We are long-time critics of generally accepted accounting principles' historical cost tradition in financial reporting. We argue for providing market values to give users the up-to-date information they need to make rational investment and credit decisions.We have had some fun in the past describing the limitations in cost-based GAAP by offering other acronyms: PEAP (Politically Expedient Accounting Principles), WYWAP (Whatever You Want Accounting Principles) and POOP (Pitifully Old and Obsolete Principles). Despite the humorous undertone, our ultimate point is serious: The current financial reporting system needs to provide better support for the capital markets. After all, they are the foundation of our economy, and work best when market participants are quickly and fully informed.
LAME ACCOUNTING
With this column, we add another acronym to our list of GAAP alternatives. It's LAME, short for Legally Adopted Measurement Errors. LAME reflects another problem associated with the profession's traditional (but unjustified) reliance on historical costs.
When GAAP reports don't provide useful information, market participants are obviously hurt, but so too are third parties like judges who must deal with accounting-related litigation. Specifically, many court cases require gains and losses to be measured, and errors are bound to happen if judges try to copy accountants who haven't done a good job of measuring economic reality.
DAMAGES OR NO DAMAGES?
Our wake-up call for this situation was a recent decision in a class-action suit seeking damages on behalf of investors in Organogenesis, a medical products company whose share price plummeted leading up to its bankruptcy in 2002. The plaintiffs' filings allege that management's deficient and deceptive reporting left investors in the dark about the company's bleak future prospects well before the bankruptcy occurred.
(As an aside, the plaintiffs haven't argued that reporting historical costs for Organogenesis' assets and liabilities grossly impaired their ability to assess the company's future accurately. We think they should have, because much of what they wanted to know, but didn't, would have been revealed by reporting contemporary market-based measures.)
What troubles us about the case was the District Court's reliance on historical costs in deciding whether a particular plaintiff suffered a loss as a result of management's alleged mischief. After a series of multiple purchase and sale transactions, the plaintiff held a lot of shares of Organogenesis during the period leading up to bankruptcy. It was the court's task to determine how much the plaintiff lost, if anything. The brief makes it clear that the court followed a U.S. Supreme Court precedent (Dura Pharmaceuticals) that held that, "A securities fraud plaintiff must properly allege actual monetary loss - not merely harm related to purchasing at an increased price - in order to state a claim for securities fraud."
SUPREMES MISS THE POINT
We were stunned when we first read that quote, because the Supreme Court established a standard that it's not enough to pay too much for a security as a result of fraud or to watch helplessly while its price dwindles. Rather, you must have a sale transaction that produces an accounting loss.
To us, this doctrine is a Legally Adopted Measurement Error because every shareholder suffers a loss in the event of fraud, not just those who sold shares to a third party. We believe that this linkage to transactions reflects unsound economics and creates an unfair legal precedent that will make it difficult for anyone but a few lucky losers to recoup their losses from miscreants.
For reasons that were once practical because of limited access to reliable value information, the traditional accounting model typically records profits and losses only at the sale, not upon occurrence of events and activities that actually change value. For example, skilled furniture makers add value to a piece at each step in the production process, such as cutting, assembling and finishing. Despite the obvious increase in value, the traditional model doesn't recognize any income until sale. Although there were, and perhaps still are, sound reasons for these practices for reporting revenue, they cannot be applied usefully in all situations in which gains and losses take place.
In fact, real gains and losses happen when wealth changes, not when sales occur. The main effect of a sale is a change in risk, such that the seller now holds a receivable or cash instead of inventory. Critics of value-based accounting assert that recognizing income before realization is speculative. Surely, measuring market values poses some challenges, but there is great soundness in the economic concept that income occurs without a sale. And remember, the issue before the court was the existence and size of a real loss, not an accounting loss.
WHEN IS THERE A LOSS?
The disconnect in GAAP about realization and recognition is more evident in accounting for investments where there are few obstacles to knowing the value of assets at different dates. In this situation, real gains and losses can be measured when values change, even if the assets aren't sold. Again, all the sale accomplishes is risk reduction when the investment is converted to less volatile cash.
This point is precisely where the courts stumbled so badly. Mimicking accountants who serve a different goal, the judges asserted that a loss occurs if and only if investors buy at one price and sell at a lower one. This construct is deeply flawed, because it circumscribes out all losses by those investors who continued to hold the stock. Economically, their loss is no greater or less than those who sold.
FIFO/LIFO?
In the Organogenesis case, the District Court got caught up in yet another foolish accounting trap. The plaintiff described before had acquired shares in numerous transactions at different prices, both higher and lower than the price when he unloaded some of the shares. When realization accounting is applied in this situation, the existence and size of the loss is based on which costs are offset against the proceeds.
The need to assign an original cost to the sold shares raises the accounting issue of whether to use a FIFO or LIFO assumption. The plaintiff used FIFO and showed a realized loss; the defense, backed by Cornerstone Research, used LIFO and showed a realized gain. The court sided with the defense, refusing to certify this plaintiff into the class of damaged investors.
This tidy answer based on GAAP may appeal to accountants and judges, but it flies in the face of economic reality and justice. If management engaged in deception and inflated a stock price that plunged when the ruse was revealed, everyone who bought or owned the stock suffered a loss, regardless of their purchase price and regardless of whether they held or sold.
SO LAME
Unless overturned, the district court's decision will create a precedent that precludes some injured investors from seeking redress. If they can't show a realized loss based on LIFO, they will be out of luck.
This LAME accounting was supported in the defendant's pleadings with citations of the Internal Revenue Code's use of realization to measure taxable income. What the plaintiff's attorneys and the court missed was that the code was not created to measure economic damages created by fraud.
Instead, the code is designed to levy taxes on citizens when they have the wherewithal to pay. That is, Congress has deemed it fair to impose taxes only on investors who sold shares, because they have cash they can send to Washington. Applying this tax policy to litigation is a mistake that should be of deep concern to all accountants and attorneys.
Collectively, these court decisions have taken bad accounting and enshrined it in ways that should never happen. It is not too different from a conclusion that a company's loss from a property theft is limited to its GAAP book value when the crime occurred.
We wish there was an "undo" button to get rid of LAME accounting, but there isn't.
Paul B. W. Miller is a professor at the University of Colorado at Colorado Springs and Paul R. Bahnson is a professor at Boise State University. The authors' views are not necessarily those of their institutions. Reach them at paulandpaul@qfr.biz.
