According to The Wall Street Journal, the U.S. market has repeatedly produced bubbles around specific sectors and themes in recent years, but the collapse of those pockets has usually not pulled down the broader stock market.
The latest memory-chip bubble rose and broke within four months
The report said the latest memory-chip bubble expanded rapidly and then burst within about four months. The move came with sharp volatility and a crisis at one hedge fund. Even so, the S&P 500 remained only 1.6% below its record high, while the equal-weight S&P 500 set a new high last week.
Pullbacks in AI-related stocks have also been almost fully offset by gains in other sectors.
A series of theme-driven bubbles over the past decade
Over the past decade-plus, the U.S. market has gone through bubbles in 3D printing, China-related stocks, low-volatility products, SPACs, clean energy, cannabis, space, crypto assets and AI-related shares.
Several names cited in the report posted steep declines. Strategy has fallen 83% from its peak, Trump Media shares have dropped 89%, and SK Hynix was at one point down 55% before rebounding last Friday.
Loose money and leverage amplified price swings
The Wall Street Journal said easy money, speculative demand and expectations tied to new technologies helped fuel those bubbles. In recent years, margin debt and leveraged ETFs have added to the volatility.
Losses have mostly stayed with investors, not banks
One reason these localized bubbles have not caused severe economic damage is that most of them were not financed with large amounts of debt. After they burst, losses were borne mainly by investors, and the banking system did not take a meaningful hit.
Macro strategist Russell Napier said the banking system remains healthy, leaving credit available to help create the next bubble.
AI investment could move the market into a more dangerous zone
The report said AI investment is pushing markets into a more dangerous area. Data-center spending over the next four years could reach $7 trillion. If productivity gains from AI do not justify investment on that scale, capital misallocation could seriously damage the economy.
As AI buildout becomes more dependent on debt financing, a broader AI investment bubble, if eventually proven to be one, could hit the financial system when it breaks. In that case, the wider market would be unlikely to stay insulated.