Portfolios concentrated in U.S. tech stocks should diversify to real assets before the AI bubble bursts, according to Cambridge Associates.
A new white paper from the investment advisor notes that the U.S. now represents roughly 64 percent of the MSCI All Country World Index, up from about 42 percent in 2010. Celia Dallas, chief investment strategist and partner at Cambridge Associates, told Institutional Investor that artificial intelligence “has driven this concentration.”
While Dallas sees a lot of opportunity in AI, that doesn’t mean everything is priced to have strong performance. “A lot of high expectations are already in the market,” she said, noting that most of the top 10 stocks are in AI and account for roughly a quarter of global equities.
“Looking at how investors get concentrated in the best performers of the past tends to increase risk in ways you don’t anticipate,” she added.
Amid this significant technological shift, Dallas expressed concern over the inevitable overbuilding that occurred with the railroad and telecom booms. “It would be shocking to have a massive technological shift without massive overbuilding,” she said. In addition, she argues that the early winners in AI aren’t necessarily going to be durable long-term players (see also the rise and fall of boom-and-bust brands like Pets.com, Nokia, and Friendster).
While forward earnings multiples for the Magnificent Seven have come down meaningfully from their peaks, investors are still paying high prices for a small set of companies expected to deliver the bulk of future growth.
Dallas acknowledges that it can feel riskier to move away from large-cap tech stocks especially when so much money is going into these investments. But “having that concentration in portfolios can be pretty dangerous to long-term expectations” when the tide inevitably turns.
“It’s hardest to rebalance when you most need it,” she added.
Investors have overallocated to a U.S. market that has become unusually concentrated—the U.S. represents roughly 64 percent of the MSCI All Country World Index, with the top 10 U.S. companies, most of them technology related, accounting for 24 percent. Cambridge forecasts that once the enthusiasm for U.S. equities fades (like it did in the 1990s), the U.S. dollar will likely fall as well.
One idea Dallas likes is investing around electricity grid resilience, like lithium-ion batteries that can support the energy transition. Electricity infrastructure and grid modernization sits squarely in the area of shortage and is not solely relying on AI to be profitable. As a result, real assets, which have declined significantly in investor portfolios over recent decades, should also play a larger role to provide the infrastructure needed to support a more capital-intensive and “electrified” economy.
Meanwhile, within equities, while not cheap in absolute terms, Cambridge is tactically leaning into global and non-U.S. small-caps, values and cyclical stocks. “Don’t focus on the rearview mirror,” Dallas said. “Evaluate strategically how portfolios are prepared for the next decade rather than what’s happening now.”
