Labor market tightened further in October, unemployment fell to 4.1% from 4.2%, payrolls rose 159K, and wages climbed 5.2% year-over-year, giving the Fed cover to keep rates restrictive even as credit stress builds underneath.
▸
High-yield spreads widened to 3.48 percentage points while the 10-year Treasury yield slipped to 4.84%, a divergence that signals investors are pricing rising default risk in weaker companies even as growth expectations hold, the private credit crunch the Fed's dashboard doesn't see.
▸
Tomorrow's CPI print is the hinge, core CPI forecast at 0.2% month-over-month (down from 0.3%), and a hot number would lock in a December rate hike (now ~77% priced) while a soft one could crack the hawkish narrative.
The big story
The U.S. labor market just handed the Fed a reason to stay hawkish, and simultaneously exposed a credit crunch the Fed's models can't see.
Friday's October jobs report showed unemployment dropping to 4.1% from 4.2%, nonfarm payrolls rising 159,000, and average hourly earnings climbing 5.2% year-over-year. In normal times, that's a "Goldilocks" print: strong enough to calm recession fears, cool enough to not force an emergency hike. But these aren't normal times.
The same morning, high-yield corporate spreads, the extra yield junk-rated companies pay over Treasuries, widened to 3.48 percentage points, up 16 basis points in a single day and the sixth consecutive day of widening. Investment-grade spreads barely moved at 0.93 percentage points. That split is the tell: the market is pricing rising default risk specifically in weaker, more leveraged borrowers, not a broad growth scare.
Here's what the Fed's dashboard misses. The five largest hyperscalers, Microsoft, Google, Amazon, Meta, Oracle, spent $691 billion on capital projects in the trailing twelve months through September, up 93.5% year-over-year. That capex now consumes 95.5% of their operating cash flow.
Aggregate free cash flow collapsed from $201 billion to $33 billion, an 84% drop. Amazon alone burned $34.8 billion of free cash flow in Q3 and raised $48.1 billion in long-term debt in a single quarter, lifting total debt from $65.6 billion to $173.1 billion. Interest expense tripled to $1.69 billion.
That spending isn't flowing to merchant chipmakers like NVIDIA, AMD, or Marvell anymore, it's increasingly funding internal ASIC programs (Google's TPUs, Amazon's Trainium, Microsoft's Maia) and sovereign build-outs. The Philadelphia Semiconductor Index (SOXX) is down 29.7% in one month, a 1-in-6,000 event statistically. Intel and Marvell each shed roughly half their value in four weeks. The merchant semiconductor revenue base is being hollowed out even as aggregate AI capex accelerates.
The credit impulse is running through custom-silicon supply chains and non-bank funding channels that don't show up in bank lending data or broad money aggregates. High-yield spreads are pricing it. The VIX spiked from 16.6 to 28.2 in four days before today's 7% snap-back, a liquidity air pocket in the making.
The Fed's September hold cited "oil price spike and pending Powell exit"; today's labor print reinforces the hawkish bias. Commerzbank now flags a "modest chance of a December hike." But a Fed hike on top of 95.5% cash-flow absorption would be double-tightening for the most capital-intensive sector in history.
The surface story is "AI bubble bursting." The structural story is a capex supercycle that has outrun its own merchant supply chain and is now self-funding through custom silicon, leaving a credit vacuum in the old channel. The Fed is flying blind on this transmission. Watch semi-cap equipment orders (KLAC, LRCX, AMAT) and high-yield primary issuance for semi-adjacent names. If both freeze, the credit impulse turns negative before the Fed sees it in payrolls.
What's going on today
Markets open the week with a split screen: the labor market is flashing strength while credit markets flash stress. Friday's jobs report, 4.1% unemployment, 159K payrolls, 5.2% wage growth, gave the Fed exactly the data it needs to keep rates at 3.75-4.00% or even hike in December (now ~77% priced per CME FedWatch).
But beneath the headline, high-yield spreads have widened for six straight sessions to 3.48 percentage points, the VIX spiked 70% in four days before retracing a third of the move, and the semiconductor complex, the merchant arm of the AI build-out, is in a historic drawdown: SOXX down 29.7% in a month (z-score -3.6), Intel and Marvell each -50%, Micron -37%. The Nasdaq-100 closed Friday at 26,990, down 13.4% on the month; the S&P 500 at 7,377, down 5.5%.
Oil keeps grinding lower, Brent -7.2% on the week to $97.28, WTI -6.5% to $90.16, despite ongoing Iranian tanker attacks in the Strait of Hormuz and a rare Houthi strike on Saudi Aramco's Riyadh tower. Gold holds near $4,108, up 3.5% on the week, as sovereign buyers (China, India, Gulf central banks) continue accumulating.
Crypto is bleeding: Bitcoin -3.4% on the week to $74K, Ether -13% to $2,105, Solana -10% to $86, with Binance still seeing net outflows post-hack and spot Bitcoin ETFs recording $1.28 billion in outflows over the last five sessions. The dollar index eased to 101.8 from 102.3 a week ago. Tomorrow's CPI report (8:30 AM ET) is the week's hinge: core CPI forecast 0.2% month-over-month vs. 0.3% prior; a hot print locks in December tightening, a soft one could crack the hawkish narrative.
The big picture
Bond yields are drifting lower at the front end, the 2-year Treasury yield fell to 4.46% from 4.58% last Monday, the 5-year to 4.59% from 4.69%, while the 10-year slipped to 4.84% from 4.90%, keeping the 10s-2s spread at a modest 38 basis points and the 10s-3m spread at 67 basis points. The curve is upward-sloping but not steep, signaling modest growth expectations without recession panic.
Yet high-yield spreads tell a different story: at 3.48 percentage points, they've widened 16 basis points in a day and 80 basis points from the late-September low of 2.68%, a move concentrated in semi-adjacent issuers and the non-bank lenders that funded their inventory buildup. Investment-grade spreads are unchanged at 0.93 percentage points. That divergence, IG calm, HY stressed, is the fingerprint of a sector-specific credit crunch, not a broad macro slowdown.
Equity breadth has deteriorated sharply: only 58% of S&P 500 constituents trade above their 50-day moving average (down from 72% on November 4), and 52% above the 200-day (down from 69%). The advance-decline ratio is 1.33:1. The Nasdaq-100's 13.4% monthly drop is a statistical rarity (z-score -2.5); the Nikkei's 14.9% monthly plunge is even rarer (z-score -3.2). Yet defensive sectors are holding: consumer staples (XLP) up 10.3% on the month, healthcare (XLV) up 9.5%, utilities (XLU) up 4.8%. Financials (XLF) are up 6.3% on the month, led by insurance and payment networks, a quiet rotation into cash-flow durability.
The dollar index (DXY) sits at 101.8, down 0.4% on the day but flat on the week; the broad trade-weighted dollar is 121.7, near the 122 level that historically triggers emerging-market funding stress. USD/JPY holds at 158, sustained by the 375-basis-point Fed-BOJ rate differential (Fed 3.75-4.00%, BOJ ~0.25%) despite BOJ hawkish rhetoric signaling a December hike. EUR/USD at 1.12. Gold at $4,108 is up 3.5% on the week while the dollar eased, the sovereign-buyer bid (central banks adding ~1,000 tonnes over the last four quarters) is decoupling gold from its traditional inverse correlation with real yields.
Oil's grind lower continues despite geopolitical noise: Brent -7.2% on the week to $97.28, WTI -6.5% to $90.16. U.S. crude inventories remain low at 410 million barrels, gasoline at 209 million, distillate at 99 million, critically below the five-year November average, and refineries run at 94.5% utilization. The diesel crack spread has widened to roughly $32 per barrel from $22 in September, but spot diesel prices haven't fully passed through to CPI yet.
Every $10 sustained crack spread adds ~0.15 percentage points to core CPI via transportation and goods weights. With core CPI at 2.57% year-over-year and core PCE at 3.12%, the Fed's "last mile" problem is hardening. Natural gas bucked the trend, up 6.1% on the week to $3.03/MMBtu, suggesting tightening after earlier storage builds.
Nikkei plunges 15% in a month as global tech unwind overwhelms weak-yen tailwind. The Nikkei 225 fell to 58,776 on November 9, down 14.9% in one month (z-score -3.2) and 12.2% in three months (z -2.0), erasing most of its 2026 gains and sitting 19% below its June 5-year high near 72,000. The selloff tracks the semiconductor-led rout: SOXX dropped 29.7% in a month and 37% from its peak, dragging Japanese chip-gear makers (Tokyo Electron, Advantest) and tech-heavy indices globally (Nasdaq-100 -13.4% 1m).
USD/JPY at 158, a level that historically lifts exporter earnings, failed to cushion the decline, signaling that fundamental deterioration in global tech demand and risk appetite is dominating currency effects. Positioning extremes (rare z-scores across INTC -50%, MRVL -50%, CAT -31% in one month) suggest leveraged long unwinds are amplifying the move. For the decline to extend, semiconductor order data must keep weakening and the Fed must hold rates at 3.75-4% (November 8) with long yields near 4.84% (10y, November 5), keeping discount rates high. A reversal requires visible capex stabilization from hyperscalers (currently spending 95.5% of operating cash flow) and a VIX retreat from 23.7 (November 9).
Around the world
Iran's Revolutionary Guard Corps declared on November 9 that "oil smuggling" through the southern route of the Strait of Hormuz has been completely halted following the destruction of "enemy navigation systems", a direct challenge to the U.S.-promoted shipping lane that normally carries ~20% of global oil consumption. Nine vessels have been confirmed attacked in the Strait and Gulf since November 1, including a tanker on fire off Oman on November 7 and another struck by a projectile in Hormuz an hour later.
U.S. Central Command says it has redirected 147 vessels and disabled five merchant ships since reimposing a blockade in July. Yet Brent has fallen 16% from the October 30 high of $117.64 to $97.28, the steepest weekly drop since the pandemic. The market is pricing de-escalation optimism, Trump downplayed further strikes on November 4, calling them "highly unlikely" if nuclear talks progress, even as physical attacks persist. Saudi Arabia's East-West pipeline (Petroline) now moves up to 7 million barrels per day, bypassing Hormuz entirely, but the risk premium hasn't fully collapsed; shipping insurance rates remain elevated.
In Japan, the Nikkei bounced 2.3% overnight to 58,776 but remains 14.9% below its October level (z-score -3.2). The BOJ is signaling a December rate hike after inflation reached 2.1%, but the 375-basis-point Fed-BOJ differential continues to anchor carry trades. Mitsui Chemicals agreed to acquire semiconductor materials maker JSR jointly with the Development Bank of Japan for $8.8 billion, a sovereign-backed play for critical supply-chain control.
China extended its trade truce with the U.S. to January 10, 2027, suspending the additional 24% tariff on U.S. imports (a 10% duty remains) and pausing tit-for-tat port fees. The PBOC set the yuan reference at 6.7188, nudging the currency stronger.
Chinese banks are boosting dollar deposit rates to smooth the yuan rally. Taiwan activated market-stabilization measures alongside South Korea to stem AI-related equity bleeding. The U.S. approved Nvidia H200 chip sales to approved Chinese buyers (Baidu, BYD, Tesla Shanghai) with a 25% tariff, a functional engagement channel under the truce.
Europe's Sentix investor confidence weakened further in November as global growth momentum fades. The ECB holds its deposit rate at 2.5% and main refinancing rate at 2.65%, unchanged through October.
Eurozone bond yields fell while U.S. yields rose in October, a fragmentation dynamic where stress concentrates in vulnerable sovereigns (France) rather than the currency. Fitch upgraded Greece and Cyprus, confirming periphery resilience. The ECB warned extreme water scarcity puts up to 24% of eurozone output at risk, a structural climate constraint on growth.
Russia-Ukraine: trilateral peace talks resume in Abu Dhabi November 10-11. Russia intensified strikes on Ukraine's energy grid, November 8 attacks cut power to 600,000 in Dnipropetrovsk and 250,000 in Kharkiv. Russia gained a net 20 square miles in the past four weeks after two months of losses. The EU's 21st sanctions package (adopted July 23, implemented by Switzerland November 4) expanded transaction prohibitions to 33 Russian banks and 41 shadow-fleet vessels. Europe imported 6.9 billion cubic meters of Russian gas in Q1 2026, up 16% year-over-year, but Russia now supplies only 11% of European gas vs. 45% pre-invasion.
Companies making news
Nvidia board investigates CEO espionage allegations. Jensen Huang apologized for an affair with an executive suspected of working for Chinese intelligence; the board has opened an investigation. Nvidia shares fell 8.8% on the week despite a 2.8% Friday bounce; the stock is 21% below its one-month high. The scandal introduces leadership uncertainty at the single most critical node in the AI hardware supply chain.
Marvell to acquire Lumentum for $115 billion in all-stock deal. The merger creates a data-center connectivity giant combining Marvell's custom ASIC dominance (>70% market share in design services) with Lumentum's laser-based optical interconnects. Marvell shares are down 50% in a month; the deal signals consolidation as the merchant semi model fractures under hyperscaler internalization.
Amazon Q3 capex $77.3 billion, free cash flow -$34.8 billion, debt raised $48.1 billion. Total debt surged to $173.1 billion from $65.6 billion at year-end 2025; interest expense tripled to $1.69 billion. Management guides FY2026 capex ~$220 billion and 2027 "meaningfully more." The quarter confirms the hyperscaler funding model has shifted from cash-flow self-funding to debt issuance, 62% of Q3 capex was debt-financed.
TSMC Q3 revenue $46.7 billion (+51% YoY), gross margin 66.4% (down from 67.7% on 2nm ramp costs). 2nm process = 8% of wafer revenue; 2026 capex $60-64 billion; 2027 capex estimates $75-85 billion (Morgan Stanley $83B, BofA $85B). 2nm capacity more than doubling by 2027; wafer prices rising 3-10% from January 2027. The margin dilution is the cost of securing leading-edge supply for AI; the price hikes pass through to hyperscalers.
McDonald's Q3 EPS $3.57 vs. $3.36 estimate (+6.2%), revenue $7.16B (+1% YoY) missed $7.29B estimate. Global comparable sales +1.5%; U.S. comparable sales -0.2% on lower traffic, though September turned positive. The slight revenue miss and traffic decline suggest consumers remain price-sensitive even at the value end, a signal for discretionary spending broadly.
Novo Nordisk Q3 sales DKK 79.2B (+4% YoY), operating profit +27% to DKK 32B. Obesity treatment sales +22% in Q3; 1.3 million new patients added, ~48 million total reached. Full-year guidance raised to -2% to +2% at constant exchange rates. The GLP-1 franchise continues expanding the addressable market; healthcare spending is shifting toward chronic-condition management.
Arm Holdings Q3 EPS $0.45 vs. $0.41 estimate (+11%), revenue $1.44B (+27% YoY) beat $1.38B. Guides Q4 ~$1.54B vs. $1.53B estimate. Shares chopped post-earnings, the beat was priced in, and the high valuation (112x trailing P/E) leaves no margin for anything less than massive outperformance.
Qualcomm Q3 EPS $2.21 vs. $2.23 estimate (-0.9%), revenue $9.95B beat $9.69B. Total revenue -4% YoY but automotive +61% to $1.59B. Targets $40B non-handset revenue by FY2029. Diversification away from saturated smartphones is working; the auto/AI pivot is real.
Uber Q3 revenue $14.77B (+10% YoY), GAAP net income $512M vs. $6.63B year-ago. The plunge reflects a $476M pre-tax loss from equity investment revaluations vs. a $1.5B benefit a year ago. Core mobility/delivery growth is solid; the noise is mark-to-market volatility.
Today's earnings watch: KKR, CEG, B, FERG, IX pre-market; OXY, SKYD, TTWO after-hours. Cisco reports Thursday November 12; Baidu Monday November 16.
From Washington
The Fed enters the week with the December hike probability at ~77% (CME FedWatch), up from 67% a week ago, driven by Friday's strong jobs report. Chair Kevin Warsh has moved away from explicit forward guidance, aiming for seven FOMC meetings in 2027 and six in 2028, and stated at the October 28 meeting that "interest rates get in the cracks" and affect "almost everything," including the AI investment surge, challenging the notion that any sector is rate-proof. San Francisco Fed President Mary Daly said November 7 the Fed "will likely have to tighten policy further" with a deliberate, gradual approach.
St. Louis Fed President Alberto Musalem stated AI investment "will continue to contribute to inflation." The September Summary of Economic Projections showed a median 4.1% year-end 2027 rate (implying one December 2026 hike then pause), but markets price above the dots at every horizon: implied midpoint 4.12% after January 2027, 4.33% by December 2027. Mortgage rates at 7.18% (30-year fixed, week ending November 5) continue to freeze housing, new-home sales run 5.4% below the 12-month average, starts-to-permits ratio compressed to 87.6%.
Treasury auction calendar: 3-year note auction today; 10-year and 30-year later this week. The October 29 7-year auction cleared at 4.894% with a 2.46 bid-to-cover; the October 7 10-year reopening yielded 5.300%. The Chicago Fed National Financial Conditions Index (NFCI) rose to -0.459 for the week ending October 30, the sixth consecutive weekly increase and highest since mid-September, a cumulative 0.059-point tightening (11% toward neutral) driven by credit and leverage sub-indices. Hedge fund Arini Capital sold $800M in debt after margin calls on October 31, a visible leveraged-player deleveraging episode.
Under the hood
The AI capex supercycle is bifurcating: hyperscalers are accelerating spend (95.5% of operating cash flow) while the merchant semiconductor complex collapses (-25% to -50%), creating a private credit crunch that HY spreads (3.48pp) are pricing but the Fed cannot see because strong labor (4.1% unemployment, 5.2% wage growth) gives it cover to stay hawkish.
This is a scissors crisis: the Fed's policy rate (3.75-4.00%) responds to aggregate labor/inflation data, while the AI-driven credit impulse runs through custom-silicon supply chains and non-bank funding channels that don't show in traditional monetary aggregates.
Hyperscaler capex hit $691B TTM (+93.5% YoY), 35.8% of revenue, 95.5% of operating cash flow, yet FCF crashed to $33B from $201B (-83.7%). That capex is increasingly flowing to internal ASIC programs (TPU, Trainium, Maia) and sovereign-build outs, not merchant GPUs. NVDA, AVGO, MRVL, MU, INTC have lost 30-50% in a month (SOXX -29.7% 1m, z -3.6). The merchant semi revenue base is being hollowed out even as aggregate capex accelerates.
HY OAS widened to 3.48pp (+16bp today, day 6 of the move) while IG OAS sits at ~0.9%. The widening is concentrated in semi-adjacent issuers and the non-bank lenders that funded their inventory buildup. Plumbing shows VIX spiking from 16.6 to 28.2 in four days before today's -7.2% snap-back, a liquidity air pocket in the making.
The Fed's September hold cited "oil price spike and pending Powell exit"; today's labor print (unemployment 4.1%, earnings $38, +5.2% YoY) reinforces the hawkish bias. Commerzbank flags a "modest chance of December hike." But the transmission the Fed watches, bank lending, broad money, misses the AI capex channel entirely. The private tightening is already 95.5% of hyperscaler cash flow; a Fed hike on top would be double-tightening for the most capital-intensive sector in history.
Implication: the next leg down in semis isn't valuation, it's funding. Watch for semi-cap equipment orders (KLAC, LRCX, AMAT) and HY primary issuance for semi-adjacent names. If both freeze, the credit impulse turns negative before the Fed sees it in payrolls.
The surface story is "AI bubble bursting." The structural story is a capex supercycle that has outrun its own merchant supply chain and is now self-funding through custom silicon, leaving a credit vacuum in the old channel. The Fed is flying blind on this transmission.
Second-order: sovereign wealth funds and central banks (gold $4,108, broad dollar 121.7) are the marginal buyers of the custom-silicon build-out, not public equity markets. This decouples the AI capex cycle from Fed-rate sensitivity, until the HY credit crunch spills into bank C&I loan books.
Watch: KLAC/LRCX/AMAT next quarterly orders (confirm/refute semi-cap freeze); HY primary issuance for semi-adjacent names (refute if it reopens cleanly); Fed senior loan officer survey Q4 (confirm if C&I standards tighten on "tech hardware").
The long view: Is hyperscaler AI capex sustainable?
No, not at current intensity without a funding regime change. The five largest U.S. hyperscalers generated $543 billion in operating cash flow in the first nine months of 2026 but spent $557 billion on capex, a $14 billion free cash flow deficit. Full-year 2026 group capex tracks ~$770-800 billion, up from $185 billion in 2023, a 300% surge in three years. Trailing-twelve-month capex hit $691 billion (+93.5% YoY) = 95.5% of operating cash flow; FCF collapsed to $33 billion from $201 billion (-83.7%).
Amazon Q3 alone spent $77.3 billion on capex (120% YoY) against $42.4 billion OCF, producing -$34.8 billion quarterly FCF, and raised $48.1 billion in long-term debt in the quarter, lifting total debt from $65.6 billion (end-2025) to $173.1 billion (September 2026); interest expense jumped to $1.69 billion from $0.54 billion. Amazon guides FY2026 capex ~$220 billion and 2027 "meaningfully more." Incremental debt as a share of capex surged from 9% in FY2024 to ~32% by mid-2026; Amazon Q3 alone was 62% debt-funded.
NVIDIA captures 75-81% of AI accelerator revenue in H1 2026. Custom ASICs are rising: Broadcom holds >70% of custom AI accelerator design services; Google spends ~$8 billion/year on TPUs.
Memory commands ~30% of 2026 hyperscaler data-center capex; SK Hynix supplies ~62% of HBM, >50% of 2027-28 capacity booked, price increases expected in 2027. NVIDIA supply commitments ballooned to $279 billion (July 2026) from $119 billion (April 2026), primarily memory; Morgan Stanley estimates NVIDIA secured ~37.3% of global HBM supply for 2027; Samsung and SK Hynix plan 8-layer HBM4 mass production for NVIDIA starting Q2 2027. TSMC Q3 revenue $46.7 billion (+51% YoY), gross margin 66.4% (down from 67.7% on 2nm ramp); 2nm = 8% of wafer revenue; 2026 capex $60-64 billion; 2027 capex estimates $75-85 billion; 2nm capacity more than doubling by 2027; wafer prices +3-10% from January 2027. Hyperscalers are becoming active power-infrastructure developers to bypass grid constraints.
For markets: valuation multiples assume capex peaks soon and FCF inflects positive. Amazon's $77.3 billion quarterly beat and "meaningfully more" 2027 guide, Alphabet's declining OCF and $142 billion debt, NVIDIA's gross margin guided to 74% (bottoming 71-72% in Q4 FY2027), and TSMC's 2027 capex $75-85 billion crack that narrative. SOXX -29.7% 1m, VIX 25.64, 10y yield 4.84% show rotation from semis and rising equity risk premium.
If capex stays at ~$170 billion/quarter run-rate, or accelerates per Amazon/TSMC/NVIDIA guides, FCF yield compresses further and duration risk rises. $820 billion in off-balance-sheet lease commitments mask capex intensity. For a normal person's money: the AI build-out pulls forward years of semiconductor, memory, and power-equipment earnings (benefiting NVDA, MU, AVGO, TSMC, ASML, VRT, GEV) but the largest S&P 500 components are self-funding less and issuing more debt, increasing systemic leverage.
What would change the view: Nvidia Q3 FY2027 (November 17) revenue below $105.8 billion or gross margin below 71%; 10-year yield sustaining above 5.5%; HBM4 contract pricing for CY2027 below ~$1,000/stack; TSMC Q4 2026 (January 2027) guiding 2027 capex below $70 billion or flagging 2nm yield issues below 60%; S&P Global revising FCF recovery timeline earlier than 2029; Amazon Q4 2026 (February 2027) showing AWS growth deceleration below 30% YoY or capex guide cut.
Worth learning today: Futures, promises with leverage
Yesterday's lesson asked: if Core CPI comes in at 0.2% (forecast) vs. 0.3% prior, which way does the 2-year yield move? The answer: lower. A cooler core print reduces the probability of a December hike, and the 2-year yield is the most direct proxy for near-term Fed policy expectations. The mechanism: lower expected policy rate → lower discount rate for short-dated cash flows → 2-year Treasury price up, yield down. (Tomorrow's actual print will confirm or refute, we'll revisit.)
Now, futures. You see "S&P futures up 0.4%" in the pre-market. What is that? A futures contract is a binding agreement to buy or sell an asset at a set price on a set date.
The ES (E-mini S&P 500) contract represents $50 × the index level, so at 7,433, one contract controls ~$371,000 of exposure. You post initial margin (~$17,000, about 4.5%) and maintain maintenance margin (~$13,500). That's 20-25x leverage. If the index moves 1%, your equity moves 20-25%.
Two players use futures: hedgers and speculators. A pension fund with $10 billion in S&P stocks sells ES contracts to lock in today's level, they're insuring against a drop. A hedge fund buys ES to bet on a rise without putting up the full notional. The price of the future converges to spot at expiration; before then, it trades at a small premium or discount reflecting the cost of carry (interest rate minus dividend yield).
Why it matters to you: your index fund doesn't trade futures directly, but the futures market sets the opening price for the cash session. Overnight, ES trades 23 hours a day. When you see "S&P futures up 0.4% at 6 AM ET," that's the market pricing in overnight news (Asian session, earnings, geopolitics) before the NYSE opens.
The cash index opens near that level. Arbitrageurs keep them aligned: if futures get too rich, they sell futures and buy the 500 stocks; if too cheap, the reverse. So the futures move is the pre-market signal.
Quick check: why does a hot CPI print push mortgage rates up? Hot CPI → higher expected Fed funds rate → higher 10-year Treasury yield (the benchmark for 30-year fixed mortgages) → higher mortgage rates. The 10-year is the bridge.
Tomorrow's prediction prompt: USD CPI m/m (scheduled 2026-11-10; forecast 0.3%, prior 0.6%). If this print comes in above forecast, which way does the 2-year yield move, and why? We'll answer tomorrow.
What to watch this week
▸
Tuesday, November 10, 8:30 AM ET: USD Core CPI m/m (forecast 0.2%, prior 0.3%) and CPI m/m (forecast 0.3%, prior 0.6%). — The week's hinge. A hot core print (>0.3%) locks in a December Fed hike; a soft print (<0.2%) could crack the hawkish narrative and relieve pressure on rate-sensitive growth stocks.
▸
Thursday, November 12, 8:30 AM ET: USD Core PPI m/m (forecast 0.3%, prior 0.7%) and PPI m/m (forecast 0.4%, prior 1.3%). — Pipeline inflation, what producers pay, feeds forward into CPI. A sticky PPI keeps the "last mile" problem alive.
▸
Thursday, November 12, GBP GDP m/m (forecast 0.0%, prior 0.1%). — UK growth stalling would reinforce BOE caution after its first hike in three years.
▸
Monday, November 17, after market close: NVIDIA Q3 FY2027 earnings. — Revenue guide $108B ±2% ($105.8-110.2B), gross margin 74% ±50bps. The single most important print for the AI capex thesis.
▸
Ongoing: — HY primary issuance for semi-adjacent names; KLAC/LRCX/AMAT order commentary; Fed senior loan officer survey Q4 (C&I standards on "tech hardware").
Not financial advice. This brief is for informational and educational purposes only and does not constitute investment advice or a recommendation to buy or sell any security.
Data: macro indicators per FRED® (Federal Reserve Bank of St. Louis); not endorsed or certified by the Federal Reserve Bank of St. Louis. Energy data per the U.S. Energy Information Administration (EIA). Auction data per U.S. Treasury Fiscal Data. Filings per SEC EDGAR. Market prices per Yahoo Finance. Earnings calendar per Financial Modeling Prep.