There are lots of ratios that can be used when trying to compare one bank to another. Some are good, some are bad and some are just too simple to be very effective. Yesterday we discussed the Texas Ratio, so today we pick back up by looking at a few more as we continue to raise awareness.
The liquidity ratio is one of the most basic in banking. It is designed to measure how well a bank can meet its cash and collateral obligations without incurring unacceptable losses. In short, it measures whether a bank can fund its loans and support deposit withdrawals. This ratio is calculated by taking liquid assets and dividing them by total liabilities (or other subsets of liabilities, such as wholesale funding, federal funds, volatile liabilities, etc.). Regulators view liquid assets as a combination of short-term assets, marketable securities and assets that can be pledged. For many banks, this ratio is a decent start to getting a handle on liquidity, but assets and liabilities are chock-full of options and can shift substantially over time, so a simple snapshot can miss the mark. Bankers should use this ratio (or its permutations), to begin their liquidity risk analysis, but more robust analytics are also recommended to ensure the board of directors is kept informed.
The capital ratio is another commonly quoted one. It measures the amount of capital a bank has in proportion to risk-weighted assets. The thought is that the riskier the bank's assets (based on Basel I) the more they "weigh" on the bank's capital and thereby reduce the ratio. The problem with this ratio is that asset weightings are blunt by definition (for instance, nearly all loans are 100% risk-weighted despite huge differences in quality, type, geography and other factors), so this ratio is flawed indeed. Plus, it avoids the question of how much of the capital of the bank is TRUPS vs. common, vs. preferred. Without knowing how much "skin in the game" owners and managers of the bank actually have, it becomes more difficult to determine how creditworthy the bank may actually be. We understand the desire to provide common comparison points among and between banks, but this one is so blunt, it offers little value in our opinion.
Bankers are also very fond of the loan-to-deposit ("LTD") ratio. This takes total loans and divides it by total deposits. That is interesting from 80k feet at Mach 3, but what value does it really deliver? Is it worse to pay up for local deposits or use cheaper FHLB Advances? Only time will tell which strategy works better in which market for which bank, but again, general assumptions can lead to unforeseen consequences. Our suggestion is to try and operate most community banks at about an 85% maximum LTD ratio, as that is what we believe is the optimal point between reasonable leverage and balance sheet flexibility. Most banks around this level have also shown historically that they can profitable over time without tipping over when economic seas get rough. Again, different banks can operate in different ways and those with proper staffing/systems might also be able to increase leverage. Our point once again is that by only looking at a singular and static measurement stick, we cannot get a good picture of the underlying components that support each bank.
Finally, we close with the loan loss reserve ratio (ALLL to loans) used by so many analysts. The problem with this one that jumps off the page is that it combines FAS 5 and 114 into one big lump in the Call Report. Without knowing what FAS 5 looks like by itself (to compare to the FAS 114 component), it is difficult to see which banks are more at risk. For instance, consider a bank that has $10mm in FAS 5 at a 1.0 ALLL and another $10mm in FAS 114 at a 3.0 ALLL. While the Call Report will show a 2.0 ALLL overall, the component story is missing (a 1.0 on the rest of the portfolio is probably too low, while the 3.0 may point to additional problems or lending concentration risks).
We close by reiterating that ratios have their place in bank financial analysis, but keeping them in perspective and challenging underlying assumptions is more important.