Once a loan is booked, it needs to be reviewed over time for changes in credit. The problem is that many banks have only one type of commercial loan review. This standard review usually requires approximately eight hours of work from credit, loan administration, and management. When this effort is combined with data expense, the report is produced at a cost of just over $1,000 per credit. If you are one of these banks that only have one type of review, then the good news is that you can save a material cost anytime you want AND have better risk management.
Tag: Credit Management
Lots of banks have a limited number of credit grades. For the most part, the average bank has eight different categories, the first four of which are pass grades. If this is your bank, you are ripe to get hurt both in risk management and in loan marketing. Larger banks that use an infinite number of grades in their pricing model or at least use a two-tier system composed of probability of default and loss given default are at a material advantage when competing against a community bank that has a limited number of credit grades.
Sun Tzu, the ancient Chinese strategist, and philosopher had it right – Every battle is won before it is fought. How you prepare for a conflict largely determines the outcome. In banking, this is never truer as the seeds of success or failure are likely already planted in your culture, your operations, and your balance sheet. Yesterday’s post generated a heavier than normal amount of questions around the dangers of loan growth.
Banks put loans on “Watch” in order to better monitor the changes with the borrower, tenant, and property. Whereas a “Special Mention” loan has a potential weakness that deserves management’s close attention, a “Watch” loan may be thought of as a pre-Special Mention and may just require management’s loose attention. While the “Special Mention” classification as a very clear regulatory definition, “Watch” can be more of an economic category.
We are currently considering an interesting loan opportunity for a community bank. The loan is an $11mm term credit to finance the construction and operation of a cold storage facility. The warehouse will be operated by a medium-sized regional company. Our bank was asked to participate in the credit, and our analyst took only a few hours to spread the numbers. All of the cash flow, liquidity, and leverage ratios were analyzed.
When it comes to dealing with commercial property, understanding the timing of liquidation in relationship to a loan’s maturity and the time of default is important on several levels. Knowing the data allows banks to make better loan pricing and loan workout decisions. For example, loss severity is greater for loans with 75% loan-to-value (LTV) than with 100% LTV. This is counterintuitive but can be accounted for by understanding that bankers move faster to liquidate properties that have higher LTVs.
Risk parity is a portfolio allocation strategy that that every bank manager should understand because the concepts are key to understanding how a bank constructs both its balance sheet and its credit portfolio.
You can slice and dice your credit portfolio all you want, but if you are not paying attention to cross-correlations your efforts could be sub-optimal. For example, many banks separate their multifamily exposure away from their single family exposure. In some markets, these two subsectors are almost 80% correlated. A drop in housing prices usually occurs at the same time as a drop in multifamily values and in similar fashion delinquencies at banks usually move in lock-step.
This past December, the regulatory community telegraphed their intentions of focusing bank examinations on commercial real estate (CRE) concentrations in 2016: “During 2016, supervisors from the banking agencies will continue to pay particular attention to potential risks associated with CRE lending” (SR 15 17, Dec 2015).
It was back in 2014 when researchers at DeepMind directed their nascent artificial intelligence application to the game Breakout. Instead of programming DeepMind on how to play the game, the researchers programmed DeepMind to learn about learning to play the game. That is a meta level that isn’t normally programmed but the result of that effort, combined with many others, was in the back of our mind when we turned some artificial intelligent tools towards customer data.