The Federal Reserve [1] has been disclosing results of “stress tests” each year since 2009 for the nation’s largest bank holding companies (BHCs) to see how the banks would fare in “severely adverse” economic situations. Now that the Fed has been disclosing stress test results for seven years, some commentators have recently questioned whether the tests provide any new information to financial markets or if the tests are just offering up the same old data.
A recent analysis [2] by the Federal Reserve Bank of New York [3]’s Research and Statistics Group examined the impact of stress tests on financial markets and found that while market reaction to stress tests has been mixed, the tests do still provide new information that can be useful.
“Our results suggest that the market response to supervisory stress test disclosures remains significant even for disclosures in recent years,” wrote the authors of the paper, Beverly Hirtle, Anna Kovner, and Samantha Zeller.
The authors of the analysis examined alternative measures of market reaction to stress tests in a recent research paper [4]. The first measure was the average absolute value of the cumulative abnormal change (over three days) in share prices (CAR) around stress tests announcements. If investors react to the results, the CAR measure will be larger than its normal value on days without the stress test announcements, according to the authors. The second measure is abnormal trading volume (CAV), which is based on the assumption that trading volume will increase if new information affects investors’ prior beliefs. The third measure the authors used is the changes in credit default swaps (CDS) spreads (taking the absolute value cumulative abnormal spread change, or CACDS), which tend to provide more informative results that vary more with downside risk, since stress testing is oriented toward performance in stressful macroeconomic and financial market times, according to the authors.
The calculations include the nine dates for which the Fed has disclosed stress test results starting with the first tests (SCAP in 2009) and continuing through results disclosed associated with the Comprehensive Capital Analysis Review (CCAR [5]) staring in 2011 and disclosures associated with the Dodd-Frank Act stress test (DFAST [6]) starting in 2013.
The results suggest that while measures were highest for the original SCAP disclosures in 2009 and dropped off from 2009 to 2012, measures have remained roughly equal across 2013 through 2015 dates.
“In summary, we find that while the size of the market reaction to U.S. supervisory stress test disclosures has decreased since the 2009 SCAP (Supervisory Capital Assessment Program [7]), the disclosures seem to provide meaningful information to investors and other capital market participants—even for the recent stress test results,” the analysts wrote. “Our findings suggest that supervisory stress test disclosures continue to provide new information, in addition to serving as a supervisory mechanism for assessing the capital adequacy of large, complex BHCs.”