To diversify, I bought another ETF. My account now holds a broad U.S. benchmark index, adds growth stocks, and puts some money into semiconductors as well. With more product names in the account, it feels like the risk has been spread around evenly.
But what if the companies you already own keep showing up among the top holdings of the ETF you just bought? The number of products has grown, yet the companies your money actually leans on may not have changed as much as you think. Investor education materials from the U.S. Securities and Exchange Commission (SEC) also advise checking whether the top holdings differ when you own multiple funds or ETFs.[1]
Pzena Investment Management takes this problem a step further in its third-quarter 2026 report. Its point: moving like the index and diversifying risk are not the same thing. The claim is that in an environment where the growth story of AI explains the share prices of many companies at once, the number of holdings alone cannot tell you what kind of account you have.[2]
What Matters More Than Product Names Is Where Your Money Actually Sits
A single ETF contains many companies. But containing many companies is not the same as holding similar amounts of each. In indexes like the S&P 500, where weights are set by market capitalization that reflects the number of shares actually available for trading, larger companies exert a greater influence on how the index moves.[3]
Adding one more ETF on top does not automatically increase the diversification benefit, either. If both products hold the same company, you need to look at your combined exposure to that company across the whole account. That is why, rather than being reassured by each product's individual weightings, you should examine them in a way that reflects how much money you put into each product.[1]
In an analysis as of the end of September 2026, Pzena put the combined weight of the S&P 500's top 10 companies at roughly 41%. That means the choice to buy the entire index comes bundled with considerable exposure to its largest names. Keep in mind, though, that this is the figure as of that report's reference date, not the weighting at this very moment.[2]
So when assessing whether you are diversified, the first question should be closer to “How much of my money is in which companies?” than “How many ETFs do I own?”
The Other Face of the Claim That a Fund Holds Thousands of Stocks
One interesting metric appears in Pzena's report: the “effective number of stocks,” or Effective N. Pzena explained that the weight concentration of the MSCI All Country World Index (ACWI), which holds roughly 2,400 companies, is similar to that of a portfolio of about 100 stocks held in equal amounts.[2]
This passage needs to be read with care. It does not mean the 2,400 companies actually behave exactly like 100 companies. Nor does it mean the money invested in the other companies disappears.
According to MSCI's definition, the effective number of stocks is the reciprocal of the sum of each stock's squared weight. Put simply, it is a metric that translates how evenly your money is divided into “the number of stocks you would hold at equal weights.” It is not a risk measure that reflects the correlations between companies' share prices.[4]
That said, it is not a meaningless number, either. It reveals the concentration in weights hidden behind explanations that tout a large number of holdings. But if you jump straight from here to conclusions like “a crash is imminent” or “the actual diversification benefit is exactly equal to a specific number of stocks,” you have gone beyond what the metric can tell you.
When Different Companies Lean on the Same Expectation
After checking the weights, another question remains: do the earnings of these different companies really move for different reasons? This is the point where Pzena brought up AI. Even if they are separate companies, they may all be leaning on the same common factor—the continued expansion of AI investment.[2]
For example, if you are analyzing AI capital spending, you would want to look at both the rising orders on the component-supplying side and the investment burden on the side spending the money. The same expansion in investment can be a revenue opportunity for one side and an outlay that has to wait for monetization on the other. This is not a claim that every related company's share price moves in the same direction; it is an argument for looking at where each company stands within that flow.
In my view, lumping companies together under the label “AI-related stocks” is just as crude a judgment as counting the number of holdings alone. You have to distinguish who their customers are, where their earnings come from, and which part of the business wobbles first if investment slows or monetization is delayed.
What you are ultimately checking is not just the difference in company names. It is how different the conditions are that the earnings and valuations of the companies you own depend on.
Does Moving With the Index Mean Less Risk?
The title of the Pzena report that this post started from contains the phrase “tracking error is not diversification.”[2]
Tracking error is a metric that looks at how much the return gap between a portfolio and its benchmark index fluctuates. For an index-tracking ETF, managing that gap in a stable way is an important objective.[5]
But tracking the index well and having an account that falls less are two different stories. A portfolio that declines along with the index can still be faithfully tracking it. Diversification itself is no guarantee against losses when the market falls.[6]
Conversely, moving differently from the index does not automatically make a portfolio good, either. Whether that difference is the result of intended risk diversification or the result of a flawed analysis has to be judged separately.
So even if you accept the concern Pzena raises, it does not automatically justify every choice that moves away from the index.
That Said, This Is Not an Argument for Abandoning Indexes
At this point, it is worth hearing the other side's logic. Pointing out concentration in an index and proving the superiority of individual stocks or active funds are separate matters.
Vanguard emphasizes that market-cap weighting reflects the composition of the market and can be run with relatively little trading. When a company's rising share price increases its weight, the fund does not necessarily have to make additional trades to keep up with that change.[7]
So the fact that concentration exists in an index should be distinguished from a judgment that the index's investment approach itself is wrong. Vanguard, too, pushes back against claims that market-cap weighting as a structure is, by itself, the reason large-cap stocks keep getting bigger.[8]
Switching to an equal-weight ETF is less a correct answer than a different choice. In exchange for lowering the influence of large-cap stocks, you raise the weight of relatively smaller companies within the same index and take on the trading burden that comes with periodic rebalancing. Vanguard describes this not as eliminating risk but as changing the character of risk.[7]
If your thinking about diversification ends with adding yet another ETF under a different name, you may find yourself back at the original problem. You should be able to explain which risks the product you switched to reduces and what it newly adds.
Conclusion
What I felt most strongly while writing this post is that it is not enough to think of the opposite of diversification as simply “investing in only a few stocks.”
At the product-selection stage, ETF names and investment targets catch your eye. But at the account-review stage, it is better to temporarily erase those names and lay out the companies you actually own and their combined weights. Then you need to ask where those companies' earnings come from, and what could go wrong that would put them in difficulty together.
As a result, you might choose to keep the index weighting as it is. If your exposure to a particular company or business environment is intentional, concentration alone is not necessarily a reason you must change it.
But if you thought you were diversified when in fact you were buying the same expectation split across multiple products, that is a different story.
Isn't diversification less about increasing the number of ETF names, and more about thinking through how to keep your entire account from shaking for the same reason when one of your judgments turns out to be wrong?
References
[1] SEC Investor.gov, Asset Allocation and Diversification https://www.investor.gov/introduction-investing/getting-started/asset-allocation
[2] Pzena Investment Management, Third Quarter 2026 Commentary: Tracking Error is Not Diversification: How Many Stocks Does Your Index Really Own? https://www.pzena.com/apac/investment-professionals/insights/tracking-error-is-not-diversification-how-many-stocks-does-your-index-really-own-3q-2026/
[3] S&P Dow Jones Indices, S&P 500 Index https://www.spglobal.com/spdji/en/indices/equity/sp-500/
[4] MSCI, IndexMetrics, June 2020, p.52 https://www.msci.com/documents/10199/402635a5-fd5d-498e-985a-1bec8ff8d8b1
[5] Vanguard Canada, Index tracking https://www.vanguard.ca/en/tools-and-resources/etf-fundamentals/management/index-tracking
[6] SEC Investor.gov, Diversify Your Investments https://www.investor.gov/introduction-investing/investing-basics/save-and-invest/diversify-your-investments
[7] Vanguard, What to consider when choosing between index-weighting approaches, 21 July 2026 https://www.vanguard.co.uk/professional/insights/what-to-consider-when-choosing-between-index-weighting-approaches
[8] Vanguard, Setting the record straight: The truths about index fund investing https://corporate.vanguard.com/content/dam/corp/research/pdf/settingtherecordstraightthetruthsaboutindexfund_investing.pdf
Disclaimer: This post is informational content that summarizes public disclosures, analyses, and the author's views; it is not a recommendation to buy or sell any particular stock.