Forecast, Model, Repeat...

Welcome to the 5th Annual CCAR Cycle

For bank holding companies (BHCs) with more than $50 billion in assets, it’s time to get moving if you haven’t already.   The Federal Reserve just released its scenarios and shocks for the 2015 Comprehensive Capital Analysis & Review (CCAR) exercise, and these banks’ capital plans are due on or before January 5, 2015.

This year, Deutsche Bank joins the ranks, bringing the total CCAR participants up to 31 BHCs.  Using data as of September 30, 2014, these large banks must forecast their projected revenues, losses, reserves, and pro forma capital levels for the next nine quarters beginning with the fourth quarter of 2014 and concluding at the end of the fourth quarter of 2016 under five scenarios:

• Three provided by the Federal Reserve (baseline, adverse, and severely adverse)
• Two developed by the bank holding company (baseline, stress)

Additionally, based on their business models, the regulators flagged some BHCs as “global market shock participants” and/or “counterparty default participants”.  As such, they must incorporate supplementary sets of hypothetical shocks into their CCAR modeling processes.  Banks with significant foreign exposures must meet higher capital levels as part of the “Advanced Approach”.   It’s no surprise that banks can have hundreds if not several thousand people focused on CCAR.

The Federal Reserve can reject capital plans by citing either quantitative or qualitative reasons.  The former being straightforward, most likely indicating that the BHC’s capital level are too low under the stress scenarios.  However the regulators can still reject a BHC’s CCAR submission for qualitative factors related to its stress testing processes and/or overall governance.  Of the Federal Reserve’s five capital plan rejections for the last CCAR cycle, only one BHC was cited for quantitative reasons.  The other four rejections dealt with issues regarding the banks’ CCAR processes even though these bank holding companies met the Federal Reserve’s minimum capital requirements.  With more banks shoring up capital, the Federal Reserve will become more focused on evaluating a bank’s overall CCAR process for this and subsequent CCAR cycles.  It won’t be surprising if poor processes and weak controls are cited at more banks as reasons for rejection this time around.

Now what about those models?

The Federal Reserve has stipulated, “BHCs must provide a comprehensive inventory of the models used in their capital plan projections.”  This means that banks need to be extremely diligent around the governance of their models; they need to be inventoried, cataloged and diligently maintained to prove to the regulators that their models are being taken seriously and recalibrated as needed.  As in the past, the Federal Reserve is expecting models to be well documented and have clear, concise explanations of methodologies and assumptions used throughout the process.

For banks that are heavily reliant on manually obtaining CCAR inputs from disparate, diverse and inconsistent sources, and piecing them together with a battery of spreadsheets, they may be in for a wakeup call from the regulators this time around.  Since CCAR is a BHC-wide exercise, it should be viewed from an enterprise perspective.  That is, CCAR data inputs need to be sourced, cleansed, reconciled and staged into a common area.  Models need to be developed, tested, calibrated, inventoried and maintained in a common repository.  At that point, a consistent reporting process can help ensure accurate submissions.

It appears that CCAR is here to stay, and banks need to incorporate good CCAR governance into their data management, modeling, capital planning, and reporting practices.  CCAR is a complex, multifaceted and evolutionary process, and there is no one way to improve upon it or solve for it all at once.  As CCAR teams get immersed in this year’s exercise, it would be a good time to gain an understanding of how to improve the process in coming years.  Some banks may need to focus on model development while others may need to standardize on a model governance framework.  Other banks may have most of their CCAR processes in place, but can gain efficiencies and improve accuracy by automating the completion of the FR Y-14 reports.  Still other banks may need a consistent way to forecast and store the results data.

No matter how your bank goes about CCAR, there will always be room for improvement.  Is good CCAR governance a competitive differentiator?  Clearly failing a stress test, having your capital plan rejected or spending an excessive amount of resources on the CCAR exercise will put your bank at a disadvantage.  For publicly traded banks, failing may mean changes to dividend and stock repurchase plans, and for foreign owned banks, this could mean restrictions on capital returned to the parents.

What are your challenges with CCAR?  I’d be interested in hearing them.