The Accounting Onion

"Thinking Fast and Slow" about Auditing

2012-01-23T23:33:58-07:00

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.

What I Mean by "Replacement Cost" is not Literally Replacement Cost

2012-01-17T00:55:41-07:00

I get a lot of questions – and yes, some criticism – of my support for measurement of assets and liabilities at "replacement cost" (with comparative periods adjusted for the effects of inflation). My views are derived in large part from three conclusions. First, replacement cost measures are the only possible way for accounting to reflect wealth invested by shareholders in an enterprise; and consequently, changes in invested wealth. In a post and working paper that are already more than four years old, I demonstrated the truth of this proposition using a simple example.

The second conclusion is derived from the first: namely, that principles-based accounting must be primarily concerned with measurement of wealth and changes in wealth. I can think of no other principle that is more fundamental to accounting – and no way to be faithful to that principle without replacement costs. Financial reporting under complex and inconsistent rules, without principles to bind them, has been a losing proposition.

Third, I go back to the maxim that accounting affects decision making; or, "what gets measured gets done." If management is supposed to maximize wealth, then accounting must endeavor to measure the effects of management decisions on wealth. I have tried to say this in different ways over the years, but Walter Schuetze nailed it for me this weekend over lunch in San Antonio:

"Accounting cannot stop management from doing dumb things. But, we can report it as soon as possible so the world can see the results."

The FASB has succeeded in ignoring this self-evident maxim throughout its entire history. In doing so, it has enabled a national tragedy at many levels: obscene levels of compensation awarded to management, impairment of U.S. competitiveness, mass disruptions of financial markets, shareholder value destruction – not to mention the chaos and misery inflicted on honest, hardworking and capable individuals.

"Replacement Cost" in the Real World

However, I don't deny that there are a lot of implementation issues that should be addressed. Perhaps the most persistent and fundamental criticism of replacement cost measures is the lack of relevance when there is little or no prospect that the existing assets will be replaced with identical, or even similar, assets. A commonly invoked scenario is when an existing asset lacks some of the desirable characteristics of later versions that are available to be purchased – say, an aging airliner.

The conceptualization of "replacement cost" that I advocate can handle this situation simply and reasonably. To begin, let's set aside the airliner for a moment and consider inventory held for sale. For such assets, there is already a replacement-cost-based accounting treatment in U.S. GAAP (though not in IFRS), which has stood without challenge for more than 60 years.

Every accounting major will be familiar with U.S. GAAP measurement of inventories at its lower of cost or "market." For the sake of compactness, but without loss of generality, here is the essence of the protocol:

"A departure from the cost basis of pricing the inventory is required when the utility of the goods is no longer as great as their cost … the difference shall be recognized as a loss of the current period.

… [market] provides a practical means of measuring utility.

As a general guide, utility is indicated primarily by the current cost of replacement of the goods as they would be obtained by purchase or reproduction. … [However] … replacement or reproduction prices would not be appropriate as a measure of utility when the estimated sales value, reduced by the costs of completion and disposal, is lower, in which case the realizable value so determined more appropriately measures utility." [Accounting Standards Codification 330-10-35-1 through 4 (emphasis supplied)]

In essence, a measure of the utility of inventory (i.e., wealth invested) should be capped by the amount it could be sold for, less incremental costs of disposal. This is because it is hard to imagine a situation where replacement would be economically justified if, once replaced, the inventory could not be sold for more than its replacement cost.

Having established the concept in existing US GAAP, let's turn back to the problem of the aging airplane. Following similar logic as for inventory, the owner should first determine whether there is an active market for airplanes of similar vintage and features. If so, then the replacement cost can be objectively based on recent transactions. But if there is no active market, then my operationalization of "replacement cost" would cap the reported amount at what could be received from sale (under current conditions), less estimated out-of-pocket costs.

Thus, there is no need to determine whether inventory is "obsolete", or even how management intends to use and/or dispose of the asset. Moreover, the resulting measure of "market," "current cost," "replacement cost," or whatever it may be called happens to be conservative because it is likely that the utility of the investment to shareholder interests is greater. This logic is buttressed by the fact that the airplane is still owned as of the balance sheet date, providing some indication that greater utility was expected from use rather than sale.

Simple Solutions to Contentious Problems of the Day

I know we are a long way from applying replacement cost or market to airplanes, but I want to conclude by pointing out that the concept of replacement cost provides for very different solutions to a number of current controversial accounting topics. Below are two very different examples.

Leases – Speaking of airplanes, many operators access them via non-transferable leases. When the right to use an asset is not transferable, "fair value" (i.e., exit price) makes no sense whatsoever; and neither does the crude and arbitrary calculations for leased assets and liabilities that the FASB and IASB are currently proposing.

The only relevant measurement attribute of non-transferable rights and obligations arising from lease contracts is replacement cost. In fact, if you believe that non-transferable rights should be recognized as assets, I can think of no better example to show why replacement cost is preferable to fair value as a comprehensive basis of measurement.

Going concern evaluations – Just yesterday, after years of rumination, the FASB announced that it has decided it will not issue a standard that would require management to disclose its going concern evaluation (instead of auditors), even though IFRS requires it of management. If all assets and liabilities were measured under the concept of replacement cost that I have described here, this would be a non issue.

A Practical Way Forward

Although I am for comprehensive use of "replacement cost" measurements, I don't want to be seen as rigidly ideological. As the forgoing example illustrates, there are plenty of ways to significantly improve GAAP without full adoption of replacement cost measurements. By far, the most important are those inflation adjustments I mentioned in passing at the top of this posting.

In that regard, I'll leave you with two closing thoughts: (1) a bank has not earned a profit on a loan of $10,000 when it collects a total of $10,001 over the five-year term; and (2) inflation-adjusted historic costs do a pretty good job of surrogating for actual replacement costs in most circumstances.