By John Hintze
Practitioners overwhelmingly against latest FASB proposal on accounting for credit losses.
Proposals for new accounting standards involving financial instruments are invariably controversial, and the Financial Accounting Standards Board’s (FASB) proposal to update accounting for credit losses is no exception.
More than two-thirds of the 362 comment letters submitted by the May 31 deadline were from financial-statement preparers—mostly banks with an added chunk of major nonfinancial corporates filing letters toward the end. Corporate objections centered on how the proposal would distort financial-statement results and its divergence from the International Accounting Standards Board’s (IASB) own credit-loss proposal.
too many to count
One practical issue discussed in nearly all the corporate letters was the FASB proposal’s requirement to forecast losses for all debt instruments, from Treasuries to receivables—a leap into the unknown for most corporates. The proposal is vague about whether the forecasts would be required for each debt instrument or categories of instruments. In either case, forecasts would require gathering quarterly not only internal data about customer sales but macroeconomic data and models, likely from third parties.
UnitedHealth Group, for example, noted in its comment letter that the weighted average credit rating of its investment portfolio is AA, and the percentage of securities required to be written down over each of the last three years was less than 0.2 percent. Eris Rangen, the company’s chief accounting officer, said forecasting for each debt instrument would require creating at least one additional model—a credit loss or worst case scenario—for each financial asset along with new assumptions that must be evaluated and refreshed at each reporting period. This, for more than 7,000 debt securities, would likely require significant new expenditures in personnel and technology, Mr. Rangen wrote.
Donald Robertson, chief accounting officer at mid-size technology firm Sandisk, echoed that concern, saying the additional burden could add head count [to the general and administrative expense] function.”
The head of accounting at a major technology firm said the biggest challenge will be how it is operationalized. “It’s the systems stuff that takes time to implement,” he said, “designing it, getting the budget for it, and implementing it.”
Many unknowns
How much time companies will have remains unclear. Sherif Sakr, a partner in Deloitte’s financial accounting practice, said FASB appears to be moving forward with the credit-loss and classification and measurement proposals’ accounting models developed so far. The proposals represent two of the three prongs of its financial instruments project, which also includes the slower-moving hedge accounting proposal.
Duane Goldsworthy, director of reporting and control at Toyota/Lexus Financial Services, said discussions with the FASB in late May indicated a final standard, similar to the current proposal, may arrive in early 2014. “We anticipate more clarification and specificity on some of the requirements, Goldsworthy said, such as what FASB means for “reasonable and supportable forecasts.” He added, “What guidelines will there be? How far out in the future would forecasts be considered reliable, and what level of documentation would we need to support that?”
FASB noted in a July 10 update that by a nearly three-to-one margin, investors and other financial statement users prefer a model that recognizes all expected credit losses, rather than after a specified threshold is met or recognizing only some. Meanwhile, most preparers, the bulk of commenters, prefer the opposite.
too many hurdles
The diverging paths FASB and the IASB have taken in their approaches toward credit losses could slow the process, given the importance they’ve both publicly placed on harmonizing standards. Nevertheless, the project to develop an expected-loss model to replace the current incurred-loss model, which requires existing evidence of incurring balance sheet losses to recognize impairment, was fueled by the financial crisis, lending a continuing sense of urgency.
“Many came to the conclusion during and after the last financial crisis that the current model results in capturing such losses too little, too late,” Mr. Sakr said. Thus, accounting standard setters are trying to find ways to capture losses early in the credit cycle instead of waiting for actual losses.
Mr. Goldsworthy described the issue as a budget item, if not one likely to have a significant impact. He said Toyota has begun thinking about changes to comply with a final standard, although implementation will wait until it arrives. The company’s current models already enable it to forecast credit losses out 12 to 24 months, and a more lenient clarification of “reasonable and supportable forecasts” could reduce the need for significant operational changes.
Otherwise, the company’s models will have to be adapted to digest more detailed data and perhaps split them to provide short-term, medium-term and long-term forecasts. Another key change will be correlating economic data, especially from third parties, with the company’s own credit-loss forecasts.
Mr. Goldsworthy said Toyota explained to FASB how establishing the correlation going out even a year or two would be new for corporates, and those correlations become less certain with time.
“It’s less difficult to forecast what economic events probably will take place. The hard part is modeling the specific impact on the company’s portfolio,” Mr. Goldsworthy said.
Auditors, however, require strong support for those correlations, especially when third parties are involved.
“Why are we able to rely on that third party? Is the party credible, does it have controls we’re comfortable with? There are a number of steps auditors require if a company is relying on a third party,” Mr. Goldsworthy said, adding, “The further out you go, the more support you need because it becomes less certain.”
The head of accounting at the technology company pointed to accounts receivables, noting that the internal data supporting credit-loss forecasts would largely come from customer histories—data most firms likely already have to assess likely credit losses on receivables. The big change would come from incorporating economic factors impacting the company and its customers and suppliers.
“How do you justify all your assumptions?” the executive said, noting any subjectivity in determining which factors to emphasize could be construed as attempts to manage earnings. Auditors are going to ask why did you pick this and not that, and how did you build it into your model,” the executive said.
He added that any changes the company pursues to comply with the new standard will most likely start with models it developed to comply with FASB’s 2009 guidance to determine whether a security is temporarily impaired or faces longer-term troubles.
“We would try to leverage that and broaden it out to make it useful for this new model,” the executive said.