Seriously, stop worrying about market concentration

This article is an on-site version of our Unhedged newsletter. Premium subscribers can sign up to get the newsletter delivered every weekday. Standard subscribers can upgrade to Premium here, or explore all FT newsletters

Good morning. At some point today, we should know the results of the US midterm elections, or at the very least a sense of whether we have a blue wave or a purple haze on our hands. Whatever the outcome, here’s a quick list of what to watch:

• The Iran war: Will it end or escalate, and what will that do for oil and inflation?
• The AI infrastructure build-out: A rare bipartisan issue in this election, with links to state-level affordability campaigns and consequences for growth and jobs.
• Regulating AI model developers: Donald Trump thinks it isn’t necessary, and the AI bosses are divided, but a change in congressional control may create an opening for federal regulation of AI safety.
• Financial regulation: The Trump administration has been famously relaxed about crypto and prediction markets. Can this continue?
• Fiscal policy: Flipping the House, the Senate or both would create the conditions for changes in federal taxes and spending.

The Iran war: Will it end or escalate, and what will that do for oil and inflation?

The AI infrastructure build-out: A rare bipartisan issue in this election, with links to state-level affordability campaigns and consequences for growth and jobs.

Regulating AI model developers: Donald Trump thinks it isn’t necessary, and the AI bosses are divided, but a change in congressional control may create an opening for federal regulation of AI safety.

Financial regulation: The Trump administration has been famously relaxed about crypto and prediction markets. Can this continue?

Fiscal policy: Flipping the House, the Senate or both would create the conditions for changes in federal taxes and spending.

We’ll be back tomorrow with analysis of the outcome and the market reaction. In the meantime, here’s something different. Email us: unhedged@ft.com.

Not from concentrate

Worrying about market concentration is a hard habit to break. It’s the kind of thing analysts, investors, Unhedged readers (and writers) bring up all the time. We all know that a tiny clutch of stocks accounts for a massive chunk of US stock market capitalisation and earnings growth, and that you can see the same phenomenon all over the place, including in emerging markets. And we know it somehow gives us the ick.

Right now, given the recent stumbles in AI-related stocks, the hand-wringing over concentration is particularly common, again. So it’s worth looking at the extent to which it really matters. Let me get to the punchline first and say it doesn’t, at all.

Mark Kritzman of, among other places, MIT Sloan, and David Turkington of State Street Associates put out a paper on this last year called The Fallacy of Concentration. (It’s free to read if you’re interested.) It forms another part of the literature telling people to stop worrying about this — a message that people seem determined to ignore.

The title of the paper is slightly odd, given they find that concentration is very real. Looking at the effective number of stocks based on their historical weights in the S&P 500, it is clear that the index is more concentrated than it has been at any time since 1998. This chart taken from their paper shows sector concentration, stock concentration and a combination of the two. (Lower values mean higher concentration. If that confuses you, turn your phone upside down.)

Looking at longer-run data in a similar way delivers a similar result and shows that levels of concentration this high are not unprecedented, but they are unusual — previous similar episodes date back to the late ‘70s/early ‘80s and the 1950s.

OK, so, say you decide this is a problem and you want to hedge against it. The paper devises a trading strategy to tilt more into cash in higher-concentration periods, and more into stocks at other times. If you were doing that in real life today, you’d have a tiny allocation to stocks, and obviously, you’d be getting hosed compared to just holding stocks and ignoring the concentration levels.

“In the US stock market, the buy-and-hold strategy generated more than twice as much wealth as the dynamic strategy from January 1936 to June 2025, and it did so with less risk. In five other major stock markets, from January 1983 to June 2025, counteracting concentration was detrimental to performance on average, and where it was not, the benefit was insignificant,” the report says. Concentration in itself also tells you nothing about returns, volatility or drawdowns.

Why might this be? The paper posits several possible explanations, including that bigger companies are generally more diversified in product and geography, and maybe even better run. Maybe it’s also just “natural” for winners to take the spoils. They also have better access to capital and advice and so on. It might be more pronounced right now simply because the winner-takes-all dynamic is more forceful in the AI trade and because the whole world wants a slice of what these stocks are offering, and a lot of them happen to be in the same place.

Now, this paper is about concentration. It’s not about bubbles. It’s certainly plausible that tech and AI are in a bubble, and that’s a whole other issue, particularly in light of the concentration. Anything from higher borrowing costs to credit screw-ups to regulatory clampdowns or normal stuff like revenues, moats and boring old access to electricity could pierce an AI bubble, if that is what this is. But that’s another story entirely, albeit a darker one given the levels of concentration at play. (If a lower-value, more boring sector than tech were to collapse tomorrow, you have to imagine it would leave less of a mark.)

The point suggested here, though, is that in itself, high concentration tells us nothing useful, and that trying to lean against it is a waste of time.

One good read

Marauding toads

FT Unhedged podcast

Can’t get enough of Unhedged? Keep following Rob Armstrong and Katie Martin’s incisive and unfiltered analysis on the Unhedged podcast, a twice-weekly, 15-minute lively dive into the latest markets news and financial headlines. New episodes every Tuesday and Thursday.

Recommended newsletters for you

Due Diligence — Top stories from the world of corporate finance. Sign up

The AI Shift — John Burn-Murdoch and Sarah O’Connor dive into how AI is transforming the world of work. Sign up