A debate is raging right now as to whether the new loan loss accounting standard, soon to go into effect, will aggravate or alleviate the notoriously abrupt cycles of the banking industry.
Regulators and modelers say the Current Expected Credit Loss model, or CECL, will alleviate cyclicality, while at least two regional banks and an industry group argue otherwise. Who’s right? The answer, it seems, will come down to the choices bankers make when implementing CECL and their view of the future.
CECL requires banks to record losses on assets at origination, rather than waiting until losses become probable. The hope is that, by doing so, banks will be able to prepare more proactively for a downturn.
This debate comes as banks are busy implementing CECL, which goes into effect for some institutions as early as 2020.
Last year, internal analyses conducted by BB&T Corp. and Zions Bancorp. indicated that CECL will make cycles worse compared to the existing framework, which requires banks to record losses only once they become probable.
Both banks found that CECL will force a bank with an adequate allowance to unnecessarily increase it during a downturn. Their concern is that this could make lending at the bottom of a cycle less attractive.
The increase in provisions would “directly and adversely impact retained earnings,” wrote Zions Chief Financial Officer Paul Burdiss in an August 2018 letter, without changing the institution’s ability to absorb losses. BB&T said that adjusting its existing reserves early in a recession, as called for under CECL, would deplete capital “more severely” than the current practice.
BB&T declined to comment, while Zions did not return requests for comment.
The Bank Policy Institute, an industry group representing the nation’s leading banks, said in a July 2018 study that the standard “will make the next recession worse.” CECL’s lifetime approach forces a bank to add reserves every time it makes a loan, which will increase existing reserves during a recession, the group argued.
“The impact on loan allowances due to a change in the macroeconomic forecasts is much higher under CECL,” the study says.
And in Congressional testimony on April 10, JPMorgan Chase & Co. Chairman and CEO Jamie Dimon said CECL could impact community banks’ ability to lend in a recession.
"I do think it’s going to put smaller banks in a position where, when a crisis hits, they’ll virtually have to stop lending because putting up those reserves would be too much at precisely the wrong time,” he said.
Those results are at odds with research conducted by the Federal Reserve and firms like Moody’s Analytics and Prescient Models. Some of the differences can be chalked up to modeling approach and choices; other disagreements center on the very definition of ‘procyclicality.’
Moody’s Analytics believes that CECL will result in “easier underwriting and more lending in recessions, and tighter underwriting and less lending in boom times,” according to a December 2018 paper. The Federal Reserve similarly found that CECL should generally reduce procyclical lending and reserving compared to the current method, according to a March 2018 study.
Yet, both the Fed and Moody’s Analytics concluded that CECL’s ability to temper the credit cycle will vary based on the forecasts and assumptions employed by banks under the framework.
“The most important conclusion is that CECL’s cyclicality is going to depend heavily on how it’s implemented,” says Moody’s Analytics' deputy chief economist Cristian deRitis. “You can … make choices in your implementation that either make it more or less procyclical.”
DeRitis says the “most important” variable in a model’s cyclicality is the collection of future economic forecasts, and that running multiple scenarios could provide banks a baseline loss scenario as well as an upside and downside loss range if the environment changes.
The model and methodology that banks select during CECL implementation could also play a major role in how proactively a bank will be able to build reserves, says Prescient Models’ CEO Joseph Breeden, who looked at how different loan loss methods impact an economic cycle in an August 2018 paper.
A well-designed model, he says, should allow bankers to reserve for losses years in advance of a downturn.
“With a good model, you should pay attention to the trends. If you do CECL right, you will be able to see increasing demands for loss reserves,” he says. “Don't worry about predicting the peak, just pay attention to the trends—up or down—because that's how you're going to manage your business.”
In the final analysis, then, the answer to the question of whether CECL will alleviate or aggravate the cyclical nature of banking will seemingly come down to the sum total of bankers’ choices during implementation and execution.