Fifteen Stocks Now Hold 47% of the S&P 500. The 2000 Bubble Topped Out at 27%

The top 15 stocks in the S&P 500 now hold 46.7% of the index's total value, a record, according to The Kobeissi Letter. The other 485 companies split the remaining 53.3%.

That's 19.5 percentage points above the 27.2% the same top-15 group held at the peak of the Dot-Com Bubble. The numbers tell the story.

The Road to 47%

March 2000 was the last time anyone talked this way about index concentration. The top 15 then included Microsoft, Cisco, GE, Intel, Walmart, Exxon and Citigroup. Cisco briefly held the title of most valuable company on earth, then gave up roughly 80% of its value and never got it back.

The share those 15 names held was 27.2%.

Then came the unwind. Concentration drifted for a decade. It rebuilt slowly through the 2010s as Apple and Microsoft turned into cash machines. Then it rebuilt fast. Nvidia's run alone pulled trillions into one ticker. Add Broadcom, Eli Lilly, Berkshire Hathaway, JPMorgan and the rest of the current top group and you get a number with no precedent in the modern index.

Spread 19.5 points across 25 years and it looks like a slow grind. It wasn't. Most of it landed in the last three.

What Concentration Actually Does

Here's what matters: an S&P 500 index fund isn't the diversified instrument most people think they own. Buying it today gets you 53 cents of exposure to 485 companies per dollar, and 47 cents of exposure to 15.

That's not a philosophical distinction. Passive flows are price-insensitive. Money enters at whatever weight the index assigns. So when a mega-cap misses earnings, the retirement account of someone who has never bought a single stock takes the hit.

Active managers feel it too. They can't beat a benchmark without owning its biggest weights, so they own them. The reality is, that's forced positioning, not conviction. A lot of portfolios are making the same bet while telling themselves they're spread out.

This isn't 2000, though. The businesses are real. Nvidia ships actual silicon at actual margins. Apple generates actual free cash flow. Cisco in 2000 traded at 130 times earnings on revenue that later collapsed. Today's leaders sit closer to 25 to 35 times.

But that's the wrong comparison. The concentration risk here's mechanical, not valuation-driven. Even if all 15 companies execute perfectly, 47% of the index now rides on a handful of boardrooms, a handful of AI capex budgets and a handful of supply chains. So what happens when one of the fifteen stumbles?

What to Watch

Three things. First, earnings breadth. Watch whether the other 485 start contributing a larger share of S&P profit growth. If they do, concentration falls on its own. If they don't, it keeps climbing.

Second, hyperscaler capex guidance. Every quarter, Microsoft, Amazon, Alphabet and Meta tell you what the top of the index is worth. A slowdown in that line item hits the largest weights first and hardest.

Third, the 50% line. If the top 15 break above half the index, expect a much louder conversation about index construction and concentration caps. We're 3.3 points away.

What the street is missing: the risk isn't that these are bad companies. It's that a record share of American retirement money has become a single trade with fifteen legs. Watch the next index reconstitution and the next round of capex guidance. That's where this gets decided.