The Fed's assessment of the 19 bank-holding companies' vulnerability to an economic stress event underscores one of the major challenges in conducting "what-if" types of scenario analysis; namely that it seems to be fighting the last war.
This argument is not news, but the way in which the Fed executed the latest round of stress tests supports this theory. Beyond the fact that the Fed is focusing on sins of the immediate past, the outcome exposes the banks and their investors to unnecessary market volatility as we observed over the last couple of days.
The Fed should get a lot of credit for taking on the daunting task of running each bank's balance sheet and income statement through its paces, but in the end, it is not clear what exactly was achieved.
First, the Fed has not clearly articulated a rationale for the complicated stress test it selected. It appears that with its selection, the scenario is closer to Armageddon than the Great Recession or even the European sovereign debt crisis.
Imagine what social unrest would result in this country should we see an additional more than 20% decline in home prices from where they've already fallen. Compared to the nearly 26% decline in home prices observed between 1928-1933 during the Great Depression, the Fed is clearly obsessing over home prices. Compound that with a 13% unemployment rate and a massive decline in GDP and you have a scenario unlike one ever observed.
Now, to be fair, coming up with a severe stress test is in the eye of the beholder and erring on the side of caution is prudent. However, putting all the chips on a single stress test such as this one introduces a number of problems on the firms under review, their investors and over time undermines the credibility of the stress test process which as stated earlier is a worthwhile objective.
There are other ways to establish a viable framework for assessing individual bank capital strength, and it starts by introducing the concept of stress test diversification.
Rather than rely on a single stress test, an independent advisory board made up of leading scholars, industry practitioners, and fellow regulators could assist in the development of a set of stress tests that obtain widespread support among subject matter experts. This would enable the Fed to appear less insular in their policymaking, which has hampered their stature over the last few years among the public at-large. Using a small number of additional stress tests reduces the chances that the results for some firms are simply due to anomalous sensitivities rather than repeatable severe events of different forms.
With all the assumptions and model risk inherent in the current process, does anyone really believe that Citigroup’s minimum capital ratio is just .1% below the target? The process has considerable margin of error that exposes each firm to fluctuations in capital ratios that may just be driven by idiosyncratic stress test assumptions.
The second enhancement to the process must include establishing a set of possible scenarios that go beyond what we've just recently experienced.
A third enhancement to making this process truly comprehensive in scope requires introducing liquidity risk into the assessment. Its omission from the latest stress tests is one of the more surprising aspects of the methodology given its prominence in other financial crises.
In addition, if the Fed seeks greater credibility for the stress test results, then it needs to be more transparent. This starts with providing the details behind the models, assumptions and data used in its assessment to the proposed stress test advisory board so this body can conduct a thorough validation.
Moreover, the Fed should provide more details on the sensitivity of specific bank portfolios to the various macroeconomic assumptions. The regulators impose such requirements on banks for their models, so why isn’t that also the case for the Fed's stress test models?
Thus far, the Fed has succeeded at least in drawing attention to stress tests and providing some means of consistent comparison across banks. But this doesn't imply that the stress test results are 100% accurate. The market's gyrations over bank stocks in the last few days suggest that the Fed take care in disseminating highly sensitive information based on a stylized, unique economic path. Doubling down on the last crisis via the current stress test may not achieve the desired results, whatever they may be.
Clifford Rossi is executive-in-residence and Tyser Teaching Fellow at the University of Maryland’s Robert H. Smith School of Business.