DX Insights: When Is A Stress Test Not A Stress Test?

In addition to Fair Value pricing of loan portfolios and forward-looking CECL analysis, we do traditional stress testing for our clients.  The steady increase in rates has had several direct and indirect consequences for the various services we provide to lenders.  Below are some observations as we head into the end of the year:

  1. Ceteris paribus, loans are being marked down more today than during peak Covid.  While spreads were certainly higher in the late spring of 2020, we only saw portfolio price declines (CRE, Resi, C&I, Consumer) of 2-3% on average.  The one-two punch of increasing risk-free rates AND spreads may not be a knockout, but folks are certainly feeling a bit unsteady on their feet as Q3 results are going out the door.
  2. For variable rate product, the debt service coverage issue is starting to rear its ugly head.
  3. As cap rates adjust upward, though not necessarily in lockstep with rates, underlying collateral values for commercial real estate are going to decline.
  4. Similarly for residential, according to a Freddie Mac market survey a $2,500 monthly payment could buy a $534,000 home one year ago.  As of three weeks ago, that payment could only sustain a $379,000 home.

To tie this all back to the catchy title, a fair number of our traditional stress testing clients use the Fed’s Supervisory Scenarios.  The assumed conditions clearly did not contemplate the current rate or inflationary environment we are experiencing.  As a result, we are finding some loans that are pricing higher under a 2022 Severely Adverse stress test scenario than today’s mark to market price.  That was a real eye opener for us.

Thus, our advice is to scrutinize your stress test results closely and make sure they are sensible, as models are only as good as their inputs.  Also, if you have not re-underwritten the underlying collateral for your loans recently, we highly recommend it given this year’s volatility.  It is very difficult to have a fulsome understanding of your portfolio risk if you are using stale credit metrics.

Categorised in: , ,