Tactically Cautious, Structurally Constructive

Halloween delivered the right mood. On Friday the Nasdaq fell more than 3%, Google suffered its largest weekly market cap loss on record despite strong earnings, and the word “bubble” was everywhere. Beneath the noise, the picture is far more nuanced.

Start with the scoreboard, because it is calmer than the headlines. The S&P 500 ended October down less than 1%. Nine of eleven sectors rose, and the share of stocks trading above their 50 day average climbed from 20% to 60%. Volatility barely moved. This was a violent rotation inside the market, not a collapse of it.

The pain sat where leverage sat. Hedge funds recorded their largest deleveraging on record and surrendered a year of AI gains within weeks. The hyperscalers, once cash machines, now borrow to fund capex, and investors are asking the right question: where is the revenue? Barclays estimates AI needs about $1 trillion in annual revenue by 2028 to justify current spending, against roughly $400 billion today.

Credit is the signal we watch most closely. AI related debt issuance has reached $1 trillion this year, data center bond spreads are widening, and yields on the weakest junk bonds sit in the mid to high teens. Equity supply is rising too, with 2026 set to be the first year of net positive issuance in decades. Capital is no longer free, and markets are finally pricing that in.

There are offsets. Oracle has capped its future borrowing, Anthropic’s filing revealed far more cash than expected, and major investment banks expect AI financing needs to peak in 2027.

The week ahead brings the midterms, possible Iran escalation, a Treasury refunding announcement and payrolls. Each can move prices in the short term. Historically, though, the year after a midterm has been kind to equities.

So where does that leave us? Tactically cautious, structurally constructive. A further pullback toward the S&P 500’s 200 day average, around 7,300, would be healthy rather than alarming. We continue to favor quality companies, healthcare and the real economy beneficiaries of AI, such as memory and optical networking.

One lesson holds across every cycle: the biggest gains arrive in short, unpredictable bursts. Investors who exit on fear rarely buy back in at better prices. Our job is to stay disciplined, stay diversified and let volatility work for us.