Skip to content
← Back to Active Thesis Feed
🟢 Confirmed Oct 2, 2026

The S&P Isn't Ignoring Rates. Four Stocks Are.

The 10-year rose 85bp in Q3 and the S&P still gained about 2%, so the story became that stocks don't care about rates. They do. Four megacaps added about 300 index points while the rest of the S&P subtracted about 150 — the typical stock priced the rate shock, and four companies that don't need to borrow covered it up.

Citadel Securities (via TheStreet Pro and zerohedge, Oct 2): Microsoft, Nvidia, Apple and Meta added about 300 points to the S&P 500 in Q3 2026 — more than 200% of the index's total gain. The remaining ~496 stocks together subtracted about 150 points. 311 S&P 500 stocks fell over the quarter. Same quarter: the 10-year Treasury yield rose 85bp, its largest quarterly increase since 2016 (WSJ / Dow Jones Market Data), and touched 5.34% on Oct 1 — the highest since 2002. Context from Citadel's own May Toolkit: the Mag 7 represented roughly 35% of the S&P 500 and the 10 largest constituents about 40%; only 22% of S&P names had outperformed the index over the prior 30 days, a 30-year low.
View source ↗ 2026-10-02
For a month the bond-equity divergence has been read as complacency — the 10-year at a 23-year high with the S&P near its record. The attribution says the market wasn't complacent. Most of it repriced exactly as a rate shock should: 311 stocks fell in a quarter when long yields jumped 85bp. The index held because four companies offset all of it and then some. Vein 10 is competition for capital, which Fed Chair Warsh named in September as a driver of yields. This is what it looks like in equities. Companies sitting on cash fund their own AI spending and are indifferent to what Treasuries yield. Companies that need outside capital — small caps, leveraged balance sheets, floating-rate borrowers — pay the higher rate, and their stocks show it. The index average hides the split because it is cap-weighted toward the self-funders. Layer 4 is the fragility. The index's stability now rests on four names whose AI spending is the buildout itself. The stocks holding the S&P up are the same ones exposed if that spending gets questioned. The concentration isn't diversified away — it is the bet.
  • Falsifier: breadth improves while long yields stay above 5% — the median stock stops falling and the divergence turns out to be true indifference to rates.
  • Equal-weight S&P against cap-weight through Q4: the cleanest running measure of whether this gap widens or closes.
  • Megacap Q3 earnings in late October: any capex guidance cut or margin pressure hits the only part of the index that has been working.
  • Oct 7-8: 10-year and 30-year reopenings. Oct 14: September CPI. Each further leg up in long yields should widen the split, not close it.
⚡ Recasts the bond-equity divergence as concentration rather than complacency — rates are already hitting most of the market, and four self-funded stocks are masking it.
🟢 The Breadth Data Confirms It Independently 🔄 Follow-up Oct 4, 2026
Three separate sources now show the same split. About 75% of S&P 500 stocks fell in September, fewer than half sit above their 200 day average, and the equal weight index is on track for a losing streak seen only twice before, in 2002 and 2022.