If every loan made were repaid as promised and scheduled, the allowance for loan and lease losses wouldn’t be necessary. But because that’s not real life, banks must quarterly add or subtract from the ALLL—also referred to as loan loss reserves—to ensure that there is some cushion.
The rub lies in how much to set aside and when. And that’s behind the controversy over the Financial Accounting Standards Board’s pending adoption of the “expected loss” model—“Current Expected Credit Loss” or CECL—which is replacing the incurred-loss model currently in place.
“We expect the adoption of CECL to be a major change for the banking industry,” notes Keefe, Bruyette & Woods in a late 2015 report. Regulators have gone further, calling CECL “the biggest change ever to bank accounting.”
Supporters, opponents duel
Backers of the rule, which evolved over several years since the financial crisis, think the final version, due early in 2016, will be a more timely and accurate means of setting ALLL. Among their claims are that adoption of CECL will leave banks better prepared for a major downturn. They believe that the industry went into the Great Recession with insufficient reserves. Banking representatives and others dispute this.
Opponents of the measure believe it will bring no significant benefit, cause greater volatility in bank financial reporting, and be very expensive, if not unworkable or unrealistic, for community banks and even midsize banks to implement. Some also say that bank reports will become less comparable one to another—potentially meaning that for all the effort and expense, less light and less transparency will result for users of financial statements.
In early February, a key public roundtable (Part 1, Part2) was held by FASB, where representatives of the American Bankers Association, the Independent Community Bankers of America, and bankers from smaller banks met with FASB board members and staff concerning CECL in concept and in implementation by smaller banks. (Regulators and accounting industry representatives also participated.) Meanwhile, FASB staff has been at work finalizing the rule, which at press time was expected to be mandatory for SEC registrant companies in 2019 and for other organizations in 2020.
Writing in advance of the FASB roundtable, community bank attorney Jeff Gerrish blogged about the concept on www.BankingExchange.com: “I personally hope [FASB] gains some ‘perspective’ (translated: ‘common sense’) prior to that time and exempts community banks from CECL compliance.”
While such a development is rare, it is not without precedent. News reports indicated that while individual bankers at the roundtable felt some progress had been made, the two associations were not satisfied by what they heard.
“FASB’s complex accounting proposal would radically change community bank accounting methods, sharply increasing the cost of lending and constricting the flow of credit to local communities,” ICBA said in a statement to the FASB board. The association issued a statement headlined: “Hit Stop Button On Dangerous Accounting Plan.”
ABA’s Michael Gullette, vice-president for accounting and financial management, said in a statement that too much of the CECL roundtable concerned issues resolved some time ago or that were not part of the controversial concept. “As a result, little substantial discussion was given to the underlying source of complexity in the CECL model,” he noted. He said ABA would be reaching out to FASB, auditors, and regulators separately.
Don’t wait; prepare now
Indeed, CECL’s complexity has prompted ABA to advise bankers at institutions of all sizes to begin preparing as early as possible for the shift, even as the industry groups still press for changes.
“It’s sort of like entering the lottery,” says Donna Fisher, ABA’s senior vice-president of tax, accounting, and financial management, in an interview. “You buy your ticket, but you don’t quit your job. You have to assume that CECL’s going to take effect.” ABA plans a series of educational efforts to help banks to prepare. Fisher and Gullette say that many policy, systems, and vendor issues must be addressed, and this can’t be started too soon.
Large institutions certainly will go ahead. Midsize banks may have the greatest difficulty, says banking investor Joshua Siegel. He explains that the largest banks already deal with reserve issues over a broad canvas, both national and international in scope. (Siegel serves as chairman and CEO at StoneCastle Financial Corp.)
The midsize institutions, Siegel continues, will have much more to get their arms around to comply, in terms of research and computations.
Community banks, though not set up for extensive data-gathering and analysis, work on a much smaller, local stage, according to Siegel. ABA notes in a white paper, however, that smaller banks have already told FASB that estimating losses more than a year or two out is difficult for them.
There is no sweet spot, really, but Siegel maintains that CECL will be “a very complicated issue as banks get larger.” Even the largest banks will face a lot of lower-level internal proposals on what is appropriate, and this will filter all the way up to one official, who must rationalize things to one yardstick—about which Siegel says, “Good luck!”
FASB’s life-of-loan concept
Setting up and monitoring the ALLL has been an essential element of bank risk management even before the industry began using the latter term. Reserves also have been controversial, at times the subject of a tug-of-war between an SEC concerned about “earnings management”—alleged usage of reserve levels to control perceptions of the bottom line—and the regulators’ insistence that recognition of difficulties be timely. Reserves also are something analysts watch when parsing bank earnings—i.e. the percentage coming from ongoing operations versus from releasing reserves.
The evolution that resulted in CECL was full of twists and turns. FASB had weighed multiple potential models, including one that would have been in concert with international accounting rulemakers. ABA and ICBA each had their own counterproposals to accommodate change without going fully down FASB’s anticipated road.
At present, banks comply with the principle of “incurred loss” for setting loan loss reserves. That is, in the simplest terms, if an event has happened that will cause a loan to go bad, the bank must weigh the matter and adjust reserves accordingly.
CECL would be very different. Essentially, when a loan is booked, a bank would have to determine up front what the expected losses over the life of the loan could be. The reserves decision is always meant to be deliberative, but such a life-of-loan approach would require a much higher reliance on historical data with which to forecast expected losses, which would then be used to set the ALLL. Auditors would evaluate a bank’s process for establishing reserve amounts.
As ABA explains in a white paper, “A credit loss provision (expense) is recorded effectively at the time of loan origination based on what is expected to happen many years in the future. In practice, loss estimates for both models (incurred loss and expected loss) will normally be performed in pools (not individually), using historical experience as a starting point. Adjustments are then required to fine-tune historical averages and arrive at an actual estimate of losses at the reporting date (incurred model) or to the end of the loan’s expected life (CECL).”
Ultimately, the industry’s objection to CECL lies in the concept itself.
“It’s not the way that banks look at risk,” explains ABA’s Fisher.
Fisher’s associate, Gullette, uses commercial real estate loans as an example of what banks face. CRE credit is cyclical, and a portfolio, historically, may see a decade with low or no losses, followed by a year of high losses.
“How do you work with your auditors and regulators on this when you are not expecting losses most of the time?” questions Gullette. Fisher notes that some portfolios, such as credit cards, consisting of relatively homogenous credit, lend themselves better to portfolio-level evaluation and projection. But CRE and other commercial loans tend to be more specific and individual in nature, even in one market.
Gullette says that regulators have assured the industry that they will work to make application of CECL “scaleable”—appropriate to the size and sophistication of the bank. Much will only become apparent once CECL is out of the barn, and Gullette says that the industry, auditors, and regulators “will be fumbling around for a couple of years.” A FASB transition group, including banking industry representation, has already been established.
“Complexity is the name of the game here,” says Gullette.
Fisher says a major concern for smaller banks is a lack of databases that they can use for estimating life-of-loan losses.
It’s expected that community banks will have to spend more on staff and related costs to make CECL work—something ABA and ICBA believe will put them at competitive disadvantage.
Apples, oranges, carrots?
Arguments that something like CECL would have prevented the financial crisis “are pretty much an illusion,” according to Gullette.
Inasmuch as any question of prevention is more or less moot and unprovable, looking ahead makes more sense: What impact would the transition have on reserve levels, as CECL is currently understood? Estimates vary wildly, hinging on the economic outlook of the estimator and more.
Banks and their auditors will be conferring a great deal about what’s acceptable. “The standard is really about what your individual expectations are, and how you support those,” explains Gullette. “There is the basis here for a lot of confusion, and no one has given that a lot of thought.” While the transition group is in place, ABA would like to see many issues resolved before the standard is finalized.
Even then, all banks will be adding more variables to their deliberations over ALLL, as they project life-of-loan losses. In a table assessing the impact of CECL, Keefe, Bruyette & Woods states: “Comparability of financial results will be reduced both historically and between companies due to the variation of loss accounting methodologies likely to be allowed. Bank disclosure about expected loss assumptions will be a key, but expectations for quality disclosures are low.”
It also points out that management philosophies will drive more of the decision, “since a wider latitude will be given to managements to assess potential reserve requirements, and a wider possible range of outcomes could result.”
Here lies a fundamental concern for Siegel, who as a bank investor is a user of the financial statements that will result from a new take on ALLL. Indeed, part of FASB’s stated rationale for CECL is that investors want an expected loss model.
Yet Siegel, experienced with FASB deliberations, is no fan of CECL at all. With all the variation that can go on behind the scenes, he sees CECL resulting in less transparency and less comparability, so there will be more confusion, not less.
Conceptually, he points out that FASB is meant to be the arbiter of what GAAP—“generally accepted accounting principles”—are. Yet Siegel sees CECL driving more variation, and without seeing everything that went into the reserves decision, investors will have less to go on, not more. There will be less “generally” in GAAP in this regard, he says.
Siegel says there is a point of confusion in the rationale behind CECL in that it derives, in part, from the view of structured securities, which are constructed on a standalone basis. Reserves are built into those as a result. He says that what FASB fails to understand is that banks have the flexibility that securities do not. Banks can retain their earnings and add to their capital outside of the ALLL process.
“That seems to be lost on FASB,” points out Siegel, and so a whole new approach to reserves is being adopted. And banks are already under revised capital standards.
As an investor, Siegel sees CECL as a poor bargain. Potentially, higher reserves will be set, which will idle more resources. Additional money will be spent to support reserve decisions.
“Will CECL make the banking system safer?” asks Siegel. “Yes. But it will make it almost too safe.” And that will be only marginally, while lowering potential earnings, he adds.
More immediately, FASB is supposed to be weighing the cost-benefit balance of CECL. Gullette asks, “Without there being a common understanding of what CECL requires, how can FASB do an adequate cost-benefit test?”
Once CECL is out of the gate, there may be no choice. In its white paper, ABA states that “even if CECL welcomes the use of simple estimation models, a relatively complex model will be required to satisfy auditor demands.”