Get ready to fire up those credit-loss prediction models.

The Financial Accounting Standards Board has issued a proposed accounting standards update that will require companies to predict credit losses and book higher reserves, if needed, even before cash flow and earnings actually begin to suffer.

The proposal, which applies to all companies but will hit financial firms the hardest, would move U.S. Generally Accepted Accounting Principles away from the current approach that requires companies to book credit losses as they occur or seem imminent, toward a greater reliance on forecasting to book losses long before they actually happen.

As the financial crisis unfolded, shareholders were pummeled with bad news as banks recorded crippling losses from troubled loans, largely after the fact. Investors called on FASB and the International Accounting Standards Board to come up with an accounting approach that would warn them earlier that a bank is headed for trouble based on the make-up of its debt securities and loan portfolio. “The changes we are proposing will address the concerns we have heard that today's GAAP recognizes losses too little too late,” says FASB Chairman Leslie Seidman.

Under existing GAAP, companies do not recognize a loss until it is deemed probable, says Faye Miller, director of the national professional standards group at McGladrey. “Now FASB is removing that probability threshold to say at any given point in time your allowance for credit losses should represent the cash flow you don't expect to collect on an instrument throughout its lifetime,” she says.

That is expected to lead to reserves on even high-quality instruments where losses are not considered probable at the outset, says Miller. The proposal would require companies to give at least some consideration to the possibility of a loss, based on their own historical data as well as expected economic conditions, and book a reserve accordingly.

The new standards could have a dramatic effect on the balance sheets of banks and other financial services firms. Seidman says FASB's analysis of data on the largest financial institutions suggests reserves could rise some 50 percent under the new standard, given current economic conditions.

Nichole Burnap, associate accounting analyst at Credit Suisse, agrees the new rules may make the “day one” loss, or the reserve, larger. “Right now a company may recognize an estimated loss based on what it expects to incur over the next 30 to 90 days,” she says. “This proposal is asking for an estimate over the life of the financial asset.”

The proposal also calls for a single approach to the accounting for all debt instruments, says Miller. “Right now there are five or more models,” she says, “It's different depending on whether it's a debt security or a loan receivable, and whether the assets are originated by the entity or purchased subsequently.”

“The changes we are proposing will address the concerns we have heard that today's GAAP recognizes losses too little too late.”

—Leslie Seidman,

Chairman,

FASB

Beyond Wall Street

That means even operating companies that are not financial institutions will need to take a close look at how the new forward-looking approach might affect their accounting for all of their debt instruments. Burnap believes the effect will differ depending on how companies enter into receivables or other financing arrangements, and how they interpret existing accounting rules around probability. “If they assume performance until formal default, then that would change under this proposal,” she says. “They would have to, on day one, recognize a larger amount of expected losses and update those estimates each quarter.”

Perhaps the trickiest areas for companies to apply the proposed new approach will be in long-term receivables, says Burnap. “Companies will have to estimate expected losses based on their historical experience and forecasts about the future,” she says. “Even if contracts are performing now, there might be some pressure to recognize losses given a downturn in an economic forecast.” Companies would also have to apply the approach to any securities they hold as available for sale to maturity, she says.

CALCULATING EXPECTED CREDIT LOSSES

Below is an excerpt from Credit Suisse's report on FASB's credit loss proposal.

FASB's approach with Calculating Expected Credit Losses sounds pretty straightforward. It would require companies to recognize an allowance for credit losses based on the amount of contractual cash flows that the company does not expect to collect given the company's historical loss experience, current conditions, and management's forecasts about the future. The allowance would then be updated every quarter to reflect changes in the company's

estimates of expected credit losses.

The concept is simple: for many financial assets on the balance sheet, recognize an

allowance for credit losses that reflects the amount of cash that you don't expect to

receive. However, applying that concept can get tricky and will probably vary amongst

companies as the proposal does not dictate how the expected losses should be measured

(whether on a pooled or individual asset basis), what metrics need to be considered in

determining an appropriate reserve amount, or what economic conditions need to be

forecasted. Fortunately FASB provided a number of examples of how a company can

calculate its estimate of expected credit losses, including the one shown in Exhibit 1 which

uses historical loss rates that are consistent with “current economic conditions.”

There is an exception (there's always an exception) for all financial assets that are marked

to market through OCI (i.e., Available for Sale) an allowance for credit losses does not

have to be recognized if fair value is above or equal to cost and expected credit losses are

insignificant.

In case you were curious, net interest income would still be recognized on a contractual

basis where the interest rate is applied to the book value of the asset (before deducting

the allowance for credit losses). As for purchased impaired assets, the discount that's

embedded in the purchase price due to credit losses “should never be recognized in

interest income.” FASB also tries to standardize the meaning of nonaccrual status, so

that when it's not likely that “substantially all of the interest will be received” the loan stops

accruing interest.

Source: Credit Suisse.

That forecasting aspect of the standard is a little troubling to even the financial institutions that are keeping close tabs on the proposal. Banks are quite adept at forecasting a year or perhaps 18 months into the future, says Mike Gullette, vice president of accounting and financial management at the American Bankers Association. However, they're not accustomed to long-range forecasting, especially on long-lived loans or other financial assets.

“We really don't know how to look that far out,” Gullette says. “That leads to a concern about reliability.” Banks and other financial companies are also worried about the volatility that could arise, leading to steeper capital requirements under banking regulatory rules—all stemming from potentially unreliable long-range forecasting.

In the banking sector, time horizons that are comfortable to forecast depend not just on time, but also circumstances, adds Gullette. “If they know there are certain things on the horizon that are of higher risk, bankers will not have any problem saying yes, we should be accruing and reserving for that,” he says. Beyond a few years in the future, however, it's difficult to forecast anything with great certainty, he says.

Despite efforts to converge U.S. and international accounting rules, IASB's proposal, expected in the first quarter of 2013, will take a different approach. IASB plans to propose a “three bucket” model, where instruments would move among buckets depending on when losses were expected. FASB considered adopting that approach, but later dismissed the idea when outreach efforts revealed it to be problematic.

FASB says its interaction with preparers and investors revealed confusion over how items would move among the buckets, but the board says it will study feedback to IASB's proposal. "We continue to cooperate with FASB on this important topic and intend to publish proposals for public comment in Q1 2013 based on a simplified version of the previously agreed approach,” IASB said in a prepared statement.

Peter Bible, a partner at EisnerAmper, says he believes FASB's proposal represents a simpler approach than current GAAP and results in a better measurement. He likes the idea of having a single approach to measure all debt instruments. “If I were a CFO or a controller at a bank, this would make my life easier,” he says.