Last month, the Federal Reserve released the economic scenarios that it will use for purposes of its 2018 CCAR and DFAST stress tests.  Because both the overall severity of the scenarios and their individual components each directly impact CCAR results– in particular, they are key determinants of how any bank’s balance sheet will perform under the test – their ultimate impact on banks, and those who rely on them for credit, is enormous.  In practice, the scenarios will in large part determine how much capital banks subject to the tests must hold, whether they may pay dividends or repurchase shares, and if so, at what levels.¹ 

As my colleague Francisco Covas has explained here, the 2018 scenarios are notable because they are significantly more severe than prior years, and thus will likely act as a more stringent restraint on bank capital actions, absent balance sheet adjustments or (less likely) capital issuance.  In this post, we focus on an even more fundamental procedural problem:  Given how important the 2018 scenarios are, how is it we know so little about how and why the Fed chose them?

In trying to assess the Fed’s rationale for its CCAR scenarios and whether they are appropriate, there is surprisingly little to go on. 

Is there a Standard for CCAR Scenario Design?

There are few meaningful legal standards against which to the compare the 2018 scenarios.  The relevant statute under which the stress tests are conducted, section 165(i) of the Dodd-Frank Act, itself says very little – only that the stress tests should evaluate firms’ performance under “adverse economic conditions,” and include “at least 3 different sets of conditions … including baseline, adverse, and severely adverse.” 

The Fed’s actual stress testing regulation is not much clearer, and does little more than restate the relative adverbs.  Specifically, that regulation provides that:

• Scenarios are “those sets of conditions that affect the U.S. economy or the financial condition of a [company] that the Board annually determines are appropriate for use in stress tests, including, but not limited to, baseline, adverse, and severely adverse scenarios;”
• The baseline scenario is a “set of conditions that affect the U.S. economy or the financial condition of a company and that reflect the consensus views of the economic and financial outlook”;
• The adverse scenario is a “set of conditions that affect the U.S. economy or the financial condition of a company that are more adverse than those associated with the baseline scenario and may include trading or other additional components”; and
• The severely adverse scenario is a “set of conditions that affect the U.S. economy or the financial condition of a company and that overall are more severe than those associated with the adverse scenario and may include trading or other additional components.”

In sum, it is difficult to imagine any possible scenario that would not be consistent with the Federal Reserve’s regulation.

The Fed provided more color in its 2013 Policy Statement on Scenario Design.²  That statement indicated that it “intends to use a recession approach to develop the severely adverse scenario,” under which the Fed “will specify the future paths of variables to reflect conditions that characterize post-war U.S. recessions, generating either a typical or specific recreation of a post-war U.S. recession.”   A key aspect of that approach is an explicit assumption that the unemployment rate under the severely adverse scenario will rise at least 3 to 5 percentage points, but in any event to at least 10 percent, with the rest of the macroeconomic variables set so as to be consistent with that rise in the unemployment rate.  (As Francisco Covas and Bill Nelson explain here, even this basic framework is no longer conceptually coherent.)

Within that general rubric, however, the Fed’s 2013 Policy Statement effectively delegated back to itself unfettered discretion to change all the underlying parameters at any time, in its sole discretion, and without explanation:

“Although the Board does not envision that the broad approach used to develop scenarios will change from year to year, the stress test scenarios will reflect changes in the outlook for economic and financial conditions and changes to specific risks or vulnerabilities that the Board … determines should be considered in the annual stress tests.”

Put more simply, the Fed’s 2013 definition of the various scenarios was:  generally consistent with post-war recessions, unless decided otherwise.

The lack of meaningful standards against which to compare the 2018 scenarios might not be that concerning if the Fed had otherwise undertaken a process to propose, explain, seek public input on, and then finalize those scenarios.  But the Fed’s CCAR scenarios have never been subject to public notice and comment.  (The Federal Reserve has sought public comment on its stress testing rules and policy statement on scenario design; but as described above, those provide only the broad parameters of how the scenario may be formulated, and leave the Fed with enormous discretion to “fill in the blanks” each year.  Indeed, the wide variance in the actual stress scenario used in any given year shows just how wide that discretion is.)

The unilateral way in which scenarios have been announced in February each year during the CCAR/DFAST exercises means, of course, that there is no opportunity for public comment and input into any year’s scenarios.  That in itself is a problem; the Administrative Procedure Act clearly requires that any agency policy action of the import and impact of the annual CCAR/DFAST scenarios be promulgated only through notice and comment rulemaking, rather than by fiat.

But of equal concern, it also means that the Federal Reserve provides no explanation whatsoever as to why it chose the scenarios it did, or how those scenarios enhance the overall safety and soundness of the U.S. banking system.  This leaves us without answers to some remarkably simple yet crucial questions about the scenarios:  Why a 65% drop in the stock market, and not 50% (or 70%)?  Why a 30% decline in housing prices?  Why does the Treasury yield steepen as it does?  Even if unemployment must rise to 10 percent, why does it rise so suddenly, contrary to all past experience?  What about the rest of the 28 variables?  In what way, precisely, is the 2018 scenario “a typical or specific recreation of a post-war U.S. recession”?  What information did the Fed take into account in making each of those decisions?  What alternatives did the Fed consider and reject, and why?  

Inquiring minds want to know.