The Accounting Onion

"Thinking Fast and Slow" about Auditing

My writings in this blog have focused on the absence of linkages between financial reporting and the economic concepts like wealth and value creation/destruction. But, accounting is more than applied economics. Individuals acquire accounting information before making decisions; and heuristics and biases affect the quality of those decisions. I spent much of my academic career thinking about these concepts, and Daniel Kahneman's new book, Thinking, Fast and Slow, brings me back to those halcyon days.

In that vein, I would like share with you this autobiographical vignette from Chapter 7, entitled "A Machine for Jumping to Conclusions":

Early in my career as a professor, I graded students' essay exams in the conventional way. I would pick up one test booklet at it time and read all the students' essays in immediate succession, grading them as I went. I would then compute the total and go on to the next student. I eventually noticed that my evaluations of the essays in each booklet were strikingly homogeneous. I began to suspect that my grading exhibited a halo effect [defined as the tendency to like (or dislike) everything about a person – including things you have not observed], and that the first question I scored had a disproportionate effect on the overall grade. The mechanism was simple: if I had given a high score to the first essay, I gave the student the benefit of the doubt whenever I encountered a vague or ambiguous statement later on. … I told the students that the two essays had equal weight, but that was not true: the first one had a much greater impact on the final grade than the second. This was unacceptable.

I adopted a new procedure. Instead of reading the booklets in sequence, I read and scored all the students' answers to the first question, then went on to the next one. I made sure to write all the scores on the inside back page of the booklet so that I would not be biased (even unconsciously) when I read the second essay. Soon after switching to the new method, I made a disconcerting observation: my confidence in my grading was now much lower than it had been. The reason was that I frequently experienced a discomfort that was new to me. When I was disappointed with a student's second essay and went into the back page of the booklet to enter a poor grade, I occasionally discovered that I had given a top grade to the same student's first essay. I also noted that I was tempted to reduce the discrepancy by changing the grade that I had not yet written down, and found it hard to follow the simple rule of never yielding to that temptation. …

… The consistency I had enjoyed earlier was spurious; it produced a feeling of cognitive ease and … [I] … was happy to lazily accept the final grade. By allowing myself to be strongly influenced by the first question in evaluating subsequent ones, I spared myself decisions of finding the same student doing very well on some questions and badly on others. The uncomfortable inconsistency that was revealed when I switched to the new procedure was real: it reflected both the inadequacy of any single question as a measure of what the student is and the unreliability of my own grading.

The procedure I adopted to tame the halo effect conforms to a general principle: decorrelate error! [pp. 83-84; italics supplied by me]

The lessons that any student or professor can take from the story are obvious and compelling; and grading happens in auditing, too – a lot.

Auditors Have Feelings, Too

The standards that form the foundation for the practice of auditing were laid down long before the study of heuristics and biases in decision making gained traction among psychologists, and eventually economists. But, while a revolution in psychology and behavioral economics has been taking place, the key driver of audit reliability shifted from objective verification to subjective "grading" of judgments made by management when preparing financial statements.

For example, auditor term limits can be regarded by behaviorists as an effort to mitigate serial errors as the auditor accumulates too many positive vibes from the client. The resistance to any any form of auditor firm term limits by auditing firms is effectively either: a denial that a confirmatory bias – as if auditors were inoculated from this human fallability; or perhaps like Kahneman's grading story, a statement that auditors have identified the problem and figured out a solution – much as Kahneman dealt with his exam grading issues.

But, neither argument seems valid. Delving a little deeper into auditing procedures, auditors 'grade' numerous judgments made by management. Yet, there are no explicit policies that I am aware of that have been instituted for decorrelating those grades. According to the confirmatory bias, order matters a great deal. Maybe, auditors should be making and documenting their judgments and estimates independently of management (and even without looking at prior years' work papers), before deciding whether or not a judgment by management "appears reasonable."

Or, perhaps, auditors will maintain that supervisory procedures within the firm ensure that judgments are unbiased and independent. But, that doesn't appear to be the case. Audit managers 'grade' numerous tasks performed by audit staff members, and it seems to me that little or nothing is much different about the way it is done than the way Kahneman performed his grading – until he realized something was amiss.

In the current environment, can it be said that standards governing "independence," "technical competence" and "sufficient evidence" are enough to produce reliable audits? The weight of evidence brought to bear by psychologists suggest that auditing judgments may not be reliable without instituting explicit procedures for decorrelating errors in auditors' judgments.

Kahneman's story is memorable for his self insight in recognizing a halo effect was influencing his own judgments, the creativity to devise a successful strategy for mitigating bias, and the discipline to implement a costly strategy. If only the accounting profession would recognize its duty to the public to strive to act in a similar fashion, we would all be wealthier and wiser.

FASB Alumnus Trashes GAAP (and IFRS)

When I started my blog almost exactly 3 years ago now, my opening strategy was to work down a rather limited personal inventory of pet peeves. I wanted to stay above the fray, but that proved impossible. I soon became hopelessly drawn to commenting on the helter-skelter stampede of recent developments driven by the Big Four's IFRS putsch and the undeniable role that lousy accounting standards played in the financial debacles of 2007 and beyond. 

But, who really cares what I think?  I suspect that the folks being paid the big bucks to make the tough calls on accounting standards don't pay a lot of attention to to the likes of Tom Whatshisname, even were I to announce that the sky is falling. But, I don't take it personally. Over the past 40 years, any PhD not drawing a salary from the Big Four has been viewed with more suspicion than respect by the standard setting establishment.

I mention all of this now, because there is a new voice, whose credibility and qualifications cannot be so easily dismissed.  That voice belongs to FASB alumnus David Mosso, who has written an 80-page monograph entitled Early Warning and Quick Response: Accounting in the Twenty-First Century).  If you don't want to believe me, take it from him: GAAP is broken.

Mosso is not just some average pedant. He has pretty much done and seen it all when it comes to accounting standards setting, most significantly having spent nearly twenty years at the FASB, and having been a board member for ten of them:

"I spent three decades in helping to create the accounting mess called generally accepted accounting principles, or GAAP. This book is an attempt to make amends." (page 1, italics supplied)

"Eighty years of tinkering [with a broken accounting model] has not done the job. … In the case of business failures, the company's auditors are usually the scapegoats, but I suspect that accounting standards are often more, or at least equally, at fault." (page 78, italics supplied)

I'm not trying to promote Mosso to oracle status, but coming from him—as opposed to whatshisnames like me—this should been seen more as an indictment than merely opinion. (By the way: just to be sure he wasn't just trashing GAAP, I asked David via email what he thought of GAAP/IFRS convergence. He replied that it was like the collision of two garbage trucks.)

Mosso's book has three principle themes:

1. Financial reporting is a vital function; however, the politicization of standard setting has been far more harmful than beneficial. Words like "junk," "trash," and "garbage" permeate the references to current GAAP.
2. GAAP must be fundamentally changed. Mosso's proposed system of accounting rules consists of little more than a few principles, which together comprise his "Wealth Measurement Model" (WMM). The gist of that model is to recognize "all" the assets and liabilities of an entity and to measure them at fair value, which apparently out of convenience is specified to be in accordance with existing FAS 157. Mosso claims that the comprehensive nature of WMM makes it capable of generating timely signals of looming problems to both investors and policy makers.
3. Upon implementation of the Wealth Measurement Model, subsequent accounting standard setting could be limited to a sort of rapid-response version of the Emerging Issues Task Force for addressing implementation issues arising out of innovations in business practices and new technologies.

I don't have much to say about the third theme, because it's not a topic that I spend much time thinking about. I do acknowledge, however, that the current standard-setting system is seriously dysfunctional. And, I can't resist pointing to Mosso's contention that accounting standards should be run by academics, because "…accounting professors are the only group with the untainted standing to represent the public interest." Read, all ye good people, what this brilliant and eloquent man has thusly writ!

As to the pile of trash that GAAP has come to be (#2, above), I licked my chops when in the first chapter Mosso broadly hinted that his readers would receive the down low skinny on those back room machinations, which led to the accounting standards to bring him the most shame. However, there was absolutely none of that in the book. Mosso's only specific mea culpa was this tiny little snippet:

"When standard setting was formalized, standard setters seem to have taken the existing equity classification of [preferred stock, stock purchase warrants and stock compensation options] for granted without much scrutiny. I acknowledge that I did not challenge the conventional classifications when the conceptual framework was under development."

To his significant credit, however, Mosso may be the first real insider to broach the topic of standard setting dysfunctionality so comprehensively, even though I feel more than a little let down by his refusal to dish. Since Mosso is asking his readers to consider proposed solutions from an admitted creator of the problems themselves, I feel entitled to more of something like this:

"We wrongly promulgated X (and I voted for it) because of pressure from Y. The resulting harm was Z1, Z2 and Z3.

I gag every time I think of the quantities of toxic waste that my innocent son and his babe-in-the-woods peers are force-fed from their 1000+-page "intermediate accounting" textbooks. It would be nice if at least one of the authors of the master cookbook strongly advised students that many grains of salt should be added before consuming.

The Wealth Measurement Model

There is a great deal to admire about the WMM. In particular, I strongly support limiting owners' equity to basic ownership interests, and throwing goodwill and deferred taxes onto the trash heap.

But, I do have two important concerns; in regard to how Mosso would measure assets and liabilities, I believe he is not being ambitious enough; and in regard to which assets and liabilities Mosso would recognize, he is far too ambitious.

The Measurement Issues—Although Mosso states that assets and liabilities should be measured at their fair values, I suspect that deep down, he realizes that replacement cost is the attribute of an asset that corresponds to "wealth." I draw my inference of Mosso's true views from Chapter 5, wherein he discusses the insights to be derived from comparisons of the market value of a firm's equity to the replacement cost of its net assets – i.e., a variant of Tobin's Q ratio. Mosso's analysis clearly indicates his awareness that "wealth" embodied in productive assets like machinery is not correctly measured by the current amount for which the machinery could be sold, but for how much it would cost to replace its productive capacity. So, I find it slightly ironic that Mosso condemns the politics of standard setting, but himself seeks a political compromise with the hope of persuading proponents of fair value to embrace his WMM.

The Recognition Issues—These are an even bigger concern to me than the choice of fair value over replacement cost. Mosso contends that accounting standards can be written such that "all" assets and liabilities could be found on the balance sheet. I'll admit that the possibility is tantalizing, because it would eliminate off-balance sheet financing, and it would also transform what is now merely a "balance sheet" to a true "statement of financial position." Unfortunately, Mosso's vision is unrealistic; and here's a couple of examples to illustrate why:

• A competitor's misfortune provides an opening for an entity to introduce a new product. Such an opportunity is what is known as a "real option." An accounting definition of "asset" that would comprehend those circumstances would be far too broad to be of practical use.
• Without going into too much detail here, Mosso would require the recognition of "constructive" liabilities (even though that's not the term he uses), and for reasons I explain here, I would not support that.

Therefore, the set of assets and liabilities that would be eligible for recognition needs to be bounded in a manner that can be consistently applied and audited. For me, legal rights and obligations form natural and convenient boundaries.  It has been my position that assets and liabilities recognized should be limited to legal rights of ownership, rights of use, and rights to receive assets in the future.

A Final Thought

When a member of the guild of royal sausage makers issues a health advisory, we need to find something else to eat. David Mosso's new recipe does not perfectly suit my tastes, but it's definitely restaurant quality.

A Tale of Shorn Sheep, Slaughtered Pigs and Keystone Cops

No One Would Listen: A True Financial Thriller is Harry Markopolos's own story of a futile eight-year quest to expose the $65 billion Madoff fraud. As an SEC alum myself, I was not looking forward to reading a blow-by-blow description of gross incompetence at my alma mater as it were.  I stared at that brown and black book jacket sitting on my desk much as I stared at one memorable dinner plate from my childhood, holding a generous dollop of slimy, similarly colored mushrooms that seemed to dare me to return its limpid gaze.

And, where would the "thrill" come from as the scheme sans apparent motive abruptly came to an end? Madoff himself had effectively written the anticlimax when he turned himself in to the authorities before he would suffer the indignity of being properly caught in the act. I did know that I would be reading about grave mistakes having been made, from which everyone with more than a passing interest in U.S. capital markets would have something to learn. Thusly did I know I would be compelled to read No One Would Listen, much as I knew the two-headed power of our home well enought to predict with absolute certainty that I would be eating those horrible looking mushrooms before TV time.

The mushrooms actually tasted pretty good (my Mom is still a wonderful cook), and much to my surprise, No One Would Listen was might tasty, too. The red meat was, as expected, the SEC's incredible ineptitude; a case of truth stranger than fiction served blood rare. However, the role of "feeder funds" was more complex than simple meat and potatoes aiders and abettors; although to be sure, there was enought of that.

Some feeder funds were managed by third- and fourth-generation rich kids who were as lost as babes in the woods when it came to rigorous financial analysis. All they really cared about was that Madoff had that je ne sais quoi they thought they could trust; to give the blue bloods the excess returns that they were entitled to by their place in the natural order.

A more insidious group of feeder funds portrayed themselves to investors as due diligence mavens, when in actuality they (along with the audit firms involved) did little more than go through the motions. Like the blue bloods, they merely accepted everything Madoff supplied them without any real verification.

All Ponzi schemes raise the ineffable question as to how otherwise intelligent folks turn off their brains when offered the prospect of money without work or risk. But, too little attention seems to be paid to the thinking and actions of those who confidently recognized a true-good-to-be-true offer for what it is. Markopolos writes that for him, one of the most disturbing aspects of the Madoff affair was that hundreds of analysts like him must have been aware of the mass fleecing going on. Yet, only Markopolos and his small group of compatriots apparently tried to do anything about it. I don't know who should feel worse: the suckers who were taken in by Madoff; or those who knew full well what was under foot, yet chose to stand idly by. Ah well, more about that later, when I discuss incentives for whistleblowers.

Can The SEC Do Better?

Markopolos has a number of insightful suggestions for reforming the SEC, and even if you don't care to read the entire book, the lengthy epilogue is worth reading to prime your own pump. I'm going to comment on two of the areas that primed mine: radical restructuring of the SEC staff, and changes to incentives for both staff members and whistleblowers.

With respect to reconfiguring the staff, I agree that there are far too many lawyers, but replacing them with experienced finance professionals, as Markopolos proposes, is a lot harder than it sounds. The problem is that the government can't pay competitive wages to competent and experienced professionals unless it can supplement a paycheck with non-pecuniary benefits, e.g., on-the-job education and 'relationship building.'

SEC lawyers and accountants accept below-market pay sometimes because of lifestyle considerations, but much more often they do so because it creates other options for them. The private sector values the technical and institutional knowledge acquired while at the SEC, and it must be said their tight relationships with SEC staffers doesn't hurt. Such is evidently not the case with financial economists. The best and the brightest are hired straight from school onto Wall Street and soon, if not immediately, command salaries and bonus prospects that are many times greater than SEC staffers of similar vintage. Moreover, SEC experience won't make them any better, brighter or more marketable. The unfortunate reality is that the SEC will have to obtain its financial expertise elsewhere—perhaps through consulting arrangements under which it can pay market rates.

Regarding incentive bonuses to staff, Markopolos suggests that the SEC create compensation bonus pools from collected fines. With all due respect, I think that's an even bigger stretch. If an SEC staffer, who has been working full-time and full-out on a Madoff-size case unexpectedly discovers exculpatory evidence, how will that be handled? If the staffer's family budget depends on getting a bonus out of the case, will she try to bury the evidence? While I would like my government to be highly effective enforcers of the law, I fear much more a government where there is even a slight appearance that law enforcement officers have a personal financial incentive to prosecute and convict the innocent.

Where Markopolos and I agree completely is on the cryingly obvious need to give greater incentives to financial fraud whistleblowers. One of the things I find most pleasing about the securities laws governing financial reporting is that the SEC's enforcement capacity is augmented by private litigants seeking to recover damages and penalties from companies and individuals that violated the securities laws in some fashion or another. Thereby, shareholders and the lawyers who retain them (sometimes in reverse order) are given strong economic incentives to police violations of the securities laws. To be sure, some defendants will be harried by frivolous claims, but that cost seems to pale with the benefits.

So, why not extend this logic to violations by investment advisers and funds? Let's provide an incentive for individuals, be they employees or others, to come forward with direct evidence of financial fraud, and reward them generously when a case is prosecuted successfully. Markopolos points out that a bare bones program is already in place at the SEC for insider trading cases, but somehow the SEC has avoided shelling out much money to its sources. The IRS and other government agencies have much more developed and successful whistleblower programs that the SEC can emulate and/or adapt.

Markopolos also points out that the processes by which whistleblower complaints are vetted needs to be made more efficient, and I would take his suggestions even further. I would require that whistleblowers make their concerns known via a formal, non-public filing process (to be unsealed after a period of time). The instructions to a whistleblower complaint form would set forth the minimum level of specificity of allegations and strength of supporting evidence in order for the SEC to even consider taking further action. The whistleblower could also choose to remain anonymous by engaging a lawyer to handle the filing, and there might be a small filing fee, say $1,000. It would also be specifically established that false filings of whistleblower complaints would themselves be subject to sanctions.

To effectively process the whistleblower allegations and to fairly distribute awards, the SEC would establish and publish its procedures for, among other things, assigning files to teams of enforcement staff members who possess the technical competence to deal with the issues and, if necessary, engage outside consultants. Just as in the Madoff debacle, the inspector general's office would investigate the valid complaints that fell through the cracks, but it would also continuously monitor the handling of all whistleblower filings.

Fiddling with Fair Value as Rome Burns

If you took a look at the House Bill that went down in flames today, you might have caught Congress fiddling with GAAP as Rome burned. See Section 132 of the bill proposed to empower the SEC to suspend application of FAS 157, Fair Value Measurements at any time, and for any company. Doesn't Congress know that the SEC already has the power to make and break GAAP?

See also Section 133, which directs the SEC to submit a study to Congress on FAS 157 as it applies to financial institutions, and preferably to make FAS 157 and the FASB a scapegoat for bank failures. If Congress were serious about studying accounting's role in how our economy's "fundamentals" have been heading due south, they shouldn't be looking at FAS 157, because it only addresses "how," and not "when" fair value is to be measured. "When" is, of course, the real question, which means that Congress should be questioning instead the accounting standards that the members' banker friends actually do favor: FAS 115 (marketable securities), FAS 140 (securitizations), and FAS 114 (loans). You know, the ones with the loopholes. Such is politics.

The Blog Must Go On

I feel somewhat abashed to be discussing the niceties of financial reporting in the wake of current events, but if Congress can do it, why can't I? My topic for today is to follow up my previous post on current value accounting, discussing whether the current values to be measured should be replacement costs or fair values. Specifically, I want to share with you my reactions to Federal Reserve Chairman Ben Bernanke's view of "hold-to-maturity" versus "market value." Two other comments on my advocating for replacement cost accounting (and a history lesson from Bob Jensen) also got me thinking a little bit more about what I wrote.

First, Bernanke gave congressional testimony arguing that two prices existed for troubled loans: a "hold-to-maturity" price and a current market price. At first blush, this sounds something like the difference between replacement cost and fair value, but it isn't. This is important to me, because I think that some people (for example, see the comment of George Weintraub on my previous post) may erroneously associate replacement cost measurements with 'value in use,' which I believe would be the FASB's nomenclature for what Bernanke has called "held-to-maturity." There can be no question that value in use invariably presents one with a reliability problem due to the subjectivity of estimating future cash flows and discount rates (i.e., 'mark-to-model'), but that is not necessarily the case with replacement costs.

Moreover, I'm not sure that Bernanke, whose intellectual capacity I respect and admire, is making much sense here (and I suspect he knows that). If we are talking exclusively about financial assets and a global marketplace, how could it be that the market price understates the present value of holding a loan until maturity? If the subject were changed to producing or consumption goods, one could expect a difference between value-in-use and value–in-sale. However in the case of financial assets, such a presumption is perhaps a step too far. Teaming up with Treasury Secretary Paulson to sell such a dubious notion to the public may in the long run affect Bernanke's reputation for independence, just as Greenspan's reputation was damaged by his flip-flop on Bush's tax cuts.

Moving on to replacement costs and financial assets, the 'bid-ask' spread captures the difference between replacement cost and fair value. The 'bid' price is fair value, because it is the amount that someone is willing to pay to purchase the asset from you. The "ask" price (a higher price) is the replacement cost, because it is the amount that a market participant who already has the asset is willing to sell it to people who want it. But, even if there is no formal market maker to provide both bid and ask prices, the concept still applies for the purpose of comparing replacement cost and fair value accounting. For example, and moving from financial to real assets (and I owe this one to Bob Jensen), bluebook prices for vehicles are the functional equivalent of average ask prices by used-car dealers.

Second, Jim Noel, my longtime friend and sounding board, expressed his doubts after reading my previous post that replacement costs were an appropriate basis of measurement for bank balance sheets. That's because the single most important question to ask about a bank, as the financial crisis painfully illustrates, is liquidity. Indeed, the strongest argument for fair value is that it is the gold standard for measuring liquidity.

My response to Jim is to revert to core principles (at least mine) of financial reporting: investors primarily seek information about wealth and changes in wealth. In normal circumstances, liquidity is of secondary importance. As a case in point, maximum liquidity leaves one with just cash and no wealth producing assets.

Granted, there are circumstances when liquidity is paramount, but that should not mean that principles-based accounting needs to be either replacement costs all the time, or fair value all the time. For example, when ARB 43 requires a departure from historic cost to measure inventory, the general basis of measurement is replacement cost. However, replacement cost valuations are capped by 'net realizable value' and floored by 'net realizable value less incremental costs to sell.' A second, and more general example, is the required switch to liquidation values when an entity's ability to continue as a going concern becomes doubtful to a sufficient degree.

The Fair Value Debate Needs Some Rules

My general sense is that some order is needed to make sense of the challenge to fair value. Whatever your position is on the issues, I would like to suggest that you think about providing answers to the following three questions—in the order listed:

1. Which assets and liabilities should be measured at current value?
2. Which 'attribute' of an asset or liability should be measured—i.e., replacement cost, fair value, historic cost, present value of future cash flows?
3. What rules would be needed to assure that the attribute chosen is reliably measured?

Which assets and liabilities should be measured at current value -- So much of the balance sheet and earnings volatility that the fair value naysayers are complaining about is brought on by our 'multi-attribute model', where some assets and liabilities are measured in terms of actual past cash outflows and others are measured in terms of estimated future cash inflows. The status quo is simply not stable, so we have to decide whether to move backwards to historic cost or forward to a single set of current values.

Of course, I believe that means we have to finally bring ourselves to measure liabilities as well as assets at their current values. This shouldn't be so hard. A simple economic principle that I pointed out in a previous post is that all of the liabilities of an entity are perforce assets of other entities. If we can measure these financial relationships when they are on the lender's balance sheet at current values, then why in the name of symmetry can't we use that valuation for borrowers? Oy, it makes me crazy.

Which attribute should be measured – Some say fair value, some say present value of future cash flows, some are sticking with historic cost, and I say replacement cost. Whatever your druthers, reported earnings will not pass a reasonableness test until we stop being content with adding apples and oranges on the balance sheet.

How to reliably measure the current value attribute chosen? -- I hope George W. (the Skeptical Texas CPA) and others may now see that the difference between replacement cost and fair value is really not a difference between market quotes and mark-to-model. Rather, it's how best to measure what it is you think ought to be measured. Maybe fair value and replacement cost measurements will require mark-to-model every now and then, but for reasons I pointed out in a previous post, replacement cost measurement will require less of it.