Peeling away financial reporting issues one layer at a time

Loan Accounting? The Public Can’t Handle the Truth!

Even if the average citizen might have shrugged with resignation over the Senate’s abandonment of gun control legislation, surely she must be in awe of the power of the NRA to bend democratically-elected senators to its will.  On a less-visible front, the FASB and IASB are doing their best imitation of U.S. Senators to appease even more powerful patrons than the NRA — very rich bankers.

The tune the accountants are singing about loan accounting is from the same songbook as the politicians, but played in a different key: inaction must be disguised as well-considered action.   Actually, though, one FASB member has been doing something.  Tom Linsmeier collaborated on a recently-published study* with three professors from Stanford, Michigan and Michigan State:

“Many have argued [principally, the ABA] that financial statements created under an accounting model that measures financial instruments at fair value would not fairly represent a bank’s business model….

We find that leverage measured using the fair values of financial instruments explains significantly more variation in bond yield spreads and bank failure than the other less fair-value-based leverage ratios [under U.S. GAAP or under regulatory rules for “Tier One” capital] in both univariate and multivariate analyses. We also find that the fair value of loans and deposits appear to be the primary sources of incremental explanatory power.” [emphasis added]

Let’s assume, and I have every indication that this is true, that the research yielding the above conclusion is seen to be high quality by business school academia.  If so, how could anyone associated with, or acknowledging its quality, support any other accounting treatment besides some version of current value?

The implied answer of Linsmeier et. al. is in the last paragraph of their paper:

The results of our study should not be used in isolation to suggest that all financial instruments should be recognized and measured at fair value. … There are other costs and benefits [to banks] … that we do not consider. Most notably, our study does not address the potential implications that fair value accounting has on procyclicality or contracting.

Cynical me wonders whether the authors inserted that bit of boilerplate as a favor to Tom Linsmeier: to keep him in the good graces of his overseers at the Financial Accounting Foundation.  I’m glad he collaborated on the study, and I’d like to think that this was the price. All I can say is that when I read it, I began to fantasize about how Jack Nicholson’s Colonel Jessup in A Few Good Men would have responded to my question:

You investors can’t handle the truth! Son, we live in a world that will always have toxic loans, and those loans have to be guarded by bankers whom you have no choice but to trust.  We bankers have a greater responsibility.  Our duty is to apply accounting rules to loans that is assured to spew out numbers that nobody can audit, much less describe.

You want fair value?  We invent terms like “allowance” and “historic interest rate” as the backbone of a life defending our bonuses.  You use them as a punchline.  I don’t have time for you and your questions!  I would rather you just rolled over your CDs and trust the government’s deposit insurance program.  Whatever you do, I don’t give a damn about what information you think you are entitled to!

I doubt the IASB was channeling Colonel Jessup, but the key language in its latest proposal says much the same thing, without actually saying anything at all:

The main proposals would require an entity to recognize expected credit losses … using current estimates of expected shortfalls in cash flows…  An estimate of expected credit losses would always reflect the probability [italics in original] that a credit loss might occur… Accordingly, the proposals would prohibit an entity from estimating expected credit losses solely on the basis of the most likely outcome (that is, the statistical mode).”

So, now we know how a loan may not be measured under IFRS, but that still leaves us pretty much in the dark as to how a loan would be measured under the proposal.  While there is a token laundry list of the items that management must some factor into the measurement, the process evidently passes muster with those Colonel Jessups running  too-big-to-fail banks in Europe; otherwise, it surely wouldn’t have seen the light of day.

Consequently, methinks two things are safe to predict.  First, every bank will have the license to do pretty much as it pleases with its loan loss allowances and revenue recognition, just as it does now; and financial statements won’t be “comparable” — the putative holy grail of convergence being trumpeted by it supporters.

Second,  the next financial crisis will hit like a ton of bricks without fair warning from banks’  financial statements.  For the Colonel Jessups of the rarified world of finance, “Code Red” will once again be the order of the day.

* * * * *

Let me state my earlier question another way: What good reason is there to conduct research comparing fair value measurements to other alternatives, if not to inform accounting standard setters?

Whose job is it to deal with procyclicality, anyway?  Where is that discussed in the Conceptual Framework?

Dr. Linsmeier, would you care to enlighten us?


*Blankespoor, E., Linsmeier, TJ, Petroni, KR and Shakespeare, C.  Fair Value Accounting for Financial Instruments: Does it Improve the Association Between Bank Leverage and Credit Risk?, November 2012. Available at: .

Error correction—An earlier version of this post reported that the co-authors were from Stanford and Michigan.  I erroneously omitted Michigan State.

1 Comment

  1. Reply Kalin April 20, 2013

    Dear Tom,

    Your comment on this paper inspired me to have a thought about the role of accounting science today. I’m short of space now but … What is the form of its existence, the one in your example? Should we draw a line between accounting science and accounting ethics? As the latter have to be in close bond to the results, aim of the science…and do not forget social consequences when dealing with the resultant of application of new Ifrs. Does we have such research?

    Best regards,

    Kalin K.

Leave a Comment