Wall Street has reached an inflection point on the AI trade, the investment theme that has single-handedly powered all four years of this bull market. Questions about Big Tech's massive AI spending and its place in the economy are dragging stocks down into the year's final stretch, when history shows they should be riding high.
Since the Nasdaq Composite hit a record peak just shy of a month ago, on Oct. 6, the index has slumped more than 7.5% -- and is dangerously close to drifting into correction territory, defined as a 10% pullback from a previous high.
The PHLX Semiconductor index, Wall Street's benchmark for the chip stocks fueling the AI buildout, has already suffered a correction and then some. Since its record high in late June, the Sox has tumbled nearly 28% and is technically in a bear market -- a more than 20% loss.
And the S&P 500's technology sector, the benchmark's heaviest weight, is also in correction territory, closing out Friday with a 11.1% slump from its early October peak.
Three things are behind these astounding losses: mixed earnings from a host of megacap tech companies -- Alphabet, Meta Platforms, Microsoft, and Amazon; the rise in capital costs linked to their massive AI spending plans, and a lack of clarity about all that spending will translate into bottom-line earnings.
And that, in many respects, explains why Wall Street is heading into the holidays with big, dark clouds hanging over it. The S&P 500's gains so far this year are just 10.7%, compared with 16.5% at the same time last year. Then, there's the extremely good chance -- odds are 75%, according to CME FedWatch -- that the Federal Reserve will raise interest rates at its December meeting. And oil are just a headline away from spiking to a price that could stop the economy in its tracks.
BofA Global Research's Claudio Irigoyen knows about the angst because he hears hard questions being asked.
"Investors are increasingly concerned about the AI trade, and the more capex keeps growing, the more risks loom," said Irigoyen, who is head of global economics research. "Will firms be able to monetize their investments, or could the speed of adoption disappoint?"
And RBC Capital Markets began having serious doubts about the tech sector a couple of weeks ago, downgrading it to Market Weight because of its "tendency to lag in broader market pullbacks that usually accompany the start of a Fed hiking cycle."
Still, there is the glass-half-full perspective.
Deutsche Bank raised its rating on the sector back to Overweight on Friday after downgrading it to Neutral just three weeks Ago. Tech's long October slump, the bank's analysts explained, has created an attractive entry point for investors.
Citigroup's take is that investors are "underestimating AI-driven revenue growth at the hyperscalers." The bank projects 30% revenue growth next year on what it calls "accelerating enterprise adoption to sustain the AI trade."
So where does the lack of consensus leave stocks? Up or down between now and the end of the year?
Of course, it depends on who's answering the questions.
For example, Freedom Capital Markets' Jay Woods thinks investors are simply becoming more thoughtful about what stocks they buy.
"The market has been on a wild ride," said Woods, who is the brokerage's chief market strategist. "October left treats for the value investors, but pulled the rug out from many high growth ones."
"But investors aren't necessarily abandoning stocks," he added. "They're becoming much more selective about which ones deserve their money."
Melius Research's Ben Reitzes, on the other hand, thinks a slowdown is inevitable because of the heavy borrowing by AI's biggest players.
Reitzes calls their capital needs the market's primary risk factor, trumping those tied to power, chip supplies, and politics.
Yet, he writes, those same companies don't realize that the capital shortage will ultimately cause a slump. Martin, are we talking market slowdown or slowdown in borrowing.
"We have at least five trillion-dollar plus companies (soon to be seven) that have to issue hundreds of billions in securities (stock, bonds, etc.) and they still need the help of their suppliers, private-equity firms and banks," he said on Monday. "All this to fund a multi-trillion buildout in an inflationary environment."
This year has seen $655 billion in new borrowing, an unprecedented amount for the sector.
Hyperscalers are switching to equity markets for their capital needs. For example, Oracle said Monday that it plans a $20 billion bond plan and $10 billion stock sale. shares sales. Last month, Facebook parent Meta Platforms announced a massive $95 billion equity raise. Earlier this year, SpaceX raised $85.7 billion.
For those companies that don't have an investment-grade ranking from a major credit-rating firm, borrowing costs have double this year -- to the highest level since the pandemic.
Saxo Bank's Charu Chanana boils down all that's going on to this: Can Big Tech turn all that AI spending into actual revenue?
"So far, much of the AI investment cycle has focused on building capacity: chips, data centers, networking, memory and power infrastructure," said Chanana, chief investment strategist.
"The next phase should increasingly focus on what businesses can actually achieve with that infrastructure," she said. "And there is still limited evidence of broad, measurable productivity gains across the wider economy."
Trading was mixed on Monday. In afternoon trading, the Sox was down about 2.4%, but the Roundhill Magnificent Seven ETF was up 0.2%. The Nasdaq slipped 0.3%.
That suggests things are on hold for a while -- until Tuesday's midterm elections and September quarter earnings from Apple, Advanced Micro Devices, and Broadcom this week.
But what is clear at this point, and for the foreseeable future, is that tech isn't the tide lifting all boats. And that radically changes the waters investors have to navigate.
Comments