When people tell me they’re diversified because they own an index fund, I understand the logic. You’re spreading money across hundreds of companies. That should smooth out the risk, right? The problem is that owning the index today looks very different from what it looked like five or ten years ago. A small group of names now controls such a large slice of the S&P 500 that calling it a diversified position is, at best, incomplete.
The Magnificent Ten Now Own the Index
In our recent quarterly Scholar Big Picture conversation, we dug into something that has been on my mind for a while: how concentrated the S&P 500 has become. The Magnificent Seven (the large-cap tech names everyone knows) were already above thirty percent of the index not long ago. Add a few more AI-adjacent names and you are likely looking at the top ten stocks representing well over forty percent of the entire index.
That is not diversification. That is a bet on a very narrow slice of the economy, even if the fund prospectus says otherwise.
What Makes This Different from Past Cycles
Concentration has happened before. The dot-com era is the obvious reference point. But when you look at what percentage of the S&P 500 internet stocks actually represented back then, it was nowhere near what we are seeing now with AI and large-cap tech.
Here is why that matters: when the Nasdaq collapsed in the early 2000s, the broad S&P 500 was down roughly thirty to forty percent. But the Nasdaq itself fell seventy to eighty percent, because that is where the overvalued names were concentrated. The S&P functioned as a partial buffer.
If a similar correction happened today, the diversification buffer is thinner. When forty percent of the index is tied to a single thesis, a shock to that thesis does not just hurt the tech sector. It moves the whole benchmark.
The AI Bet Embedded in Everything
What makes the current concentration especially worth watching is that the AI thesis is not just in the obvious names. Major companies across the economy are pouring enormous sums into AI-related capital expenditure, particularly data centers. If you believe the total addressable market for AI is as large as some projections suggest, that spending makes sense. If the projections turn out to be optimistic, you are not just looking at pressure on the dedicated AI names. You are also looking at pressure on every company that placed a large strategic bet on that same thesis.
In other words, owning a “diversified” portfolio that is heavily weighted to an index may mean you have a much larger single-thesis exposure than your allocation percentages suggest.
How We Think About This in Your Portfolio
There are a few things worth considering here. First, it is worth auditing your actual exposure. How much of your portfolio is effectively a wager on AI continuing to deliver? That includes direct holdings, index exposure, and any growth-tilted funds you may own. The sum is often higher than clients expect.
Second, we have been having conversations about tilting toward value over growth as a partial hedge. A portfolio that emphasizes lower-valuation companies is less exposed to the concentration risk that comes with the current index weighting. It is not a magic fix, but it does reduce the all-in-on-one-thesis character of a pure index approach.
Third, this is also why we talk about the role of cash and liquidity. If there were a significant market correction, the historical data suggests that thirty percent drops have historically been reasonable entry points for deploying additional capital into equities. At forty percent, the case becomes even clearer from a historical standpoint. But those historical norms assume a distribution of returns that may look different when the index is this concentrated. That changes the conversation.
What a Correction Could Actually Look Like
Most investors today have not experienced a meaningful drawdown. The financial crisis was nearly twenty years ago. People who were invested during it had far more human capital than financial capital at the time. The felt impact was real but limited compared to what a similar percentage drop would mean today, when portfolios are much larger.
There is a reasonable scenario where a thirty percent pullback in the current environment would land harder than the same number did historically. I am not predicting a crash. I am saying the distribution of outcomes may have fatter tails on both sides than standard historical models would suggest, and that should factor into how your portfolio is structured.
If any of this is prompting questions about your own index exposure, feel free to reach out. These are exactly the conversations we should be having before something moves, not after.
This post is adapted from a recent episode of the Scholar Wealth Podcast. For more perspective on market concentration and portfolio risk, listen to the full podcast episode here.