20 Mar How Much Better Will Banks Be if CECL Dies? The Answer Might Surprise You
By Adam Mustafa, Invictus Group CEO
The Federal Deposit Insurance Corp. this week did something unusual: FDIC chair Jelena McWilliams sent a letter to the Financial Accounting Standards Board, asking it to delay implementing the current expected credit loss (CECL) standard while banks deal with COVID-19 economic challenges. The public announcement calls for FASB to delay the 2020 start date for publicly traded banks, as well as impose a moratorium on the effective date for banks that must comply in the future.
This very public action was designed to put extreme pressure on FASB. Rarely does one regulatory body take a strong public stance against another. I expect pressure to only increase on FASB to delay CECL. Bank regulators want to remove all pressures on banks to incentivize lending to customers in need.
But if FASB caves on CECL, it would be a colossal embarrassment. FASB has been arguing that CECL would be countercyclical. In other words, it would force banks to hold higher reserves in good times, allowing those reserves to be utilized during tough times so banks wouldn’t have to replenish them until the economy recovers. McWilliams implies that CECL will end up doing just the opposite. She contends if CECL remains in place for the first quarter of 2020, it would deplete banks of precious capital that can be better utilized to support borrowers in need because of the virus.
I have no idea if FASB will stand its ground. My big concern is that it won’t matter very much. Last week, I wrote about how CECL combined with the coronavirus will be an earnings nightmare for banks. Unlike the previous standard, the incurred loss methodology, CECL requires banks to reserve for the entire expected remaining life of the loan. It also requires banks to adjust their expected losses based on ‘current conditions’ and ‘reasonable and supportable forecasts’. For all these reasons, loan loss provision expenses in the first quarter would likely be massive.
So why am I also saying that dumping CECL wouldn’t matter very much? It’s because auditors would be under massive pressure to make sure loan loss reserves spike anyways. Publicly traded community bank executives who think their loan loss reserves will remain the usual 80-125 basis points (between 0.80% and 1.25% of loans outstanding) are kidding themselves. Auditors would be putting the balance sheets of their respective firms at risk. They are also under the watch of their “Big Brother”: the Public Company Accounting Oversight Board (PCAOB), which was created by the Sarbanes-Oxley Act (remember Enron and WorldCom?). My concern is that the auditors will (understandably) transfer this pressure to the community banks themselves, by either directly or indirectly communicating an expectation for a significant increase in loan loss reserves, even if we return to the old incurred loss methodology.
They will ultimately do this through the Q-factors, or qualitative factors. Banks try to use qualitative means to estimate losses that are already probable within their loan portfolio, but are currently unknown. In laymen’s terms, the losses exist, but the banks don’t know about them yet. In recent years, the Q-factors have come to dominate the calculation of the entire reserve because banks simply don’t have material amounts of historical losses. The PCAOB has criticized auditors for the lack of quantitative support behind the loan loss reserve in recent years, creating a bizarre conundrum for community bank CFOs and their auditors. If the loan loss reserve appears too low, the banks get called out. But if they increase their reserves via q-factors, they also get dinged.
This conundrum will now be magnified. I suspect that for auditors, picking their poison will be easy. They will strongly want to avoid being the auditor who let a bank client keep its reserves at usual levels, only to see massive losses materializing in the future. The lessor of two evils is to push banks to further increase their loan loss reserves because of COVID-19 through the mysterious Q-Factors. So, at the end of the day, there will be massive pressure on the loan loss reserve to increase substantially, whether CECL survives this onslaught.
The reality for publicly traded banks is that investors are going to struggle with their judgments on the final reserve, no matter what. This is why CECL was created. Investors practically ignored bank loan loss reserves heading into and in the middle of the 2008 Financial Crisis, making their own estimates instead. This explains why many banks traded below their book value back then and why many are doing so now. If a community bank shoots too low with its reserve, investors are unlikely to believe it. Therefore, most community bank CFOS will find it inevitable, if not appropriate, that the loan loss reserve will have to increase. How much is a mystery, whether CECL remains intact or if we return to the incurred loss methodology.
At the end of the day, most community bank CFOs would prefer to get rid of CECL immediately. But the reason is far less about preserving earnings and capital and more about the time and effort of the calculation itself. The incurred loss methodology is the devil they know, and as the old saying goes, “better the devil you know than the devil you don’t.”
In the long run, CECL or no CECL, the banks that will separate themselves from the rest of the pack with respect to estimating the loan loss reserve will be those that have a competitive advantage with data and analytics. Better data and analytics help banks to more accurately estimate their loan loss reserve, while unlocking strategic value from the process. Stress testing needs to be fully integrated with the loan loss reserve calculation because banks simply don’t have many losses in good times. Those banks that have integrated stress testing with their loan loss reserve calculation certainly won’t get it perfect because nobody knows the economic consequences from COVID-19, but they will be more directionally accurate with their estimate and have far more conviction over their calculation.