Now that the details of this year’s CCAR stress tests are starting to fade, the thing that most people remember are Zions Bancorporation (NASDAQ:ZION) failing grade and a handful of rejected capital plans. That this was the first year the Federal Reserve used independent projections, instead of relying on banks’ own balance sheet projections, seems to have been forgotten as critics call for more transparency from the Fed.

“We estimate that un-explained changes to the Fed’s application of a generally consistent (if not slightly better) stress scenario in 2014, wiped out a cumulative $65b of excess capital from the large-cap banks relative to results from the 2013 CCAR test,” writes Bernstein Research senior analyst John McDonald. “This includes implied capital hits of $30b at Citigroup, $18b at Bank of America Corp (NYSE:BAC), $11b at JPMorgan Chase & Co. (NYSE:JPM), $2b at PNC Financial Services Group Inc (NYSE:PNC), $2b at U.S. Bancorp (NYSE:USB) and $1b at Wells Fargo & Co (NYSE:WFC).”

excess capital lost ccar 0514

Variability shows the CCAR needed to be more streamlined

McDonald argues that the variability between 2013 CCAR test results and the 2014 results, as well as the variability in ‘lost’ excess capital between different banks shows that the Fed’s process is opaque and could have a chilling effect on banks trying to figure out the best capital return plans for them and their investors. The second part may be true, but the variability is actually the reason for switching to independent projections in the first place.

The CCAR is complex, as it has to be to forecast the impact that a crisis would have on the banks, and its application leaves room for interpretation. If different assumptions are built into the projections for each bank, then side-by-side comparisons become less meaningful. The variation between the banks’ internal models and the Fed’s results shows that streamlining was needed, even if it came as a shock this year.

CCAR stress test company v fed results 0514

CCAR shouldn’t be a guessing game: McDonald

“With the banks having added thousands of staff and invested millions of dollars in this process and shown great improvement in their overall approach and results, it seems time now for the Fed to allow the CCAR/DFAST process to evolve and mature into more of a cooperative effort and less of a guessing game,” writes McDonald.

While McDonald has skipped over the most likely cause for the year-on-year changes to the CCAR results, he’s right that it doesn’t make sense to force banks to guess how much regulatory capital they need to have on hand. Releasing more information on how the Fed models a financial shock and why the estimates were so different would allow for better long-term planning. The Fed might think that banks were being overly optimistic with their internal projections, but the banks have a right to know how they are going to be assessed by regulators.