Stocks dropped on Friday not because anything especially “bad” happened, but instead because the consistent drip of positive AI and economic news was interrupted. And given the market’s stretched condition, it led to some profit-taking.
The two events most directly responsible for the declines on Friday were AVGO earnings (which weren’t bad, but weren’t as good as expectations) and the jobs report, which was a touch “hot.” And while it’s “fine” for the economy, the market is on edge about rate hikes, and the strong number boosted those expectations.
However, there was another reason for the pressure on tech late last week that’s not getting as much focus as it should: a looming oversupply of AI shares. There are three massive “AI” public offerings looming: SpaceX (imminently), OpenAI, and Anthropic (later this summer/early fall). SpaceX, while not exclusively an AI company, houses xAI, which is a large part of the SpaceX offering.
The amount of capital being sought by these three firms is enormous: $75 billion by SpaceX and around $60 billion from both OpenAI and Anthropic. That’s nearly $200 billion in share supply coming to market in the next several months. Keep in mind that’s on top of the $80 billion share secondary offering GOOGL just announced.
For context, the largest IPOs we’ve seen in the U.S. market have all been less than $20 billion, so the scale here is simply enormous. Now, there is a quasi-insatiable demand for AI companies and SpaceX, but there are legitimate fears that this amount of share supply could 1) cause forced selling in other parts of the market as there simply isn’t enough money to absorb the massive offerings, and/or 2) create excess tech/AI capacity that exceeds investment demand and causes weakness across the market.
This fear gets back to a simple truth about markets: Stocks go up because there are more buyers than sellers. With nearly $200 billion in share supply looming in the next few months, we are going to need a lot of buyers to absorb it all. To be clear, that doesn’t mean people should get outright bearish on AI. Share supply won’t impact the key question on AI, i.e., whether all this capex continues and produces a positive ROI, return on investment.
But for a market that is undoubtedly stretched, and for a tech sector that’s gone quasi-parabolic, the looming supply deluge is a near-term negative we must consider.
Practically, I think that means understanding this market can pull back and quickly (easily 5% or more), but still have very positive underlying fundamentals. And if that happens in the coming weeks as we approach these listings, it shouldn’t shock anyone. From a “hedge” standpoint, allocating to non-tech parts of the market (healthcare, industrials, defensives, RSP, value, and international exposure) will help insulate gains from any tech supply-driven selloffs (these are the sectors that traded well amidst AI weakness early this year).
In January this year, we did just that. We moved from tech-heavy segments of the market to bonds, value, mid-cap, and international. We continue to believe that allocation will weather any storm that we will face this year and still participate when the market moves up.