Stress test scenarios miss the boat

Professor Jonathan Parker

The Dodd-Frank act requires that large financial institutions be more closely monitored for exposure to systematic risks.  A key part of this monitoring is stress testing.  Regulators announce a set of adverse scenarios, and financial institutions calculate and report their losses in each scenario.  Regulators get to see how exposed the institutions are.

The OCC has now released the scenarios for 2014.  You can easily download and look at them here.  The scenarios detail lots of economic outcomes: each scenario has 28 variables that mostly turn bad.  But I am really disappointed – this regulatory approach is not addressing the financial-crisis type exposures at all.

There are only three scenarios, one of which is the baseline scenario. The other two are really two typical recessions – one bad, one worse.  What use is this?  Where are those once a century scenarios, like a house price collapse?  Where are the unlikely but possible bad scenarios like a run on the US dollar or US debt?  How about a large fiscal inflation?  In none of the adverse scenarios do long interest rates.  My suggestion: reverse the detail – more scenarios, each with fewer variables.  There are many factors in asset pricing for a reason – there are many sources of risk.  Regulators seem to think there is only one.  Be more creative. Think about unlikely bad outcomes  That’s the entire benefit of the exercise.  Make banks think about these scenarios; make regulators aware of how bad (or not) the unlikely scenarios would be.

Large financial institutions do not have to evaluate their losses in the scenarios that keep me up at night.  I hope the regulators also do not sleep so well.

Jonathan A. Parker is the International Programs Professor in Management and a Professor of Finance at the MIT Sloan School of Management.