September inflation caught the market off guard: annual CPI printed 3.7% against 3.6% expected, with core at 2.6%. According to CNBC, the 10-year Treasury yield spiked intraday before settling at 5.19%, while the dollar hit an 18-month high. The rotation out of growth accelerated into value and energy. Futures repriced the odds of a month-end rate hike from 22% to 43% in a single day.
Morningstar's October outlook puts the US market at a 10% discount to fair value, but the number misleads: the entire discount sits in seven giants. Strip out Nvidia, Alphabet, Broadcom, Microsoft, Meta, Amazon and Tesla and the rest of the index trades at fair value. Buying the index and picking cheap stocks stock by stock are no longer the same game; the gap has never been wider.
The prescription is a barbell: a modest overweight in growth and value, funded by an underweight in core blend stocks. The checklist is crisp: wide economic moat, pricing power, near-term cash generation and a rising dividend. The slogan is simple: own the AI leaders, avoid distant-dated profitless growth stories and borrowers loaded with floating-rate debt.
Earnings week is packed: SpaceX and Netflix report on October 20, Tesla, Meta and IBM on October 21. Tesla delivered 486,532 vehicles in the third quarter versus 461,974 expected, a 5% beat that still trails last year by 2%. Zacks flags automotive margins and pricing pressure as the real test. FactSet expects Tesla EPS of $0.45 on $27.9 billion of revenue and Meta EPS of $6.67.
Big banks were mixed: JPMorgan posted a rare miss at $5.57 versus $5.93 expected, while Goldman crushed it at $17.01 against $13.02 as equities trading revenue surged 72%. Citi delivered its best third-quarter revenue in a decade and Wells Fargo says the consumer is holding up. At AboutSchwab the records kept coming: $190.4 billion in core net new assets and $7.2 billion of revenue. The shared warning from bank chiefs is the sheer scale of credit flowing into AI infrastructure.
At ASML the machine keeps humming: third-quarter sales of 11.7 billion euros, a 57.5% gross margin and 5 billion euros of net income. The company guides 2026 to 44-45 billion euros and the fourth quarter to 14.2-15.2 billion. Per CNBC, some customers are already discussing orders for 2029 and 2030 delivery, and 2028 capacity may rise beyond the planned 30% increase.
The week's sharpest move came in telecom: SpaceX announced a nationwide 800 MHz portfolio of up to 14 megahertz of paired spectrum from Grain, subject to FCC approval. Elon Musk called it a very big deal. Per CNBC, AT&T fell 9.8%, Verizon 8.8% and T-Mobile 13.3% the next session. Starlink counts more than 11,000 satellites across 205 countries; the company targets a mobile launch in late 2027.
Morningstar analyst Hodel stays calm: fair value estimates for the big three carriers stand, but Uncertainty ratings move from Medium to High. These licenses cover only about 6% of US low-band spectrum; Verizon's depth exceeds triple that in most areas. A real network needs fresh licenses in every band plus thousands of sites. The base case is unchanged: Starlink does not threaten the trio's wireless business, though it could push spectrum prices higher.
The episode closes with four long-term buys still trading cheap; names aside, the lesson is identical: wide moat, light debt, rising dividend. Broadcom tops Morningstar's most-undervalued wide-moat large-cap list. Defensive pockets such as health care, short-duration earners and businesses that can pass inflation through to prices stand out. This is a stock-picker's market: weigh the stock, not the basket.
| Indicator | Reading |
|---|---|
| September CPI | 3.7% (vs 3.6% exp.) |
| 10-year yield | At 5.19% |
| Tesla deliveries | 486,532 units, 5% beat |
Key moments
AI commentary
"My read on this episode is that it belongs to stock-pickers, not index-huggers: the discount has pooled in seven giants while everything else sits at full price. I would hunt for bargains through moat and dividend discipline, and keep October 21 earnings and the FCC calendar on the same screen."
AI assessment
Keep the counter-view in your pocket: if rates march toward 6%, long-duration growth multiples break first and the cheap-seven thesis wobbles. A pause in AI spending would defer ASML's 2030 order dreams too. And should SpaceX actually start building towers, telecom dividends would feel the squeeze; Hodel's High uncertainty flag is there for a reason.
The practical takeaway is plain: watch Tesla and Meta results on the evening of October 21 and the FCC vote on October 29. When picking stocks, screen for fair-value discount, economic moat and indebtedness together and never fall in love with a single metric. This content is not investment advice; consult a professional before deciding.
This content is for information only and is not investment advice; consult a licensed professional before making investment decisions.
Sources
8 links; 3 of them also cited by 5 other stories. Stories sharing a link do not confirm each other; a source's origin is not inferred from how often it is cited.
- @youtube.com YouTube — Morningstar, Inc.
- @morningstar.com Morningstar: Q4 2026 Stock Market Outlook
- @cnbc.com CNBC: SpaceX spectrum license hits AT&T, Verizon, T-Mobile
Also cited by: Musk Receives US Top Science Honor as Starlink Spectrum Deal Rattles Carriers
- @cnbc.com CNBC: Treasury yields jump on hot CPI inflation report
- @cnbc.com CNBC: ASML order interest through 2030 on AI demand
Also cited by: A $3.5T Cloud Over AI Buildout: Notes for Micron Holders · Seven Fallen Stocks Ordered by True Worth
- @cnbc.com CNBC: Bank earnings takeaways JPMorgan Goldman Wells Fargo Citi
Also cited by: Regional Lenders Carry Market Tail Burden · Memory Stocks Crack, Banks Take the Stage: My October 13 Game Plan
- @pressroom.aboutschwab.com AboutSchwab: Record net new assets and earnings Q3 2026
- @zacks.com Zacks: Tesla Q3 earnings automotive margins preview
inflation · earnings · spacex · undervalued stocks