On Dec 11, the Wall Street Journal’s lead article in its Money and Investing Section was on the news that the Bank of America (BofA) passed its stress test. This was big news because earlier in the year, the Fed had decided that BofA’s stress test models did not meet standards. BofA had to spend significant time and money making sure they passed this time.
The largest banks are bound to draw the most attention. But, what is interesting about this story is the importance of stress testing and financial modeling. The stress test models are really just a set of models that predict losses and revenue given future events in the economy. For example, how many more defaults can a bank expect if unemployment jumps to 7% and what losses will it take? Having good models shows the regulators that your bank can handle downturns in the economy. But, also, having good models can help you gain insight into your business that you otherwise wouldn’t have.
What does all of this mean to the banking industry?
First, it means that these models are important. If you don’t do them well or the models do not give good results, you will get a lot of negative attention. And, the second big point is that more banks will be required to build these models. Initially, only the largest banks were subject to the stress tests. Now, banks with assets between $10B and $50B are required to build stress test models. I was talking to a banking expert, and he was thinking that it will only be time before the smaller banks need to build these models as well.
The stress test models are data and modeling intensive. Our experience shows that you need to understand the statistical models very well and be able to make judgement calls about the data you need to gather and clean to feed the models.