue
Ordinary investors are prone to misunderstanding mainframes, thinking of them as obsolete hardware. In reality, IBM Z mainframes still serve as core equipment for critical transactions in many banks, airlines, insurance companies, and government systems, known for their reliability, security, high throughput, and high migration costs.
More importantly, mainframes are not merely hardware sales. They usually come bundled with transaction processing software, which handles high-frequency critical transactions, as well as long-term maintenance services. These software products have high gross margins and stable renewals, forming part of IBM's software business quality.
Quoting IBM's CFO, Reuters stated that the mainframe hardware and transaction processing software stack has dragged IBM's growth down by more than 5 percentage points, compared to the previously expected impact of about 1-2 percentage points. This statement should not be simplistically construed as software alone dragging down performance, but it indicates that the hardware cycle has already transmitted to the software and service revenue rhythm.
For the market, the danger signal is not that IBM is selling fewer mainframes but that the hardware cycle, software licensing, and customer budget decisions have all been simultaneously impacted by AI supply chain disruptions in the same quarter. Income streams that were originally priced separately suddenly appear more correlated.
This is also why investors are concerned about the pressure spilling over to some software and IT service companies. The market is not assuming an immediate downturn for all software; instead, it is reexamining one question: If customers overspend on AI budgets, which traditional system upgrades will be delayed?
Krishna and Idlet's Counterargument Boundary
Krishna's counterargument focuses on rephrasing "disappearing demand" as "shifted transaction timing." He emphasizes that there is no evidence of customers migrating away from mainframes, the z17 project is still strong, Red Hat's growth is also improving, and some delayed transactions have already been completed in early Q3.
This narrative can help alleviate even the most extreme pessimistic view. If the Q3 and second-half delayed orders continue to materialize, the second quarter will appear more like a short-term misalignment caused by AI hardware supply constraints, compounded by IBM's own failure to adjust promptly.
CFRA analyst Brooks Idlet also takes this view. He believes that IBM's software issues are more likely a reflection of the company's hardware problems rather than a trend in the software industry as a whole. The purpose of this view is to try to sever the contagion link between the "IBM plunge" and the "software industry budget crisis."
However, the rebuttal has not completely eliminated the risk. IBM itself has acknowledged that customers shifted budgets toward AI infrastructure at the quarter-end. Zero-sum budget relationships have occurred at least in some large accounts and at certain time windows. The only difference lies in whether it will happen again.
If AI server, memory, and storage supplies remain tight, customers may defer traditional system projects more than once. While a one-time deferral can be recovered, multiple deferrals will alter the annual growth trajectory and may also depress the market's valuation multiples for stable revenues.
Order Fulfillment Will Determine Valuation Recovery
The IBM plunge cannot simply be attributed to "AI killing the mainframe." Current evidence supports a more nuanced assessment: AI supply constraints exacerbated IBM's mainframe cycle and execution issues in the second quarter, leading the market to reevaluate the stability of traditional enterprise IT income.
The validation will come in the third quarter and second half of the year. If the delayed demand Krishna mentioned continues to materialize, dragging down mainframe-related hardware and transaction processing software, the full-year growth can still fall within the new guidance range, and the market will consider this impact more of a timing issue. At that time, the approximately 25% drop on July 14 may have included excessive panic.
Conversely, if delayed order fulfillment does not proceed smoothly, or if more enterprise software and IT services companies mention in their earnings reports that customers are deferring traditional projects for AI infrastructure, Idlet's "IBM-specific hardware issue" explanation will be weakened. The market will no longer trade on IBM's individual misstep but on the sustained squeeze of AI capital expenditures on traditional enterprise IT budgets.
Welcome to join the official BlockBeats community:
Telegram Subscription Group: https://t.me/theblockbeats
Telegram Discussion Group: https://t.me/BlockBeats_App
Official Twitter Account: https://twitter.com/BlockBeatsAsia
