Despite recent volatility in artificial intelligence (AI) related stocks, the current capital expenditure (capex) cycle has not yet peaked and companies supplying AI hardware still have strong earnings visibility over the next year.
This is according to Jean-Louis Nakamura, head of conviction equities at Vontobel Asset Management, who told FSA in an interview that ongoing fears surrounding the AI trade have become excessive.
“We believe the capex cycle will continue for some time,” he said. “The companies engineering the frontier models AI are doing this because they believe the leading company in this new extraordinary market will win it all.”
“So it’s a race for monopolistic power to try to convince as many as possible clients to join their [AI] model to become more or less prisoner of their own ecosystem,” he said. “Once they are a prisoner of the ecosystem, they will be in better position to set their price.”
“Exactly the same way that Microsoft with the cloud, or Amazon with the cloud have done this before.”
As such, he sees the current capex cycle continuing for at least the next two years, with the ratio of capex to sales peaking at the end of next year, but even after that, for it remaining high compared to history.
He said: “This explains the tensions in the semiconductor and memory market, because investors see it with a lens of a super cyclical market where the cycle is not very long, it lasts two years, then oversupply comes, demand slows down and prices collapse.”
“This time is different. Not only you have super unusual strong demand, but you have super tight capacity that will be very difficult to expand within the next two and a half years.”
He pointed to some segments of the market, such as high bandwidth memory and substrates, which have seen their lead times move from three months to a year, with no clear easing in sight before even 2030.
“I think that semiconductors and memory manufacturers have been punished by the market over the last few months because of this fear that capex will slow down and maybe we were at the peak of the semi market a bit prematurely,” he said.
“We believe that there is a potential for those companies to again surprise the market with not only very strong result in absolute terms, but with positive surprises versus expectations.”
Diversification is difficult
However, despite his confidence in the AI hardware trade, Nakamura acknowledges the difficulty investors face in trying to diversify their exposure to AI.
“If you want to diversify away from AI trends, it’s not so obvious,” he said. “What we see today in the market is that there some part of the healthcare sector, especially hospital services and some segment of biotech that could be interesting.”
“Financials have done well as long as interest rates were rising, but we tend to believe that we are probably at the peak of long-term interest rates.”
“Of course, other industrial sectors aside from AI might do well, only if the growth in the manufacturing sector broadens out.”
He said that although there are some pockets with good earnings visibility, such as in automation, the segment which has the better visibility over the next 12 to 18 months is hardware related to AI and semiconductors.
But Nakamura also made the case that investors can further diversify by going deeper into emerging markets, which have been surprisingly resilient to recent macro shocks.
“We had the shock on international trade with the rise in U.S. tariffs, this spring we had the shock from the Middle East with a rise in oil prices, and more recently we had the shock with a sudden rise in long-term yields,” he said.
“This kind of shock, 20 years ago, will have sparked a panic in emerging markets, fall of emerging currencies, massive capital outflows, local central banks forced to tighten interest rates, and of course, massive underperformance of the equity market.”
“This time, not only nothing happened, but emerging markets continue, except maybe over the last few weeks, to outperform persistently developed markets.”
He believes one reason for this has been due to the transformation of most of the emerging market economies over the past 15 years.
He pointed to their more mature domestic demand, more robust macroeconomic conditions with lower more stable inflation and relatively balanced external accounts.
“The situation of those economies has transformed radically, and today their dependency in terms of external trade to other developed economies is much more limited than it used to be the case,” he explained.
“The way investors should look at them is not the same way they would have looked at them 15 years ago,” he said. “15 years ago, emerging markets was this small pool where you put assets to chase for high return as a counterpart of high risk, high volatility.”
“Today they have matured, and they should be more as a way to diversify your global equity exposure.”