Three weeks of semiconductor pain had a clear pattern: the selling was technical, driven by leverage unwinding, profit-taking, and rotation. Monday's selloff is different. Three new and simultaneous pressures arrived over the weekend, and unlike the prior correction, each one has a direct path to higher inflation and therefore to a more hawkish Fed.
Trump posted on Truth Social reinstating what he called "THE IRANIAN BLOCKADE," explicitly framing it as stopping Iran's ships and customers from entering or leaving the strait. Separately, Trump announced a new 20% "cargo tariff" on imports, a direct goods-price inflation shock arriving the day before June CPI prints. Iran declared Hormuz closed and began proxy-targeting refineries against US allies including Bahrain and Kuwait, widening the conflict beyond the strait itself. Ship-tracking data from Kpler showed only six vessels transited the waterway on Sunday, the lowest count in five weeks. WTI crude surged 4.6% to nearly $75 a barrel at the open. The KOSPI fell nearly 9%, dipping below 7,000 for the first time since early May and reentering bear market territory, 25% below its June 22 peak. SK Hynix posted its biggest single-day decline on record in Seoul, more than 15%, one trading day after its Nasdaq debut. The S&P 500 lost 0.6% and the Nasdaq fell 1.2%.
The prior three weeks of selling did not have a closing catalyst. This week has three, all pointing in the same inflationary direction.
What the Hormuz Data Actually Shows
The conflicting narratives from Washington and Tehran are less important than the ship-tracking numbers. Trump said the strait remains open to commercial traffic. Iran said it is closed. Kpler's data aligns with Tehran's position: six vessel transits on Sunday is a number that functionally describes a closed waterway, not an open one. The question is not who is claiming what. It is how many barrels are actually moving.
The risk transmission is not just about oil prices. Tanker earnings rose to $340,000 a day last week, $50,000 above the prior week, and that number reflects the supply of willing shipowners, not just the price of crude. When shipowners won't transit regardless of what any government says about the strait being "open," the physical supply chain breaks regardless of the diplomatic posture. Insurance markets, not press releases, determine whether tankers actually move.
The context that makes this escalation more dangerous than the previous rounds: the Oman-mediated transit corridor that served as the primary commercial alternative during the June ceasefire period is now less certain. If that corridor also comes under threat, there is no remaining safe routing out of the Persian Gulf for Gulf producers. Iraq had already been forced to reduce output at some fields in June because it could not find enough ships to load its cargoes. A full Hormuz closure reproduces and extends that dynamic at scale.
The 20% Cargo Tariff: The Most Overlooked Pressure of the Day
While Hormuz and the KOSPI collapse are dominating headlines, the cargo tariff announcement may be the most consequential development for Tuesday's CPI read-through. A 20% tariff on imported cargo is a direct and broad-based goods price shock. Unlike an energy tariff or a sector-specific measure, a cargo tariff applies to the cost of physically moving goods across borders, which means it compounds on top of whatever underlying goods prices are already doing.
The FOMC minutes released last week explicitly noted that "core goods price inflation had risen relative to a year earlier, which the staff judged as largely reflecting the effects of tariffs and AI-related price pressures." A new 20% cargo tariff adds a fresh tariff layer on top of whatever the June CPI print shows tomorrow. Even if June comes in at or below the 3.9% consensus headline, the market now knows that July's data will incorporate a new goods-price impulse that was not in the picture when the Fed met in June.
This is why the cargo tariff matters more than it might seem in isolation. It changes the trajectory after the CPI print rather than just the print itself. Warsh's testimony at 10:00 a.m. Tuesday will almost certainly be asked about it directly, and his answer, specifically whether the Fed views cargo tariff-driven inflation as transitory or persistent, will be the most actionable thing he says all week.
SK Hynix Down 15% on Day Two: The IPO Premium Vanishes
SKHYV opened last Friday at $170, 14% above the $149 offer price. By Monday, the stock is trading near $156, down roughly 6.7% on the day and down about 8% from its opening. The 15% collapse in Seoul's SK Hynix common shares pulled the ADR lower through the arbitrage mechanism, and the direction of flow is the opposite of what the listing was supposed to demonstrate.
The irony is stark. SK Hynix's Nasdaq debut was framed as a validation event for AI memory demand and a catalyst for the memory sector's recovery from its three-week correction. Instead, it listed on Friday and fell on Monday alongside a KOSPI that is now 25% below its June peak. The IPO premium that opened at 17% has compressed to approximately 5%, and whether it holds depends on whether the Hormuz situation stabilizes before Tuesday's CPI print.
The SK Hynix situation illustrates a structural problem with the memory sector's current setup. The stock's fundamental case, 58% global HBM market share, the strongest contracted revenue base in the company's history, and Q2 earnings on July 29 that are expected to show the strongest quarter in company history, has not changed. But the multiple at which those fundamentals are valued is being compressed by two simultaneous forces: a geopolitical energy shock that raises the Fed's interest rate path, and a wave of leveraged ETF products (SKUU and SKDD began trading Monday) arriving on the stock's first full trading day while it is already under maximum pressure.
The Week's Real Stress Test: CPI and Warsh on Tuesday
Everything that happens today is noise relative to what arrives Tuesday morning. June CPI prints at 8:30 a.m. ET. Fed Chair Warsh delivers his first congressional testimony at 10:00 a.m. ET. Four major banks report before the market opens: JPMorgan ($5.59 EPS, $51.09 billion revenue consensus), Bank of America ($1.13 EPS, $30.78 billion), Citigroup ($2.72 EPS, $23.74 billion), and Wells Fargo ($1.73 EPS, $21.86 billion).
The CPI consensus is headline at 3.9% year over year with core at 2.9%. Those numbers would represent meaningful deceleration from May's 4.2% headline and would confirm the thesis that Hormuz normalization in early June was pulling energy costs lower. The problem is that the normalization that was supposed to show up in June's data is now being reversed. The ceasefire collapsed on July 7. June's data was collected before that collapse. The CPI print on Tuesday will show a world that no longer exists, and Warsh's testimony immediately after will be the first opportunity to explain what the Fed thinks about the gap between the data and the current reality.
The FOMC minutes released last Wednesday confirmed that the committee identified AI infrastructure demand as a current contributor to core goods price inflation and that "some policy firming would likely be warranted" to return inflation to 2%. Those minutes were written when Brent was in the low $70s and the ceasefire appeared functional. Brent is now at $76 and the strait is effectively closed again. The distance between the world the FOMC minutes described and the world Warsh faces Tuesday morning is the most important macroeconomic fact of the week.
If Warsh signals that the Fed can "look through" the energy shock because it views it as transitory, rate expectations will ease and the equity market will find some relief. If he signals that persistent supply-side inflation from Hormuz makes the September hike more likely, the 10-year yield holds above 4.56% and equity multiples remain compressed. The CPI number at 8:30 sets the table. Warsh's reaction at 10:00 determines the direction.
The Software Rotation: Where the Money Is Actually Going
The one constructive signal in today's tape is the internal market rotation that the hardware pain is enabling. Software names including IGV, Palantir, Salesforce, ServiceNow, Adobe, Reddit, and Shopify are holding up or advancing as institutional capital rotates out of high-beta memory and semiconductor names into AI software plays with more defensive earnings profiles. This is the same infrastructure-to-application rotation that showed up briefly in early July, now amplified by the Hormuz re-escalation.
The logic is straightforward. Software companies do not have physical supply chains exposed to Hormuz. They do not import cargo. Their margins are not directly compressed by oil at $75. When geopolitical risk rises and tariffs add cost inflation to hardware supply chains, software becomes the AI trade's defensive expression. The broadening of market breadth into software is actually a healthy signal for the overall market structure, even as the semiconductor names bleed. It means the AI thesis is rotating within the ecosystem rather than exiting it entirely.
History suggests this pattern is seasonally reinforced. The past five years have seen July consistently serve as a consolidation month for the highest-momentum names, with high-beta stocks underperforming while the broader market digests the first half's moves. The difference this July is that the consolidation has a specific and ongoing geopolitical catalyst compressing the high-beta names rather than just normal seasonal rotation. When Hormuz stabilizes, the mean-reversion trade from software back into semiconductors could be sharp.
TSMC Revenue Data: The One Bright Spot
Into a market dominated by geopolitical fear, Taiwan Semiconductor Manufacturing published a notably constructive data point Monday morning. TSMC reported a 67.9% year-on-year rise in June sales, bringing first-half 2026 revenue to 2.4 trillion new Taiwan dollars, approximately $75 billion, representing 35.6% growth compared to the same period in 2025. TSMC reports full Q2 earnings Thursday, where management's commentary on AI chip demand and capacity utilization will serve as the most comprehensive real-time assessment of whether the AI infrastructure spending cycle is intact.
TSMC's June revenue number is important because it is the freshest available data on actual semiconductor demand, as distinct from stock prices that have been compressed by leverage and geopolitical sentiment. The divergence between TSMC's 67.9% revenue growth and the KOSPI's 9% Monday collapse is the clearest available illustration of the gap between the fundamental business and the market price. That gap does not close in a single day, but it provides the analytical anchor for understanding what is actually happening in the sector versus what the stock prices are temporarily pricing.
Why This Keeps Happening and When It Stops
The pattern of the past three weeks has been escalation, partial de-escalation, re-escalation, with each cycle arriving just as market sentiment was beginning to stabilize. The structural reason is the one that Height Securities' director of research identified last week: the Trump administration "faces two dueling imperatives around acquiescing to Iranian regime plans to control maritime flows in the Strait and nuclear ambitions" and "has avoided decisively addressing these imperatives." Until one of those imperatives is resolved, the cycle continues.
The market's stabilization condition requires at least one of three developments: a verified and durable reopening of Hormuz transit, a CPI print that shows inflation decelerating fast enough to take September's hike off the table, or a set of earnings reports from the bank sector and TSMC that demonstrate the underlying economy is strong enough to absorb higher rates without credit deterioration. This week provides all three potential catalysts simultaneously.
The base case from the WeekAhead report, a 50% probability scenario, envisions CPI and PPI near consensus, Warsh keeping language conditional, and Hormuz remaining impaired but not fully closed. In that scenario, the S&P 500 holds between 7,500 and 7,600. The bear case, now at 25% probability, involves core CPI above 0.3%, PPI surprising higher, and Warsh giving no assurance the Fed can look through supply risk. In that scenario, the S&P 500 breaks below 7,500 and the 10-year yield moves above 4.60%.
The market is not broken. It is stuck in a corridor between a fundamental case for AI infrastructure that remains structurally intact and a geopolitical inflation shock that has no clear resolution timeline. The corridor has two exits: a diplomatic one that reopens the strait, and a data one that shows inflation decelerating despite the energy shock. Tuesday provides the clearest view yet of which exit is closer.
Explore more exclusive insights at nextfin.ai.
