DISMAL SCIENCE & BANK RISK MANAGEMENT

Economics is often called the Dismal Science. The inception of this term came about because of a mid-19th century economic model. It forecasted imbalances between the growth of food production and population growth would lead to massive starvation. That particular prediction was dismal indeed and thankfully wrong. The science of economics relies upon models. Sometimes they work pretty well and sometimes they leave something to be desired. A recent article on this subject caught our eye. The discussion was around the standard economic models that central bankers and policymakers typically use to make projections. The standard mainstream macro model is called a Dynamic Stochastic General Equilibrium (DSGE) model. The problem with DSGE models is that it has come to light they are startlingly inaccurate. This is because they don't represent the modern financial system, nor do they allow for the boom and bust cycles that occur in real life. DSGE models also specifically exclude banks. Banks are seen as simply facilitators between depositors and borrowers, rather than for-profit entities trying to make money by finding opportunity in the differences between liability costs and return on assets. Needless to say, DSGE models are really just Darn Stinky Gauges for Everyone, so change is afoot among the dismal scientists. A Princeton study showed the internal risk models used by banks often lead banks to take on more and more risk as asset prices rise. A Yale economist has long warned that small changes in the appetite for lending against a class of assets, can lead to large effects on the price of those assets. The result of those two ideas creates a scenario where loose lending standards would allow speculative borrowers with limited cash to bid prices far higher than the underlying asset would normally support. This is pretty familiar territory when you consider the run-up to most real estate bubbles. To improve, updating the models to accurately represent the financial sector would be a big step and might produce more accurate economic predictions from central banks. This macro-economic concept would be appropriate to apply to community bank internal risk models as well. During real estate bubbles, the run-up in asset prices primarily is related to the collateral. Today, we see a rise in risky behavior now on the asset side of the balance sheet of banks. Competition for the strongest borrowers has led those borrowers to demand lower and lower interest rates on their loans and banks are capitulating to keep the business. The question is whether the bank is being adequately paid for the level of risk it is taking on. To protect your bank, it is wise to take a scientific approach to measure risk in your lending practices if possible. Start by measuring the probability of default and loss given default of a loan, especially before making a deal in such a highly competitive market. Another area where we see community banks reaching for return is in the investment portfolio. The Fed is holding interest rates on the safest assets very low, so margin compression is extreme. That is driving banks to take on interest rate risk by purchasing longer maturity securities or significant embedded structure or credit risk (step-ups, municipalities or corporate bonds) to achieve yield hurdles. This calls for extreme caution right now and exceptionally detailed due diligence. No matter the modeling techniques you currently use, there is always room for improvement. Consider whether your models have been validated and are accurate to go a long way towards keeping your bank clear of dismal results.