THE HUMAN BRAIN AND LOAN RESERVES

When it comes to motor coordination, humans are not all that quick. Poke a sharp pencil in your CFOs back and it will take almost 2 milliseconds for them to react. Dogs, cats, insects and wildebeests happen to be a whole lot faster (by 60%), despite having smaller brain sizes. With this lack of quickness, it is amazing humans can catch a ball, shoot a clay pigeon out of the air or flip the channel the microsecond Maury Povich comes on. The answer lies in the way the brain processes information. Humans create internal templates of the physical world and then look for inputs such as the speed of a ball or distance. What humans lack in process speed, they make up for by having a pre-planned mental model. This explains why you may not be any good at your first game of squash, but can put an overhead topspin serve into the far corner of the let court at your 150th tennis game. One area that lenders are learning to develop an internal map about is when it comes to borrower loan reserves. Lenders have often required commercial borrowers to set aside monies in a separate account to pay for items like taxes, insurance, improvements and vendor fees. Many lenders go a step farther and require their borrowers to build up a "safety" or "buffer" account to protect debt service should their be a hiccup in cash flow such as a tenant bankruptcy or loss of a lease. These reserve accounts not only mitigate credit risk, but also increase the profitability of a loan with their deposit value by an average of almost 1% of risk-adjusted return on equity. For April as an example, the average $1mm non-construction commercial loan for banks produced a 21.07% risk-adjusted ROE if it had reserve balances of approximately 5% of the notional balance and 20.21% if there were no reserves required. One issue that has changed the trajectory of loan risk is the fact that these loan reserves have been dropping for both publicly traded loans as well as loans for independent banks. On average, reserves on loans are down 5% from 1Q 2007 compared to 1Q 2006. This is a result of 2 reasons. One, cost overruns on construction projects and higher labor/materials cost on maintenance have dropped reserves more than forecasted. Two, borrowers have been more successful in negotiating lower amounts with banks due to competition. Loan reserves in Northern Virginia are down the most, followed by NY, Los Angeles, Portland, Seattle, San Francisco and most metro areas in Texas. Chicago, Kansas City, Atlanta and Minneapolis are up. Reserves on office loans are way down, while reserves on retail, mixed use, hospitality and industrial are moderately down. Reserves on multifamily and specialty loans are slightly up. What all this means is anyone's guess, as we just don't have that internal map figured out just yet. At the next downturn, it is our hope that we can figure out what reserve levels mean to lending risk as with more than 150 years of lending, no one has publicly answered the question. While credit risk is increased by holding lower reserves, borrowers that negotiate these levels down are usually stronger and in healthier markets, thus mitigating much of the exposure. Since the converse is also true, the higher levels in Phoenix may just well portend defaults. The reality is that at the next downturn, we will be as slow as your speared CFO to react to this reserve level data until we figure it out.