Loan Impairment: A Made-Up Number By Any Other Name …
According to the American Bankers Association’s August 5th letter to Treasury Secretary Timothy Geithner, the banking industry will be fundamentally changed if the FASB were to require banks to account for its loans at fair value. Everybody who thinks that banking practices should be fundamentally changed, please raise your hand.
“The FASB’s proposal will … decrease the ability of investors, regulators, and bank customers to undertstand the business of banking, due to the focus on estimates of the market’s perception of value rather than the cash value to the bank.” (italics supplied)
Let’s break this passage from the ABA’s letter into two parts: what the ABA appears to be for—reporting loans at the “cash value to the bank;” and what they are against—a “focus on estimates of the market’s perception of value.”
“Cash Value to the Bank” – Old Rotgut in New Bottles
My inspiration, such as it is, for this post came from a discussion of the ABA letter with a friend (who must remain anonymous). My friend was first to ask what the phrase “cash value to the bank” was intended to mean. As the letter offered no hint of possible alternatives to the FASB exposure draft that the ABA would support, we initially surmised that the phrase in question was the ABA’s oblique way of stating that it actually wanted no change to existing GAAP, without actually uttering those preposterous words — for in these trying times, it is hard to say with a straight face that GAAP’s loan impairment rules are anything but utterly disconnected from reality.
So, we asked the ABA; and, we were right. Here’s the gist of my email correspondence with Donna Fisher, ABA’s senior vice president of tax, accounting and financial management.
My Question: “Does the term ‘cash value to the bank’ refer to existing U.S. GAAP as it applies to a bank’s loan portfolio?”
Ms. Fisher’s Response: “Yes, we are referring to current accounting, in which the amount of principal to be received is the amount that is reported. As in current GAAP, this would not include interest income, but would include fees and discounts. Loan impairment would continue to be shown as a contra-asset on the face of the balance sheet.” (italics supplied)
I suppose there is still some uncertainty as to the intended meaning of the term in question. Is Ms. Fisher referring only to the carrying amount before or net of impairment charges? Methinks she must be referring to carrying amounts for loans net of impairment charges; for if she were not, then the ABA shouldn’t object to reporting fair value as the net of the principal amount (taking into account fees and discounts) and an adjustment to bring the principal amount to the fair value of the loan. I’m 100% certain that the ABA would not be happy with that!
I think I’m on pretty solid ground when I state that most everyone except for bankers and their auditors have eventually come to realize that extant accounting for loans is a dog’s breakfast of wishy-washy criteria and made-up numbers. But, I haven’t reached my self-imposed 1000-word quota yet, so let’s review the lowlights of what investors and bank regulators are currently being fed.
Residential mortgages—The accounting standards for “large groups of smaller-balance homogeneous loans,” including residential mortgages and credit card receivables are about as squishy as accounting can be. All banks will perforce make some accrual for defaults, but GAAP contains no guidance whatsoever for measuring the accrual, other than it must be capable of being measured reliably—whatever that means, and especially since we are told nothing about what the net loan amount is supposed to portray.
The bottom line for loans evaluated on a portfolio basis is that GAAP is an open invitation to bankers to make up a number. Even before the financial crisis, bankers were exploiting with virtually impunity these gaping gaps in GAAP via what former SEC Chairman Arthur Levitt famously referred to as “cookie jar” or “rainy day” reserves. I recall that the SEC staff in 1999, under Levitt, questioned specific banks about loan reserves that were as high as 8% during periods when actual loan loss experience was under 1%. Golly, I wish we could have those days back again, but the point is to show that even after being chastened by the S&L crisis, many bankers were defiantly back to their old accounting tricks, and with nary a peep from their auditors.
Loans identified for individual evaluation—The protocol for recognition and measurement of impairment goes like this:
- Recognize an impairment when it becomes “probable” (helpfully defined as “likely to occur”) that a creditor will be unable to collect all amounts due according to the contractual terms of the loan agreement.
- The amount of the impairment is to be measured as the difference between the carrying amount of the loan and the present value of the expected cash flows (as estimated by management, of course!) to be received from the borrower.
- The discount rate used for making the present value calculation is the loan’s original yield to maturity (referred to as its “effective rate,” in order to avoid explicit mention of the apples-multiplied-by-oranges arithmetic).
First, the definition of “probable” is as clear as mud, and intentionally written so that each bank can have its own special way for determining when that “probable” threshold has been crossed. In practice, “likely to occur” is interpreted to mean significantly more likely than not, and that by itself should speak volumes. Why must investors wait for the bad stuff to be just inches from hitting the fan? By the time investors learn about collectibility problems, there is no getting out of their way.
Second, I probably don’t have to explain to anyone how ludicrous it is to discount lower expectations of cash flows with the originally-expected rate of return on investment. Can this be what the ABA refers to as “cash value to the bank”? Evidently so.
I’m sorry to have to have to inform the ABA that a made-up number by any other name is still a made-up number. I’m dying to reproduce a Dilbert cartoon here, but with copyright laws what they are, I had better not. So, here’s the gist of what those scruffy mutts were saying:
Boss: Your report is nothing but a bunch of made-up numbers!
Subordinate: I know, but studies have proven that made-up numbers are better than real ones.
Boss: How many studies say that?
Subordinate: 58.
(Thanks to Bob Jensen for bringing the cartoon to my attention.)
Third, the requirement in GAAP to consider most loans individually plays even further havoc with the specious “probable” threshold for impairment recognition. Let’s say, for example, a bank holds 10 commercial real estate loans, and due to the occurrence of some event after the loan was made, the probability of default for each loan increased from 1% to 5%. Even though the probability that at least one of the loans will default in the future is roughly 60% (= 1 – .9510) no impairment charge will be recognized. I surmise that only bankers and auditors think that makes for “fair presentation.”
Don’t Trust Markets: Only Real Bankers Know How to Value a Loan
The ABA’s position is that when it comes to accurate estimates of a loan’s value to the bank itself, investors and regulators should rely on ABA members to make those valuations. This we should accept even though two recent banking crises and a cursory reading of Freakonomics tell us otherwise in no uncertain terms.
I certainly don’t want to belabor the obvious here, but what is there about a banker’s character that makes them different from the rest of us? Yes, they are the ones on the ground working with the entire loan portfolio, but asking them to grade their portfolio is like asking a college student to grade her own work. Surely, we have progressed far beyond the naive view of bankers and their auditors as belonging to some sort of priesthood whose solemn vows place them a cut above the rest of us mere mortals. If we are to accept the ABA’s premise that market “perceptions” of the value of financial contracts are structurally flawed, then the problems with our market-based economy run much deeper than deciding how to make the debits and credits.
It’s one thing for the ABA to be against fair value accounting, but it’s quite another to support the status quo. For that, the ABA decided to invent a high-minded sounding term of art that has no actual meaning except for the dog’s breakfast that is current GAAP for loan impairments. Their motivation is to hide their support for continuing to allow bankers to pretty much decide on their own what made-up numbers they deign to report to shareholders.
And, wherefrom comes the basis for presuming that a banker can divine value more accurately than a market of buyers and sellers (or a neutral valuation expert) possessing essentially the same information that the banker has? Certainly, recent history provides no basis. A huge factor that brought about the financial crisis was that the Big Five investment bankers (supposedly the smartest of the smart), stupidly ate too much of each other’s cooking—those overrated tranches of securitized subprime mortgage loans. They all could have fallen on themselves like victims of a Ponzi scheme of their own making had the taxpayers not picked up a large portion of the tab.
Now, the ABA wants us to believe that the old recipe their members followed for cooking their own books is still tried and true, just because they have given it a shiny new name – ‘cash value to bank.’
The name I have for that is chutzpah.