11AM Hour: Apple Drops After Earnings, Election Day Expectations & Investing in the Private Markets 11/3/26

Apple is under pressure after earnings because, although it spent only $9 billion on capital expenditures—down 30% year over year and less than 2% of revenue—its exposure to the AI build-out is increasing. The company is bidding up memory chips for iPhones and Macs and paying cloud costs to run Siri AI; TD Cowen estimates every 5% increase in memory prices cuts gross margin by half a point. Apple committed $33 billion to suppliers from June through September, much of it likely memory. Prices have been raised to offset costs, but critics call that a temporary fix, while Apple says Siri monetization and device upgrades can eventually help.

Apple CFO Kevin Parekh said the company will significantly increase AI investment mainly through research and development, which was already up 32%, rather than through acquisitions or acqui-hires. CEO John Ternus told CNBC that Apple is working on monetizing AI, expects Siri to encourage device upgrades, and plans to impose usage caps. Apple also defended its Gemini partnership with Google despite questions about whether Siri’s experience—particularly its agentic capabilities—lags competitors. Analysts cited by Mackenzie Sagalos see earnings pressure continuing unless memory prices cool or Siri becomes a meaningful revenue source; UBS expects Apple’s forward multiple to fall from 36 to 31 times earnings.

Palantir reported a strong quarter, with CEO Alex Karp saying enterprise customers are increasingly adopting open-weight models. Commercial revenue grew triple digits and, for the first time, exceeded government revenue. Karp argued that frontier-model companies face “uncapped liabilities” and ultimately may have to be nationalized. The market focused on whether growth can continue: Palantir had 64 deals worth at least $10 million, down from 73 the prior quarter, while its backlog indicated slower growth and executives blamed shorter contracts. They said customers are piloting the technology before moving to longer commitments. CFO David Glazer said the company is focused on execution rather than a stock split or buyback.

Gabelli portfolio manager John Belton said midterm uncertainty, higher interest rates, higher oil prices, renewed Middle East tensions, and slowing Anthropic revenue have made technology a difficult trade, even though cloud-company fundamentals remain strong. He expects investors eventually to defend high-quality names with lower valuations once the macro pressure normalizes, but would not predict when hedges will come off or leverage return. He distinguishes AI-supply-chain companies—semiconductors, data-center industrials, and cloud services—from application companies such as Palantir. The former need Anthropic or similar firms to prove durable revenue growth and justify the capital cycle; application companies may work even if large language-model economics weaken.

Belton favors Microsoft, Amazon, and Google because their cloud businesses are large, diversified, and growing at scale rather than serving only AI companies. He highlighted Amazon management’s expectation that AWS will grow 40% in each of the next two years and said AWS’s roughly five million customers provide visibility Amazon has not previously had. Caterpillar, meanwhile, combines a cyclical construction and mining recovery with a secular AI-data-center power story. Its power-and-energy order book is booked through the end of the decade, but the cyclical rebound is taking longer than expected, explaining the stock’s decline from an 86% year-to-date gain at the end of June to a single-digit gain.

Megan Casella said Democrats could regain the House by flipping a net three seats, while taking the Senate requires a net four and at least two wins in states President Trump carried by double digits. Michigan, Maine, Alaska, Texas, and Ohio were identified as pivotal, with slow counting and ranked-choice voting in Maine and Alaska potentially delaying the result. Morgan Stanley’s Ariana Salvatore said investors care more about executive policy—tariffs, U.S.-Iran relations, deregulation, and geopolitics—than Congress for the market’s main drivers. A Republican hold would generally support AI sentiment; a Democratic sweep would be negative, although data-center investment could move to other states.

Salvatore said a Democratic sweep could prompt companies to announce mergers and acquisitions while they still expect a more difficult political regime after 2028, although rates, economic conditions, and CEO confidence also matter. She cautioned against treating midterm outcomes as a clean signal for the next presidential election because the correlation is poor and voter issues change. Affordability remains central, but Congress has limited ability to address it; tariff changes are the president’s most direct lever on consumer prices. Defense legislation, as well as potential changes to SNAP and Medicaid cuts enacted through the OBBA, are the sectors and policies she sees as most immediately relevant.

New York Angels board member Bob Peck described Anthropic’s IPO as a major test for technology-market sentiment. Using enterprise-value-to-sales comparisons with high-growth software companies and normalized EBITDA multiples from companies such as Google, Nvidia, and Facebook, he estimated a central valuation range of $1.5 trillion to $2 trillion. He said the missing October figures would not materially change his view because the S-1 already provides enough information to extrapolate revenue. Peck noted that compute was up four times while revenue rose 11 times, and more than half of roughly $850 billion in commitments extends from 2031 to 2045 and is success-dependent.

Peck, whose career followed the Internet bubble, sees parallels in the emergence of dominant companies after a period of poorly capitalized businesses. He expects some neocloud companies to be vulnerable to higher rates or market turmoil, while larger companies could exploit the next AI wave as Google and Amazon did after the Internet bust. He said the numbers are now measured in trillions rather than the tens of billions discussed in his earlier career. He views Anthropic as a likely positive catalyst for technology sentiment and expects investors to look toward a potential OpenAI IPO in 2027, though he wants to see its S-1 before judging it.

Oil prices declined for a second day as reports suggested diplomacy between the United States and Iran might resume, despite Iranian officials calling the U.S. response to their demands unconvincing and reports of a U.S. strike on Iran’s Qeshm Island. TD Securities still sees Brent at $90 to $100 and warns of spikes while Middle East and Russia-Ukraine conflicts remain unresolved. AAA put the national gasoline average at $4.15, about a two-month low but $1.30 above the start of the year; Brown University estimated the war added $1,150 in fuel costs per American household. Refiners reported strong fuel demand and 92.8% utilization, with Marathon expecting elevated margins into next year.

Ahead of AMD’s earnings, Seema Modi said investors would compare its outlook with Intel’s negative view of PC demand, including the effects of rising memory prices. They also want details on AMD’s data-center strategy, its effort to offer a full-stack alternative to Nvidia, and its relationship with Meta, whose Muse system uses AMD processors. AMD has raised about $5 billion in debt, less than Nvidia and the hyperscalers, while analysts remain broadly positive and point to customers such as Meta, Oracle, and OpenAI. Modi said Lisa Su must show that AMD can secure longer-term contracts as competition expands to Broadcom, Google’s custom chips, Amazon, and other suppliers.