The Challenge of CECL for Credit Card Issuers
November 6, 2017
For many years, the credit card industry calculated the loan loss allowance (the ALLL) as an estimate of losses projected over the next ten to twelve months. This long-standing practice is now about to be overturned. While card issuers have not disclosed the likely impact of the new approach, a doubling of the loss allowance would not be far-fetched. This change in methodology will only magnify the recent acceleration in credit losses. It is imperative for card issuers to start communicating with the investment community about this change now, before the financial impact is felt. Without a proper understanding of the accounting rule change, investors will conclude an increase in the loan loss allowance is a signal that credit quality is deteriorating, which may or may not be the case.
Why did the accounting profession feel compelled to change a long-standing practice? Back in 2008, when the world was trying to come to grips with the global financial crisis, the Financial Accounting Standards Board (FASB) embarked on a project to improve the estimation of a lender’s credit exposure. The theory was that by moving from an incurred loss model to one which captured the expected credit loss associated with a loan exposure over its complete term, banks and other lenders would be better prepared to endure a future credit crisis. FASB conducted a long evaluation and comment period which ended in June 2016. The result, Accounting Standards Update (ASU 2016-13) Financial Instruments – Credit Losses (topic 326), is not yet adopted. The new standard has far reaching implications and FASB has provided a long lead time for implementation (essentially 2019 or 2020 depending on the classification of the reporting entity).
One of the core concepts of the new ASU was the “Current Expected Credit Loss” standard – or CECL. CECL requires that the lender estimate the expected credit loss over the life of the loan exposure. While this is challenging for any lender, it is easier for amortizing term loans than for revolving loans. In particular, for credit card issuers, determining the “expected life” of a credit card receivable (loan) is inherently complex. What is the life of a credit card loan? One month, one year, ten years? Since credit cards give the borrower a payment option (minimum, partial or full), determining the life of a loan exposure is challenging. Quite obviously, CECL could result in a much higher loss allowance if the life of the average credit card loan were deemed to be longer than one year.
After receiving numerous comment letters from industry participants, the FASB created the Transition Resource Group for Credit Losses (the TRG) so that FASB and the industry could develop practical conventions for implementation of the proposed new standard.
The TRG worked on the knotty problem of the “life of loan” for credit cards. One of the key areas of focus became the application of principal payments. Beyond the borrower’s right to vary the payment amount, there was also the question of the allocation of principal payments received by the issuer. Prior to the Credit Card Accountability Responsibility and Disclosure Act of 2009 (the CARD Act), issuers used various payment allocation methods to prioritize the distribution of principal payments. CARD Act, among many other changes, required issuers to apply principal payments to the outstanding balance with the highest interest rate first, until the high rate balance was reduced to zero or the principal payment was fully utilized. Applying principal this way was clearly designed to benefit consumers by having the highest rate balances amortize faster than lower or teaser rate balances.
When card issuers began to consider how to implement CECL for their business, they faced a quandary: allocating principal payments according to the CARD Act could make the estimation of the life of loan even more complicated. Should an issuer assume that the life of a current card loan outstanding can vary with subsequent purchase activity, even if the issuer is not legally obligated to fund future activity?
In June 2017, the FASB reviewed a paper (and examples) prepared by the TRG to propose one of two possible solutions to the credit card payment allocation problem. The first suggestion, “view A,” was a first in, first out (FIFO) allocation where the oldest balance received the principal payments until it was completely paid (or charged off). This had the virtue of simplicity and consistency. The second suggestion, “view B,” was more complicated but attempted to closely follow the CARD Act allocation. While this may have the theoretical advantage of more accurately reflecting the life of a loan, the complexity of this approach was obvious.
The FASB concluded that while it could see no reason to prohibit either approach, it did see View A as more practical. The staff also recognized that View A was more consistent with the intent of CECL that the loan loss estimate was for the balance at a point in time. Moreover, since credit card lines are unilaterally revocable by the issuer, including future purchases would essentially change this from a “life of loan” calculation to a “life of loans” calculation. At a meeting of the TRG on October 4, 2017, the Board agreed to provide further guidance on the acceptability of both views in a future new guidance Update 2016-13.
The credit card industry has enjoyed years of record profitability and near historic lows in both interest expense and credit loss expense. Now, as credit losses have begun climbing, the move to CECL will have a compounding effect on the loan loss allowance. In addition to the advance warning and education that issuers should be starting, perhaps there will be other actions that issuers can take post-adoption. For example, should issuers calculate the allowance using the old methodology as a “non-GAAP” additional metric to be included in the MD&A section of financial statements? For some long-time industry analysts, this would have the benefit of allowing easier comparisons to prior reporting periods. Perhaps after one or two years of reporting under the new standard the old calculation can be dropped but including it for at least a transition period would no doubt be helpful.
There has been much media attention paid to the rising loss rates for card issuers, but there has been little discussion about the coming impact of CECL on card issuer profitability. The credit card business will remain the most profitable lending area for US banks, but expect the future levels of profitability to “reset” at a lower level. The onus is on the card issuers to help investors understand that, at least in this case, the change in profitability may be more a function of convention than credit.
About Auriemma Consulting Group
Auriemma is a boutique management consulting firm with specialized focus on the Payments and Lending space. We deliver actionable solutions and insights that add value to our clients’ business activities across a broad set of industry topics and disciplines. Founded in 1984, Auriemma has grown from a one-man shop to a nearly 50-person firm with offices in New York and London. For more information, visit Auriemma’s website at www.acg.net or contact John Costa at (212) 323-7000.