Breadth Collapse and the Nordic Bank Test
2026-10-07 · United States · US market breadth and the AI concentration, Nordic bank Q3 and the credit-cost question
How it is made. Aavistus Daily is written and voiced by our own systems, every morning, from data we gather ourselves. The voices are synthetic. Every fact is traced to a source on this page, and every dated call is settled in public on the calibration page. The presenters, Nora and Callum, are named characters, not people.
Aavistus is a new framework, still under development. We will make mistakes, and every dated call is settled here in public so you can see them; the aim is to keep calibrating until the record beats the average forecaster, not to sound right.
On Unhedged, Armstrong and Martin call this a market with only one trade; a Nordic investor tests Whalen's claim that US bank credit costs have bottomed against Nordic Q3.
Breadth collapse and the Nordic bank test
An index near records, with the market under it
On Unhedged yesterday, Robert Armstrong and Katie Martin called this a market with "only one trade". The sharp number: only twenty-eight per cent of S&P 500 companies trade at or above their fifty-day moving average. The headline index sits near a record because a short list of AI leaders is carrying it, while the rest of the market sags under rising yields. The mechanism is a straightforward rate shock on everything that isn't an AI winner.
The numbers match on this desk. The ten-year Treasury is at five point two seven per cent — up forty-nine basis points on the month — and the thirty-year at five point six four. The real ten-year yield, computed as nominal minus the August CPI print of three point four, is one point eight seven.
Where I'd qualify Armstrong is the catalyst. He leaned on a legal shock on OpenAI or a listing problem on Anthropic as the trigger that would crack the rally. Those are tails. The near-term pressure is already running — real yields at these levels are the thing cracking everything that isn't an AI winner, and the Federal Reserve is still in a hiking cycle. The upper bound is at four, and the curve prices roughly five more quarter-point moves in the next twelve months; the one-year rate one year forward sits at five point one two. The standing view that the upper bound reaches at least four and three-quarters by mid-2027 is intact.
On the long end specifically, Chris Whalen, on Thoughtful Money with Adam Taggart, projects the ten-year climbs to five point five to five point seven five inside a year. That is directionally with me and above the top of my year-end range of five to five and a half. His mechanism leans on supply-shock inflation from damaged refining capacity and fertiliser shortages; mine leans on term premium rebuilding into Treasury supply and continued Fed tightening. Both roads lead above five point four on the ten-year; his road leads further. The dispute settles on breakevens, not on nominals.
The Shiller cyclically-adjusted price-earnings ratio sits at forty-one point nine with the volatility index at fifteen — a market pricing calm on top of a valuation that is mostly a measure of eight companies. The twenty-eight-per-cent breadth reading is not a counter-signal to the view that the broad index settles lower over the next nine months; it is the mechanism of the de-rating itself, slow and index-wide, with the headline level held up by a few names.
The position. My view on the long end — the ten-year holds five to five and a half per cent through year-end, and the read breaks on five consecutive daily closes above five point four with no equity break — is intact. On equities, the case for adding US megacap at that Shiller multiple is not there, and the case for cutting the index on these numbers alone is not there either. On Armstrong's inflation-protected entry: on a stand-alone basis a real yield near three is attractive; on the full policy path, real yields grind higher before they reset, and the entry today is a trade for the second half of 2027, not for the next six months.
Nordic bank Q3 and the credit-cost question
Whalen on the same programme made a second argument: credit expenses across American commercial banks have bottomed after seven or eight quarters of declines and are now reversing upward. If that read on US lenders is right, the symmetric question for a Nordic book is whether Scandinavian bank credit costs are also at the bottom of a cycle.
The Nordic data gives a soft yes on direction, no on magnitude. Unemployment prints moved — Sweden up three-tenths to eight point nine per cent, Norway up three-tenths to four point five. On these alone, you would read Nordic bank credit costs as softening at the edges rather than reversing in a way the Q3 release has to recognise.
My Q3 read turns on US dollar funding costs: the threshold is more than fifty basis points year-over-year widening — the point at which Armstrong's hyperscaler debt-displacement mechanism lands on a Scandinavian balance sheet. Three lines matter in the Q3 release: year-over-year dollar wholesale funding; any restructuring charge or cost program disclosed with the Q3 (operating gain or stress absorption?); and the forward net interest margin guidance into 2027. If the Q3 shows dollar funding more than fifty basis points wider with a restructuring charge that reads like stress absorption, the defensive read on European financials into year-end sharpens; if dollar funding is contained and no significant restructuring sits in the release, the standing view holds without needing to be reinforced.
What to watch
The first Nordic bank Q3 release later this month, where the US rate-shock story meets a Scandinavian balance sheet. And the 29 October ECB meeting, into which the euro area September HICP print carried headline inflation back to three point eight — the test for whether the European Central Bank hikes into reaccelerating inflation.