The pitch for an S&P 500 index fund has always been simple: 500 companies, every major sector, a broad bet on the American economy. Set it and forget it. For decades that was accurate.
It is no longer a good description of what you own. The top 10 stocks in the index now account for roughly 43% of its total market capitalization, a record high and nearly double the historical average of 18% to 23% that held from 1990 to 2015, according to RBC Wealth Management data. The Magnificent Seven alone, Apple, Microsoft, Alphabet, Amazon, Nvidia, Meta, and Tesla, represented about 33.8% of the entire index as of June 2026. The top 20 companies make up around 49%.
So roughly forty cents of every dollar you put into a standard cap-weighted S&P 500 fund flows into about ten stocks. Eight of the ten largest are in information technology or communication services. Nvidia by itself is around 7% to 8% of the index.
That is not a description of diversification. It is a description of a concentrated sector bet that happens to be packaged as the default safe choice in most retirement plans.
What actually changed: the top 10 used to be unrelated
Concentration alone is not new, and by itself it is not necessarily alarming. Indexes have had dominant names before. What is different this time is the correlation, and this is the part that matters most for risk.
In past eras, the largest holdings spanned unrelated industries: an oil company, a bank, a retailer, a drugmaker. If oil fell, the bank was unaffected. The concentration was real but the risks were genuinely independent, which is what diversification is actually for.
Today's leaders are closely linked by a common theme, AI. The chipmaker, the cloud providers, the platforms buying the chips, and the hardware suppliers are all exposed, in different ways, to the same underlying bet: that the AI investment cycle continues and that these specific companies capture the value from it. They are customers and suppliers of each other. If that thesis disappoints, they do not fall independently. They fall together.
That is the actual risk in a way "43%" doesn't quite convey. You are not holding ten separate large bets. You are holding something closer to one very large bet, expressed ten ways.
There is a second, structural wrinkle. More than $40 of every $100 invested flows into just 10 companies, creating a feedback loop where passive inflows disproportionately support the largest stocks, increasing their weights and reinforcing their leadership regardless of fundamentals. Every automatic 401(k) contribution buys more of what is already largest, which pushes weights higher, which sends more of the next contribution to the same names.
The part most warnings leave out: this trade has been expensive
Here is where the honest version diverges from the alarming one, and any piece that skips this is selling you something.
Diversifying away from mega-cap concentration has cost investors real money for years. Over the last year, the S&P 500 rose 18% versus 11% for its equal-weighted counterpart. The top 20 companies contributed 64% of the index's five-year return. Someone who "prudently" diversified away from the megacaps five years ago has underperformed, substantially, and has been wrong for long enough to have watched a lot of people get rich being wrong.
As BlackRock frames it, underexposure to mega-caps may leave returns on the table, and efforts to diversify away from them represent larger active calls than investors realize. Deliberately underweighting the biggest companies is not the neutral, cautious choice. It is an active bet that the market is wrong about them.
There is also a fundamental case that the concentration is partly earned rather than purely speculative. Guinness Global Investors published analysis in July 2026 noting the top 10's earnings growth share had declined from a 2024 peak toward 50%, suggesting the rest of the index has begun contributing again, which they argue is a healthy broadening even while headline concentration stays extreme.
Why your age changes the entire question
This is the piece that gets lost when the debate is framed as "is the S&P too concentrated," because the answer genuinely depends on when you need the money.
If you are 30, index concentration is largely a non-issue. You are buying continuously for decades, and a drawdown in the megacaps means your future contributions buy the same companies cheaper. Time is the mechanism that resolves concentration risk, and you have it in abundance.
If you are 60 to 70, the calculus inverts, and this is where the concentration genuinely matters. The reason is sequence-of-returns risk: a large loss early in retirement, when you are withdrawing rather than contributing, does permanent damage that later gains cannot repair, because you sold assets at depressed prices to fund living expenses. Someone drawing down a portfolio where 43% sits in correlated AI-exposed names is exposed to exactly that scenario in a way a 30-year-old simply is not.
The same 43% is a footnote for one person and a serious planning problem for another.
What people actually do about it
The options are well-established, and none of them require abandoning index investing.
Equal-weight index funds give each of the 500 components the same allocation regardless of size. They underperform when mega-caps dominate, but they don't crater when the top of the index rolls over, which is precisely the tradeoff being purchased. International exposure is the other common route, since the MSCI EAFE index is significantly less concentrated, spreading risk more evenly across companies, sectors, and geographies.
The practical first step costs nothing: look at what you actually hold. Many people own an S&P 500 fund, a target-date fund, and a "growth" fund and believe they hold three different things, when all three are heavy in the same ten companies. You cannot decide whether your concentration is appropriate until you know what it is.
The bottom line
The S&P 500 remains a reasonable core holding, and this is not an argument for market-timing or for fleeing the index. It is an argument for knowing what you own, because the label and the contents have drifted apart.
Whether that warrants any change depends on your age, your timeline, and how much of your total portfolio sits in the same ten names. Those are the questions worth answering, and they are worth answering deliberately rather than discovering in a downturn.
A retirement plan built on the assumption that an S&P 500 fund spreads your risk across the American economy is resting on a premise that stopped being fully true some time ago.