Cost-of-Funds Command Centre
The blended rate across the book is holding: the weighted-average FD rate came in at , exactly where it normally sits. One cohort has not. The affluent uninsured cohort repriced its weighted-average FD rate up to . FD interest paid on it rose to over the same window. The book behind it did not grow at all. Net CASA flow stayed negative throughout. Margin softens hardest in Arizona, where net interest margin fell to .
So what? The blended rate looks controlled, so nothing here reaches a rate committee on its own. Meanwhile one cohort is being paid more for balances it was never going to move, and every basis point of that earns straight out of net interest income. Left alone it repeats at the next rate sheet.
6.35 vs 4.49% book
5.55 vs 4.49% book
4.25 vs 4.49% book
1.36 vs 1.00×
1.99 vs 2.15%
1.79 vs 2.15% expected
Rate paid in basis points, with the balance behind each one.
The scale runs from 0 to 500.
The rate-sensitive books — online savings and the high-net-worth money market — are where balance thins as the rate paid climbs. That is deposit beta, and the two series move against each other across the whole product set.
Almost all of it is deposit-side. The four steps sum to the fall.
| Move | Running margin | Side of the book | |
|---|---|---|---|
| Prior margin | — | 2.15% | where it started |
| FD rate repricing | −0.20 | 1.95% | deposit |
| CASA savings-rate hike | −0.09 | 1.86% | deposit |
| Mix into interest-bearing | −0.05 | 1.81% | deposit |
| Asset-yield offset | +0.01 | 1.82% | asset |
| Now | −0.33 | 1.82% | deposit-side |
Almost the entire compression is deposit-side: the FD repricing plus the savings-rate hike, with a small mix shift into interest-bearing balances. Asset yields barely move it.
That state pairs the highest FD rate paid with the lowest margin.
Net interest margin
FD rate paid
Net interest margin in Arizona has fallen to , below the range the state is expected to hold, and the FD rate paid there has climbed to . Highest rate paid, thinnest margin — the clearest state-level read on where paying up is costing the most.
It is one cohort paying up, not the book
The blended book rate holds at . Only the affluent uninsured cohort has moved, repricing to . One cohort, not the franchise.
Why it matters: A franchise-wide rate defence would spend on every cohort to fix one.
Paid more, did not grow
FD interest paid on the cohort rose to over the window, while the book behind it did not grow — the tell that rules out buying growth.
Why it matters: Growth would have justified the interest bill. A flat book does not.
The clock is net interest income
Book net interest margin has slipped to , and in Arizona it has fallen further, to . Every blanket basis point on a book that is not leaving earns straight into that.
Why it matters: It is the line the bank is measured on, and the repricing lands there first.
Is the interest bill up because the book grew, or because it was repriced?
Repriced — the rate rose from 4.49% to 6.35% and the book did not grow. Deposit beta, not growth.
Which cohort and state is compressing margin the most?
Affluent uninsured, and Arizona — the highest FD rate paid at 4.71% with the lowest margin at 1.79%.
If rate is held only on the tier actually leaving, does margin recover?
Yes — a targeted defence on the at-risk affluent money-market tier holds the flight-prone balances without the blanket giveaway.