We all know that building up your nonmaturity deposit accounts is your best interest rate risk strategy. Likely it's your most profitable strategy too.
After all, NMDAs are low (sometimes no) cost, can lead to other relationship transactions and can generate fee income. Let's see you try that with wholesale funds.
The only problem is that bank examiners hate NMDAs. OK, hate might be too strong of a word. They'd never say that, but actions speak louder than words, especially when examiners are concerned.
Here's why your examiners hate NMDAs:
- NMDAs are at historically high % versus time deposits.
Just look at this FDIC chart. That's why your examiner is focused here. These trends worry them. More correctly, the likliehood of these trends reversing worry them.
- NMDAs can't be controlled like other funding.
Think about it. NMDAs are the only items on the balance sheet where you don't have a contractual maturity or rate schedule. If NMDAs are short maturities, your bank automatically models as more liability sensitive. If long, you model as more asset sensitive. Change one assumption and you flip your bank's IRR profile on its head.
- NMDAs must be estimated.
This is where the problem really lies. Because if your examiner is confronted with something that demands judgement, that you can't just "tick and tie", then they start to get a bit nervous. And if that "something" is growing in importance, like NMDAs are, they practically break out in a cold sweat.
What we're really saying here is that your examiners don't understand your NMDAs as well as they do other liabilites. And when examiners don't understand something, it gets lots of regulatory attention at exam time.
Think about it. How did they treat things they didn't understand in the past? Things like CRE, concentrations, liquidity, trust preferred, bankers' banks, brokered deposits, CMOs to name a few.
What they did was put them under the microscope, made you add additional analysis, made you track them in a more detailed way, and made you model the results. And that's exactly what's happening right now.
It's no longer good enough to use generic or default assumptions, even if they work well in your IRR model. No, now you need a deposit study. That's why every bank needs a deposit study.
Because if you're relying on your opinion or for your deposit assumptions, your examiner can always dispute them. But if you have a statistically significant deposit study they can't really argue with your proven statistics.