This is the first in a three-part series from guest blogger, Cris deRitis, PhD, Senior Director at Moody’s Analytics
An obscure change in accounting rules could have a major impact on consumer credit pricing and availability
The Financial Accounting Standards Board (FASB) issued an Accounting Standards Update on Financial Instruments – Credit Losses (Topic 326) in June of 2016. Scheduled to go in effect at the start of 2020 for SEC registered institutions, the rule more commonly known as “Current Expected Credit Losses,” or CECL, could have a profound impact on the availability of capital and the preferences of lenders to offer certain lending products.
What is CECL
How it should be implemented
What impact it could have on the lending industry
While we focus predominately on the consumer credit industry, the CECL rule will impact the accounting treatment for all credit assets on a financial institution’s balance sheet, including loans to large and small businesses, as well as mortgages used to purchase commercial real estate.
Accounting for CECL
Among other things, the CECL rule will change the way that banks, credit unions and other financial institutions will have to set aside money — or provision for loan losses. Currently, institutions use an “incurred loss” method defined by the FASB under standards FAS 5 and FAS 114 for determining when and how much capital to set aside in anticipation of losses in their loan portfolios.
Under these guidelines, a lender will originate a loan or issue a credit card to a borrower, start collecting monthly payments and then wait for any signs of borrower distress before adding to its loss reserves. Once an account crosses a “probable threshold of loss” — meaning that the borrower has missed several payments, filed for bankruptcy or displayed other signs of distress then the lender will make an estimate of the size of the potential loss on the account and add this amount to its loss reserves. If the account should cure with the signs of distress no longer being present, then the addition to reserves may be reversed. Otherwise, the intent of the reserve amount is to cover losses that were incurred when the account was charged off.
The incurred loss approach works well in a steady environment where loss rates are relatively stable. Loss reserve levels will increase and decrease along with the business cycle but would remain within a relatively tight range as long as lending standards don’t loosen appreciably and the economy does not collapse. The approach has merit in that it’s relatively easy to implement and interpret given its short-term horizon. Relatively few loans are able to recover from a state of serious delinquency so the likelihood of having to reverse provisions for loans that cure is low.
However, the lack of foresight in the methodology -- which makes the calculations easy -- is also its greatest weakness. By waiting for trouble to be revealed, lenders may need to scramble to raise capital in the face of rising losses. Bank shareholders may learn about the riskiness of the loan portfolio only after delinquency rates start rising; too late to demand changes to lending practices.
This “too little, too late” behavior is precisely what occurred during the Great Recession which saw loan loss ratios hit a record low in 2006 before skyrocketing to record highs within a few short years. This volatility and the delayed provision of information to investors were the key motivations for FASB to change the loss accounting rules in favor of a more forward looking procedure.
Loss Allowances Fell During the Housing Boom and Skyrocketed During the Recession
Under CECL, lenders will need to make a projection of expected loss for each of the loans they book at the time of origination. Rather than waiting for borrower performance to deteriorate, they will need to develop an estimate of the likelihood that any given loan may not pay back and the losses that may be incurred net of recoveries. These estimates will need to consider historical information, current conditions and reasonable and supportable forecasts in order to provide auditors, regulators and ultimately shareholders, with management’s best estimate of the expected lifetime losses for each of the loans on their books.
This is a significant departure from current practices and is understandably causing anxiety among both lending institutions and auditors. Switching to a measure of potential lifetime loss will not only increase banks’ allowances for loan and lease losses (ALLL), it will dramatically change the timing of those provisions. Whereas today a lender can use the interest and principal payments collected early on in the life of new loans to build capital in anticipation of defaults, under CECL they’ll need to add to their reserves before having collected even $1 in loan payments. This could change the economics of the transaction and lead to higher fees or interest rates.
Perhaps the greatest challenge for lenders adopting CECL is the need to generate a “reasonable and supportable” forecast. While straightforward and rationale in theory, the directive generates a host of questions:
Is it preferable to use national level, consensus forecasts or should institutions customize their views to their own geographic footprints?
Should they base their projections on one projection of the future or should they use multiple scenarios to present a fuller picture of potential risks?
What makes a forecast “reasonable and supportable”?
The policies and procedures that lenders develop around their forecasting processes will be just as important – if not more important – than the actual numbers they report. Analysts and shareholders will be scrutinizing financial statements to understand how they compare with other institutions. Auditors will be looking for any signs to unrealistic assumptions that might swing in either a positive or a negative direction.
The cost to comply with CECL will vary across institutions. Banks and credit unions that have conducted The Comprehensive Capital Analysis and Review stress tests may be able to leverage their existing loss forecasting infrastructure to generate CECL estimates. But their audit requirements may also be higher than smaller institutions that may be able to use simpler approaches for their calculations. In our next blog we will explore some of the modeling options that institutions have.
The impact of CECL on institutions’ current ALLL reserve levels – and their financial bottom-line – will vary depending on the composition of their portfolios. For example, the change in reserves on short-term personal loans may be small in moving from an incurred loss approach to a lifetime estimate. On the other hand, lenders with longer term assets such as mortgages or student loans could see a significant impact. Estimates may vary by other dimensions such as the credit score distribution of borrowers or the distribution of transactors versus revolvers in credit card portfolios. We explore these and other implications of CECL on the consumer credit market and the economy more broadly in our third blog installment.