Nomura McElligott: S&P Calm Masks Internal Bear Market, Right-Tail Risk Dominates Options Pricing
nashnova research
The S&P 500 has barely moved in two months, yet 85% of its constituents are already in a technical correction — options markets fear missing the rally, not a crash, and history says this kind of split ends with a volatility explosion.
The index hasn't moved — so why are 85% of stocks falling?
Over the past two months the S&P 500 moved just 0.8%, while its average constituent swung 8.9%. The 8.1-percentage-point gap sits at the 95th percentile of the past 30 years.
This means → the index looks like a still lake, but the fish underneath are thrashing — a handful of big winners and a mass of losers cancel each other out, pinning the index in place.
Case in point: AMD +39.5%, Intel +27.6%, while Netflix −18.2% and banks also fell — bulls and bears living inside the same index.
What are options markets afraid of? Not a crash — missing the rally
S&P 500 three-month at-the-money implied volatility — the market's forecast of future swings — sits at the 2nd percentile over the past year. Extremely low.
The 25-delta put/call skew — measuring whether traders fear a drop or a surge more — is at the 0.4th percentile; put skew is similarly depressed.
In plain terms = almost nobody is buying crash insurance, and almost everybody is chasing upside lottery tickets. The market's only fear is the right tail — missing the rally, not bracing for a fall.
Correlation has hit a historic low — why is that dangerous?
S&P 500 one-month realized correlation — how closely constituents actually move together — is at the 0.7th percentile over the past year. Stocks are marching to their own drums.
Goldman Sachs derivatives strategist Brian Garrett warns: only two periods in the past 25 years saw correlation this low — February 2007 (the eve of the subprime crisis) and January 2018 (the eve of "Volmageddon").
This means → history shows this level is unsustainable. It typically ends with a "starburst" spike in index volatility — the longer the calm, the sharper the snap.
Ten stocks drove 70% of the rally — who is crowding out whom?
Since March 30 the S&P 500 has risen 23%. Ten stocks account for 70% of that gain: Nvidia alone contributed 13%; Micron, Apple, and Microsoft each roughly 9%.
Since August 3 the "Magnificent Seven" are up 7.7%, the S&P 500 up 2.6%, and the equal-weight S&P 500 down 3%.
This reflects the "AI winners" dominating the index through a capital-and-sentiment crowding-out effect — money and attention flood a few names, while the other 490 stocks contributed just 30% combined.
What is McElligott's best trade of the year?
All year he has recommended "going long the core AI constraint" — buying the beneficiaries of a compute shortage (semiconductors) plus energy infrastructure, forming a semiconductor + energy barbell strategy.
Year-to-date the strategy has returned 65% with a Sharpe ratio — excess return per unit of risk — of 3.2. A traditional 60/40 stock-bond portfolio returned just 8% with a Sharpe of 1.2.
In plain terms = for the same unit of risk, the barbell earned nearly three times as much as the classic portfolio.
Why hedge an AI long with oil, not bonds?
McElligott stresses: with inflation still sticky and rate volatility frequent, bonds can neither serve as a long asset nor effectively hedge equity risk.
Europe's diesel shortage is creating energy-supply pressure that leaves the market just "one headline away" from a rate-volatility shock.
This means → the hedge for an AI long is not bonds but oil — energy is both the physical constraint on AI compute and the natural hedge when inflation surprises to the upside.
市场有风险，内容仅供研究参考，不构成投资建议。