As told to me by a faculty member at Ohio State, who heard it in a presentation by Eugene Flegm, former general auditor for General Motors:
Some years after the end of World War II, GM was negotiating a new contract with the United Auto Workers. The labor union proposed that GM should pay an additional amount per hour to its unionized employees, which the UAW would take and invest in a pension fund. Pension payments to retirees would eventually become the full responsibility of the UAW.
The CEO of GM consulted with his economist, who advised that the proposal was a good deal for GM: the expected cost of funding the pensions would remain the same, but the risk of funding unanticipated losses would be borne entirely by the union.
The CEO next consulted with his accountant, who informed him that current profits would suffer if the union's offer were accepted: the additional wages called for would be reflected in expense immediately. On the other hand, if GM were to keep the responsibility (and risk) of paying the pensions, recognition of the associated expense under extant GAAP would be delayed for decades – until retirees were actually paid.
The CEO chose to disregard the economist's advice and to decline the UAW's offer.
Postscript: Following similar decisions of subsequent CEOs and questionable funding decisions, GM's underfunded liability for retirement benefits grew to $50 billion as of December 31, 2008. In December 2008, under President Bush, GM received $13 billion of TARP funding, and another $39 billion under the Obama administration.
The 'Moral of the Story'
This tale of GM wealth destruction is just one of many that can be told that begins with a bad accounting rule. It's a relatively dated story, but if the first twelve years of this millennium are any indication, the problem of perverse accounting incentives has been getting much worse – despite the putative best efforts of standard setters to enhance financial reporting "quality."
Either the problem is seen by accounting standard setters to be in the "not my job" category, or if they are sensitive to it at all, then history suggests they are going about it the wrong way. That's just one reason why convergence hasn't been worth a hill of beans, and any forthcoming statement from the SEC about continuing on with IFRS adoption in one form or another is dross. I'll have more to say about that at the end of this post.
The wrong way that the Boards have been pursuing is predicated on the explicit assumption in their respective conceptual frameworks that the fundamental purpose for financial statements is to provide information to help interested parties value an entity. But, a major problem with this approach is that it doesn't provide any consistent answer as to how the recorded assets and liabilities of the entity should be valued. Instead, the approach is used as a justification for the kinds of accounting that managers can manage.
To illustrate, one of the most recent contributions to the valuation perspective for accounting standards is a new book by Columbia professor Stephen Penman, Accounting for Value. In many respects it is an excellent book and I highly recommend it; however, I take exception with his line of reasoning that rejects current values as the basis for asset and liability measurement.
Without wishing to oversimplify, there are two sources of equity value from Penman's perspective: things that we know; and things we can only speculate about. If accounting restricts itself to measuring things we know (such as the historic cost of land), then the amount of speculative value in the current stock price can be deduced from the difference between it and the book value per share. Penman argues that speculation should be left to investors, for if accountants were to engage in too much speculation, then it will be difficult to avoid paying a market price that could over value a company.
I do have questions about Penman's analytics that lead to a conclusion that historic cost accounting is to be preferred to current value accounting; but, if valuation is not the primary objective for financial statements, then it doesn't matter whether he is right or not. My intent is solely to point out that a rigorous defense of historic cost accounting depends on whether one maintains that it is appropriate to prepare financial statements under the presumption that valuation analysts are the primary user group. My position is that to determine that this assumption is inappropriate, it should be sufficient to have enough stories like GM, where transaction-based financial statements clearly caused massive value destruction. Based on my own experience and study, I would estimate that there are thousands of examples out there just like it – albeit few that are as overtly laden with societal consequences.
The right way to select a basis of measurement should begin with the assumption that financial statements exist to guard the interests of investors in an entity. Indeed, I would think that this "stewardship" function was the original intent of double-entry accounting in the first place. It entails a high level of accountability from managers to owners for the manner in which business was conducted. Normally, this should mean that managers have an obligation to utilize the (net) assets of a business in a manner that appropriately balances risk with return on the owners' investment.
Of course, neither stewardship nor any other financial reporting system is foolproof. But, if accounting has any hope of bringing perverse incentives to manipulate financial statements down to a tolerable level, the implication of a stewardship perspective is that accounting must focus, to the extent practicable, on measuring the effect of managerial decisions on wealth and changes in wealth.
While some argue that estimating current values injects too much subjective judgment into accounting, I suggest that no matter what system of accounting is proposed, judgment is an unavoidable reality; and therefore, quibbling about which approach to valuation requires larger doses of judgment is largely irrelevant. A stewardship role for accounting also implies that it is not "management's view" that should count, for that would be circular logic indeed. The only view that should matter is the view of a disinterested economist, taking into consideration current business conditions.
Getting back to that annoying IFRS question, which most of us wish would just go away, the SEC's Chief Accountant, James Kroeker, spread the word last week that he will be preparing a proposal to the Commission for moving forward on IFRS. I don't know what will happen here, particularly since not a single Commissioner has delivered a major speech on accounting standard setting – despite the fact that it is one of the most critical regulatory functions within the SEC's purview. If any of the five sitting Commissioners has a reasonably complete understanding of accounting standards and policy, I haven't seen it demonstrated.
Obviously, I'm not going to go into all of the reasons why the SEC should finally move away from IFRS in this blog post, but I do hope the Commissioners will ask themselves this one question:
If accounting standards can either create or eliminate perverse incentives for the managers of public companies – with potentially profound consequences for the U.S. economy – how can we set a goal of convergence with standards that cater to "management's view" instead of a goal of reflecting economic reality?