Why Booming Chip Demand May Not Be Great News for the Stock Market
Micron surpassed elevated expectations, but the Mag 7 is paying a price.
You've reached your free article limit
You've read 0 of 1 free Pro articles.
The chips are ripping, and the indexes are higher, but this may not be the good news that investors are hoping for.
Micron Technology (MU) is up over 16% in premarket trading after a blowout fiscal Q3 report. Qualcomm (QCOM) is up over 12% after raising guidance and announcing it will move beyond smartphones into data-center chips and servers. Other chip names, including Sandisk (SNDK), Western Digital (WDC), Lam Research (LRCX), KLA Corporation (KLAC), and Applied Materials (AMAT), are higher in sympathy.
Meanwhile, the Roundhill Magnificent Seven ETF (MAGS) is barely moving. The hyperscalers that buy what the chip companies sell are not participating in the rally. Nasdaq 100 futures are up over 2% but Dow futures are flat. The chip suppliers are winning, but the AI names that are buyers are stagnating.
The Chip Sellers Win and the Buyers Lose
What is good for Micron may not be good for the companies that buy its chips. Micron’s revenue of $41.46 billion came out of someone’s capex budget. The 84.9% gross margin on those chips reflects the chip seller’s pricing power, directly eroding the chip buyer’s margins.
Micron’s gross margin was 39% a year ago and is 84.9% today. That is the kind of shift in pricing power that only happens when demand vastly exceeds supply and the buyer has no alternative. Micron has near monopoly pricing power and that is not good if you are a buyer.
The buyers do not have an alternative. The hyperscalers must have the chips to maintain their competitive position in AI. They will pay whatever the supplier charges because the alternative is falling behind in the AI race. The chip buying is not optional even if it means thin margins and limited profits.
The Third Wave of Inflation
The bigger implication of the pricing power in semiconductors is that it is inflationary. The report from Micron puts the issue front and center and is creating what The Wall Street Journal calls the third wave of inflation.
The first wave of inflation was caused by supply-chain disruption due to the pandemic and tariffs. The second wave came from the Iran war energy shock. The third wave is the data-center buildout driving up prices for components and electricity. This is no longer just a stock-specific story about names like Micron and Qualcomm. It is a structural inflation story that the Fed has to deal with on top of the first two waves.
The scale of the demand is what makes this different from past tech investment cycles. Capital spending by the five hyperscalers will reach an estimated $741 billion this year, up nearly 75% from last year. That single year of incremental spending is larger than the entire global semiconductor industry’s annual revenue from two years ago. Eighty-one percent of professional economists now expect the AI buildout to add to inflation over the next year.
The problem is that supply cannot be created fast enough to absorb the demand. New fabs take three to five years to come online. Micron itself said its next-generation DRAM and NAND will not begin volume production until the second half of 2027. The new SK Hynix capacity from its $30 billion U.S. listing will not produce meaningful output until 2028. That means this pricing pressure is going to last for years rather than months.
Why the Fed Cannot Look Past This
Fed Chair Kevin Warsh’s hawkish tone last Wednesday makes more sense when considered in this context. He said the Fed needs to “fix five years of misses on inflation.” He sees what is coming. Nine of 18 officials now see at least one hike this year. The political pressure to cut rates that President Trump put on Warsh collides with the inflation that is now hitting from a different source than the energy shock that drove it earlier this year.
The squeeze on the AI buyers gets worse from here. Higher input prices from chip suppliers and higher interest rate costs from a Fed that is preparing to hike rather than cut. Margins will compress and that will cause multiples to compress. That is why the Mag 7 is not participating in the rally Thursday morning. The market is pricing in the margin squeeze, and the chip suppliers are the only ones that are benefiting.
Strategy
My defensive posture remains in place. My cash levels are around 45% to 50%. I am not buying the chip rally. I do not believe that the strength Thursday morning will be sustained for long. If I’m wrong I’ll reassess but I have no FOMO.
The rotation I have been writing about remains my focus. Biotech continues to work because it is insulated from the AI capex story. Distributed power plays like Sunrun (RUN), which I discussed Wednesday, fit into the same structural buildout that is driving the chip demand, because if memory supply is constrained, then electricity and power capacity is also constrained. Electrical infrastructure, industrial automation, cooling and thermal management, and the names that supply the data-center buildout without depending on AI software profits are all part of the same theme. That is where the opportunities lie.
Speaking of inflation, the May Personal Consumption Expenditures report will be released at 8:30 a.m. ET. Headline PCE is expected at 4.1% year over year and core is expected at 3.4%. Both are higher than in April. The data is likely to confirm the inflation trajectory that chip spending is creating which gives Warsh more reason to lean hawkish at the next meeting.
This is not a market for chasing the obvious bullish story. The Micron report is bullish for Micron. It is bearish for the broader stock market. The demand that produces Micron’s huge revenue growth is squeezing the margins and the financing costs of every other name in AI.
Don’t be fooled by the Micron celebration. The fact that they are a near monopoly is not good news for the rest of the market.
At the time of publication, Rev Shark had no positions in any securities mentioned.
