BID® Daily Newsletter
Jun 9, 2008

BID® Daily Newsletter

Jun 9, 2008

SUMMER PROFILE COMPARISON


It is summer which means swim suit season. If you are female reader of this column, most likely you are already concerned. If you are a male reader, you most likely don't care. Why the difference? We figured it out last night. The wife, like most women, looks at herself in a variety of ways in the mirror - front, back and sideways. Most men reading this column have never even thought about his profile. As a result, when a guy looks at his reflection, it is usually accompanied by a blissfully ignorant smirk, as he thinks to himself - I still got it. Turning sideways highlights some stark realities that many would like to ignore.
In similar fashion, last summer we highlighted our community bank loan default conditions, as well as introduced our C&I program. With our C&I program, we urged banks to diversify away from CRE and construction, by adding some nationally rated credits that we purchased for our portfolio and the portfolio of many client banks. In an effort to turn sideways, we wanted to highlight some facts from our loan pricing model from what was predicted back in the summer of 2007, before the credit downturn and what the current numbers are telling us.
Last summer, our Loan Pricing Model gave community banks a forward looking probability of default of 1.91% for CRE at a time when historic defaults were about 0.53%. At present, defaults (which are analogous to all delinquencies of 30 days and longer plus non-accruals as a percentage of loans) have been running at 1.85%, or about 0.06% better than predicted. Residential related construction, on the other hand, had a probability of default last year of 3.18% from our model (against historic delinquencies of about 1.40%). Currently, defaults are running a hefty 4.13%, so while we underestimated default probability, in our defense the magnitude and velocity of that change had never been seen before (a fact that we corrected for during our last release). For general non-real estate C&I, we had predicted model defaults of 0.92%, which can be compared to the current 1.17% default rate for community bank C&I. Finally, for our national credit C&I, we predicted defaults would have a rate of 1.05% on average, which can be compared to the current 0.76%.
The takeaways here are several. The most obvious one is that if a bank tried to make it through last year without a risk-adjusted loan pricing model, they probably dramatically underestimated risk and are looking at a sub-8% ROE now. For some banks, using a 3Y or 5Y historical default rate as a basis for loan pricing is a step in the right direction, but still lacks the view of what is ahead for supply and demand, cash flow trends and property values. As default rates start to revert to their 20Y mean, having a forward looking pricing model can instantly provide that latest available data when attempting to make a loan and ensure a bank is taking the most accurate view of risk into account.
The other major takeaway is that the difference between construction, CRE and C&I are stark. If the past year doesn't highlight the need for diversification, nothing will. Consider the fact that we know of no bank that is currently under a regulatory action for credit safety and soundness that had a diversified loan portfolio. While it still is too early in the cycle to tell, almost all banks in trouble have high residential concentrations, CRE or construction concentrations.
For added diversification, we still like our C&I program as a way to quickly add rated earning assets that derive their cash flow from national and international sources that are non-real estate related.
Like seeing yourself in a swimsuit, once you strip down, one can really see what they have to work on. Whether it is more ab crunches or making sure you have better loan diversification and risk-adjusted pricing, it can only result in a better looking profile.
Subscribe to the BID Daily Newsletter to have it delivered by email daily.