[IMGCAP(1)]The post credit crisis regulatory focus on forward-looking techniques and loan behavior models has been burdensome for some banks.

Already strained by the regulatory stress testing requirements, banks are feeling the pressure of the Financial Accounting Standards Board’s upcoming Proposed Accounting Standards Update (Subtopic 825-15), more commonly referred to as the Current Expected Credit Loss model, or CECL. With the final version expected to be issued sometime in the first part of 2015 as part of FASB's financial instruments standard, banks are starting to think about how they will prepare for the proposed accounting standards update.

There is a lot of well-intended, but misguided advice out there. In order to properly prepare for the upcoming guidance, we have set out to dispel some of the myths that have shadowed CECL since the exposure draft was first issued in 2012. Here we address some of the biggest misconceptions that we have come across with respect to preparing for CECL. 

Myth #1: CECL is a data issue. False.
It is true that data, and lots of it, will be necessary in order to meet the disclosure requirements. Accurate records of historical events and instrument-level risk attributes will serve as the foundation for the forward-looking loss models.

Life of loan activity such as payments, write-offs, sales, charge-offs and foreclosures will be required in order to create a roll-forward of the debt instruments measured at amortized cost. Instrument-level credit quality attributes will be necessary in order to describe the risk of the portfolio. However, a data warehouse alone is not going to solve the CECL problem, and banks should be doing more than just collecting data between now and the time that CECL is effective.

Truth: CECL will require data integration.
By all means start gathering portfolio data now, but that will only be half the battle. For banks that do not have a centralized platform, the challenge will be the integration of data from multiple systems.

Typically borrower activity and balances are tracked in the servicing system; risk ratings and other credit attributes are located in an internal risk management system, while estimates like the allowance calculation are stored elsewhere, sometimes in a spreadsheet. CECL will require cross-referencing data across all of these systems, presenting challenges and control risks if they are not automatically integrated. Automatic integration between all systems is the solution to the data challenges presented by CECL.

Myth #2: CECL is an accounting problem. False.
CECL is not just an accounting problem. True, CECL is a proposed accounting standards update, but the risk department needs to pay attention too.

The CECL disclosure requirements go beyond today’s credit quality disclosures (ASU 2010-20). CECL adds a new element, the future economic forecast, to today’s incurred loss model, and the risk team will play a critical role in developing the forward looking loss estimate.

Not only will banks need to calculate a life of instrument allowance using a forward-looking economic forecast, but as part of the disclosure requirements, banks will need to attribute changes in the expected loss to changes in the credit attributes of the portfolio, the composition of the portfolio, and the assumptions used to forecast future economic conditions. CECL will require a comprehensive understanding of the portfolio on both the risk and finance sides of the house.

Truth: CECL will require integration of risk and accounting.
Today most accounting and risk departments work independently of one another, interacting once a quarter to volley the allowance estimate over the net. The first step in integrating the accounting and credit departments is an automated solution that provides one source of truth in one location, which is accessed and used by both teams. This starts with integrating the data (see myth #1), but it does not stop there.

The same application should be used by both the accounting and risk teams to track the portfolio performance and to execute the accounting calculations and estimates. The more seamlessly that the accounting and risk groups operate together, the easier it will be for banks to meet the CECL disclosure requirements.

Myth #3: CECL is a modeling issue. False.
The proposed standard does not mandate a specific method to estimate the expected credit loss. In other words, a variety of methods are acceptable: discounted cash flows, historical loss rate averages, probabilities of default, roll rate and migration analysis.

However, the estimate is required to reflect past events, current conditions, and reasonable and supportable forecasts about the future. Larger institutions required to perform stress testing today may already use models, while smaller institutions, on the other hand, may not be practiced in forward-looking estimates. Regardless of your institution’s experience with regulatory stress testing, CECL will require more than just procuring the right model.

Truth: CECL impacts the end to end process.
Using a model for regulatory stress testing is very different from using a model for the estimated allowance. The allowance is a significant accounting estimate based on management judgment. The forward-looking economic forecast adds more complexity and subjectivity to the estimate, increasing the likelihood of a misstatement. Because the allowance is on the face of the audited financial statements, the updates, maintenance and execution of the models must be documented, monitored and tested under Sarbanes-Oxley.

Ideally, the models used to estimate the allowance are integrated on the same controlled application which is used by both risk and finance (see myths #1 and #2).   

Myth #4: CECL will be a challenge to implement. It depends.
It is natural to address regulatory changes piecemeal as they become requirements. Typically each new or updated requirement is small and usually addressed with spreadsheets and/or manual processes.
Collectively, though, the manual effort and lack of control creates an operational environment that is highly inefficient and prone to error. CECL is not an isolated data, accounting or modeling issue, but it will have a widespread impact on risk and finance.

If we are honest with ourselves, these CECL myths are not new challenges, but are inefficiencies inherent in our everyday operations. While accounting and regulatory updates force us to look at our processes, they also provide us with an opportunity to do things better and faster. Accounting and regulatory changes should be addressed holistically with consideration to the long-term impact rather than through another non-integrated point solution.

Truth: Solving CECL in an integrated way, leveraging today’s technology, enables banks to do things better, faster and cheaper.
Implementing CECL can be viewed as a challenge or an opportunity. Because CECL is unique to the financial reforms preceding it—in that it will have a larger and more widespread impact across multiple functions of the financial institution—it opens the door for a broader overhaul of all manual and spreadsheet processes. The biggest mistake that institutions will make when preparing for CECL is to solve for it in isolation.

The technology to solve CECL in a holistic manner already exists. The efficiency gains, improved accuracy, detailed value-added analysis, and improved controls that banks strive to achieve are realizable, but making this happen requires a change in approach and a more effective use of available technology.

It will require implementing a thoughtfully designed, integrated solution that has robust and well-architected data architecture. While the initial investment of this approach may be slightly higher than an isolated solution, banks that solve CECL in an integrated way will be able to do more, better, faster, and cheaper.

Lauren Smith, CPA, is a senior product manager at Primatics Financial.

Register or login for access to this item and much more

All Accounting Today content is archived after seven days.

Community members receive:
  • All recent and archived articles
  • Conference offers and updates
  • A full menu of enewsletter options
  • Web seminars, white papers, ebooks

Don't have an account? Register for Free Unlimited Access