Bank supervisors in the United States rate every institution on a six-part scale under the Uniform Financial Institutions Rating System, universally called CAMELS: Capital adequacy, Asset quality, Management, Earnings, Liquidity, and Sensitivity to market risk. Each component gets a rating from 1, strongest, to 5, weakest, and the composite is confidential to the bank and its regulator.
The ratings aren't public. The framework is, and it's what turns a KPI panel into an argument. Twelve ratios in a list are twelve facts. Twelve ratios sorted into six failure modes are a diagnosis, because a bank fails through one of the six and the others are usually fine right up until they aren't.
How do the CAMELS letters map to bank ratios?
The bank KPI panel you already have maps onto the letters cleanly:
| Letter | What it asks | From the existing panel | Added here |
|---|---|---|---|
| C | Can it absorb losses? | CRAR, CET1 | leverage ratio, distance to buffer |
| A | How much of the book is already bad? | GNPA, NNPA, provision coverage, credit cost, slippage, special-mention buckets, restructured book, divergence | charge-off rate as the US reading |
| M | Is it run well? | cost-to-income | the growth gap, jaws, provisioning direction, classification choices |
| E | Does it earn its cost of capital? | net interest margin, return on assets, return on equity | pre-provision operating profit over assets |
| L | Can it fund itself under stress? | current and savings account share, credit-deposit ratio | the liquidity coverage ratio and the net stable funding ratio |
| S | What does a rate move do to it? | nothing | unrealised securities losses against capital, duration, the held-to-maturity shelter |
Four letters are already covered by ratios you can compute, and you've met all of them before. Two aren't, and those two are where banks actually die.
Advertisement
How can you test the management letter?
Management is the letter everyone treats as unquantifiable. You can make it testable. Don't ask whether the managers are good. Ask what they did.
Compare loan growth with deposit growth. Chesapeake Union Bancorp grew loans 13.9241% and deposits 1.8519%, a gap of 12.0722 percentage points, and something has to fund that gap. Look at the efficiency ratio and its direction. It's 59.7222% here, on the boundary between adequate and poor for an American bank, and the trend matters more than the level. Look at whether provisioning rises or falls as asset quality deteriorates, since a bank cutting coverage into a worsening book is protecting reported earnings. Look at classification choices, especially the decision to hold securities at amortised cost instead of fair value. That choice is available, disclosed and consequential. And in India, look at the divergence disclosure, where the supervisor publishes the gap between its own assessment and the bank's.
None of those is a judgement about character. Each is a number or a disclosed action, and together they say more about management than any interview.
What does sensitivity to market risk show?
Sensitivity to market risk is the letter that was theoretical until 2023 and is now the first one experienced analysts check. It asks what a parallel shift in rates does to the value of the bank's assets and to its earnings.
The mechanism is unglamorous. A bank funds itself short and lends long. When rates rise, the market value of its fixed-rate assets falls immediately while its funding reprices upward. If the assets are classified as available for sale, the loss appears in accumulated other comprehensive income. If they are classified as held to maturity, the loss appears nowhere at all except a fair-value footnote, and the bank's capital ratios are unaffected until it needs to sell.
Chesapeake Union carries $7,000 million of agency mortgage-backed securities at amortised cost with a fair value of $5,600 million. The unrealised loss is $1,400 million, which is 31.8182% of the bank's $4,400 million of common equity tier 1 capital. Mark it and CET1 falls to $3,000 million. That's 8.8235% of risk-weighted assets against a reported 12.9412%.
What happens if rates rise another 100 basis points?
Then push it, because rates don't have to stop where they are. The portfolio's modified duration is 5.4 years, and duration prices a change in market value, so it applies to the $5,600 million the book is worth. The $7,000 million it's carried at doesn't come into it. A further 100 basis points costs $5,600 million × 5.4 ÷ 100 = $302.40 million. Adjusted CET1 falls to 7.9341%, and a second 100 basis points takes it to 7.0447%. With a capital conservation buffer putting the effective floor near 7.00%, the bank has 93.41 basis points of headroom on a marked basis after one more rate step. After two it has 4.47 basis points. The buffer stack and the applicable minima are set by the national regulator and they change, so verify against the Federal Reserve's current capital rule and the bank's own disclosed requirement for the reporting date.
The bank is comfortable on four letters, adequate on a fifth, and two rate steps from having no capital headroom at all on the sixth. That's the sentence CAMELS exists to produce.