Nomura warns: US stock indices are "severely distorted," ten stocks account for 70% of the S&P's gains, beware of a new shock from diesel.

Title context: Nomura warns: US stock indices are "severely distorted," ten stocks account for 70% of the S&P's gains, beware of a new shock from diesel. Text: Beneath the calm surface of US stocks, a rolling bear market is quietly spreading. Nomura cross-asset strategist Charlie McElligott warned in an October 8 report that the S&P 500 has barely moved over the past two months, yet 85% of its constituents have fallen into technical corrections. The index's "calm" is merely an illusion created by an extremely concentrated market structure. McElligott pointed out that just ten stocks contributed 70% of the S&P 500's 23% gain since March 30, with Nvidia alone accounting for 13%. Meanwhile, Europe's diesel shortage is transmitting through interest rate volatility to the entire market, constituting a "fuse" that could ignite at any moment. Within hours of Nomura's report, two key variables landed in succession: Trump announced he would not take military action against Iran before the midterm elections, and oil prices fell accordingly; the Financial Times then revealed that OpenAI's annualized revenue was far below the expectations previously communicated to investors, and the Nasdaq fell more than 1% on Thursday, with the AI sector taking a heavy hit. These two developments directly struck at the core logic McElligott describeda "severely distorted" structure in which a handful of AI winners prop up the index while the remaining constituents fall sharply beneath the surface. Nomura's analysis shows that current market correlation is at an extremely low level seen only twice in the past 25 years, and those two historical episodes were the eve of the 2007 global financial crisis and the 2018 "Volmageddon." Index "Going Nowhere," a Bear Market Beneath the Surface Since the July "momentum deleveraging/AI unwind" shock, the S&P 500 has moved less than 1% overall, with volatility indicators compressed across the board: SPX 10-day realized volatility is just 7, 3-month at-the-money implied volatility is at the 2nd percentile of the past year, and the 25-delta put/call skew is as low as the 0.4th percentile. However, this calm stands in stark contrast to the violent divergence at the individual stock level. Over the past month, the S&P 500 moved only 0.8%, while the average constituent swung as much as 8.9%a difference of 8.1 percentage points, at the 95th percentile of the past 30 years. AMD rose 39.5% in a single month, and Intel gained 27.6%; while Netflix fell 18.2%, and bank stocks also came under pressure. From a broader perspective, the damage to constituents is striking: 85% of S&P 500 constituents have drawn down more than 10% from their historical highs 59% have drawn down more than 20% 41% have drawn down more than 30% 26% have drawn down more than 40% 17% have drawn down more than 50% In other words, the index is just a step away from its historical high, but one-sixth of its constituents have been cut in half. McElligott characterized this as a "rolling severe correction" and criticized the view that "the stock market has not reflected interest rate risk"rate-sensitive sectors have been battered over the past month and a half: healthcare -3.5%, consumer discretionary -3.6%, utilities -4.6%, industrials -5.8%, financials -7.7%, materials -7.9%, REITs -9.8%. Over the past 30 days, only two sectors in the S&P 500technology, media, and telecom (TMT) and energyhave posted positive returns. Ten Stocks Drive 70% of Gains, Concentration Risk Extreme Nomura's data reveals the core mechanism of this "illusion." Since March 30, the S&P 500 has risen 23% cumulatively, with ten stocks contributing 70% of the gain: Nvidia alone contributed 13%, Micron, Apple, and Microsoft each contributed about 9%, and the remaining 490 stocks together contributed only 30%. Since August 3, the "Mag 7" (the seven tech giants) have risen 7.7%, the S&P 500 has risen 2.6%, while the equal-weighted S&P 500 has fallen 3%. McElligott defines this phenomenon as "Thematic Bifurcation": A handful of AI winners bear all of the index's upward momentum, while the remaining stocks continue to lag due to capital and sentiment being "squeezed out," expectations of AI disruption risk, and the drag from rising interest rates. Current market correlation is so low that there are only two comparable precedents in history. The S&P 500 one-month implied correlation is at the 1.8th percentile of the past year, and one-month realized correlation is at the 0.7th percentile. According to institutional reports, Goldman Sachs derivatives strategist Brian Garrett previously noted that in the past 25 years, only two periods have seen such low realized correlationthe eve of the global financial crisis and Volmageddon. Options Market "Only Fears Upside," Gamma Suppresses Volatility The calm at the index level is not accidental but is being actively maintained by structural forces. McElligott pointed out that the scale of index "volatility supply" is enormous, especially the selling of 0DTE/1DTE intraday options, mainly from QIS volatility risk premium (VRP) strategies, which "stuff" dealers' Gamma positions, thereby insulating the market from the possibility of large swings. Nomura's estimated 0DTE customer Gamma range reaches +$14 billion, at the 100th percentile. The returns from selling volatility are astonishing: the 10-day Sharpe ratio for selling daily 25-delta S&P put options is as high as 13.9, and the Sharpe ratio for selling daily strangles reaches 14.2. Beyond QIS strategies, the scale of derivative income funds cannot be ignored either. Derivative/income fund assets under management currently stand at $322 billion, up $222 billion since 2020, with funds and structured products together providing 415 million Vega of supply per month. Low realized volatility creates a self-reinforcing loop: volatility control strategies, CTAs, and risk parity funds' US equity exposure has rebounded to the 86th percentile, near year-to-date highs. Most anomalous is the options skew structure: put skew is at the 0.4th percentile of the past year, while call skew is as high as the 99.6th percentile (shown as the 100th percentile in Nomura's chart). Market participants are barely buying downside protection, "the only fear is right-tail upside risk"a large number of institutional investors, facing the pressure of missing out due to insufficient holdings in core AI stocks, are building up potential momentum forced to chase the rally. Diesel Shortage: The Hidden Fuse for Rate Volatility McElligott identifies energyespecially Europe's diesel shortageas the most fragile link in the entire market structure. His core judgment is: diesel crack spreads (Cracks) = interest rate volatility. The logic chain is clear: Europe and Asia face energy shortages, and global refining capacity (excluding Asia and India) is severely insufficient. Trump previously threatened a "diesel ban," forcing Europe to release emergency reserves larger than expected (France alone released 10 million barrels of diesel), but this can only sustain supply for about a month and cannot resolve the fundamental refining capacity bottleneck. The EU diesel crack spread is currently at 85, and eurozone 1-year forward 1-year rate volatility is at 112.6. The two have moved almost in lockstep since Februaryrate traders are effectively already trading diesel. European energy markets are also at year-to-date highs: German 1-year forward baseload power at 138, Dutch natural gas near 79, UK natural gas at 196, and diesel prices at $1,485. McElligott warned that even if the Iran situation temporarily eases, the structural shortage of refining capacity remains, and "no amount of Truth Social posts can refine a barrel of diesel." Three Trading Paths and Core Risks Faced with the above complex situation, McElligott proposes three trading strategies: First, chase the right-tail upside. In the upcoming month's earnings season, buy call spreads on large-cap tech stocks, betting that AI earnings will once again boost the market and push the S&P 500 toward 8,000. With call skew currently at the 99th percentile, this strategy offers asymmetric odds for institutions still underweight. Second, short the energy premium amid peace expectations. If Trump signals Iran negotiations before the midterm elections, energy and rate volatility will fall sharply, at which point one can rotate into upside options on "oversold assets" such as gold and the Russell 2000 (IWM). Third, hold rate volatility regardless. McElligott recommends hedging by going long USD/EUR rate Vega and holding long-dated Swaption strangles, because even if earnings season performs well, terminal rate expectations could still rise, and the Fed's policy path needs to be repriced to a "genuinely restrictive" range. Nomura backtesting shows that USD 20-year 20-year Swaption strangles returned more than 1% per month on average in the highest inflation quintile environment, almost entirely from Vega gains. OpenAI Revenue Data Deals a Blow to the AI Narrative On the very day McElligott's report was published, two key events successively validated his analytical framework. Trump announced on Truth Social that he would not take military action against Iran before the midterm elections, and oil prices fell accordingly, confirming the "de-escalation" scenario McElligott described. However, more impactful was that according to the Financial Times, OpenAI's annualized revenue is approximately $20 billion, far below the roughly $70 billion expectation previously communicated to investorsa gap of about $50 billion. After the news was published, the Nasdaq fell more than 1% that day, and the AI sector suffered concentrated selling. Analysts believe this data directly strikes at the core premise supporting the entire market structurethe authenticity of AI demand and the sustainability of its growthand also poses a direct challenge to the logic of chip-financing trades involving Nvidia, Broadcom (AVGO), and others. McElligott's bull case is built on the expectation that the "AI capex trickle-down effect" will ultimately benefit the remaining 493 stocksthe Mag 7's contribution to S&P 500 earnings growth has already fallen from 15% to 7%, while the contribution of the remaining 493 stocks has accelerated from 7% to 13% and then to 18%. But the starting point of this logic chain is precisely the authenticity of AI demand, and this premise has just suffered its first major blow from OpenAI's own investor documents. Currently, market correlation is at an extremely low level seen only on the eve of the global financial crisis and during Volmageddon, put skew is at the 0.4th percentile, and almost everyone is shorting left-tail risk at the cost of a 14x Sharpe ratio. McElligott's Swaption strangle may be the most cost-effective insurance available in the market today. This article is reprinted from Wall Street CN, author: Dong Jing; GMTEight editor: Zheng Yuyang.