In search of El Dorado; the FASB's quest for technical purity may be a quest for the impossible.

IN SEARCH OF ELDORADO

The FASB's quest for technical purity may be a quest for the impossible.

Many years ago, as a boy, I was housebound for some reason. For lack of anything else to do, I was reading some of the works of Edgar Allan Poe and memorized his poem, "Eldorado." It begins:

              Gaily bedight,
             A gallant knight,
        In sunshine and in shadow,
            Had journeyed long,
              Singing a song,
          In search of Eldorado.

Recently that poem came to mind as I thought of the Financial Accounting Standards Board and its search for the answers to complex accounting issues. Like the gallant knight, the members of the FASB may find their search is a vain one, particularly if they continue what seems to be a quest for technical purity.

Modern business transactions are complicated, and it follows that accounting standards designed to record and report those transactions also will be complicated. Recognizing that fact, those who set the standards should go to great lengths not to add further complexities that are not absolutely necessary.

INCOME TAXES

I was one of those who urged the FASB to reconsider the method of accounting for income taxes that had been used since 1968 after the Accounting Principles Board issued Opinion no. 11, Accounting for Income Taxes. The use of what is known as the liability method seemed to many practitioners to be simpler and more easily understood. The FASB did revise the way businesses should account for income taxes in Statement no. 96, Accounting for Income Taxes, and adopted the liability method. However, the result was a standard that was far more complicated than its predecessor and certainly more difficult to understand, implement and explain. It is so difficult to apply that its effective date has been postponed twice by the FASB. Implementation difficulties have been experienced by both small and large companies.

To cite just one example, consider the balance sheet classification of a liability for deferred income tax. Under Opinion no. 11, the liability was classified as current or noncurrent, based on the classification of the asset that gave rise to the deferred tax. I was not aware of any widespread dissatisfaction with that principle. But Statement no. 96 would require a determination of the future years in which the deferred tax liability would likely become due. The liability would then be classified in most instances as current if it would fall due within one year. If the deferred tax resulted from the use of accelerated depreciation for tax purposes, the assets giving rise to the liability would be noncurrent, but a portion of the liability would be current. Is this better, or just different? If we were to extend our quest for technical purity, one could as easily argue that current assets should include that portion of property and equipment that will be depreciated in the next year.

I may be accused of oversimplifying a complex issue, and perhaps I am. But I am convinced that the liability method of accounting for income taxes can be much simpler than it will be under Statement no. 96 without sacrificing the integrity or usefulness of financial statements.

OTHER POSTEMPLOYMENT BENEFITS

Consider one aspect of the debate about how to account for postretirement benefits such as health care. There is general (but not universal) agreement that the prevailing use of pay-as-you-go, or cash basis, accounting for this type of obligation is not satisfactory. It is generally agreed that the obligation to provide this type of benefit after retirement should be recognized as services are performed by the employee.

There are numerous and complex issues that must be addressed, but I will discuss only one: how to measure the obligation and related expense. After the transition to a new method of accounting is made, measurement is the issue that will remain.

The FASB has issued an exposure draft of a proposed statement on this subject (Employer's Accounting for Postretirement Benefits Other Than Pensions). One of the provisions in the exposure draft requires a company to select and use its "best estimate" of future healthcare cost increases in calculating the obligation and related expense. Some responses to the exposure draft have suggested the expected rate of general inflation be used instead. I say, "Why use either?"

To estimate future inflation, whether for health care or for general price levels, introduces an unnecessary subjectivity into measurement. It can detract from comparability. It takes into account something that has not happened--inflation.

At paragraph 167 of the draft, the FASB discusses the suggestion that employers be allowed to take into consideration future changes in the benefit plan, which may be adopted to offset the effects of cost increases. It states: "However, financial statements are intended to present the financial effects of events and transactions that have already happened. The obligation to provide postretirement benefits is measured as it is defined at the measurement date--future actions to change the promise are not anticipated, and it would not be representationally faithful to do so. Similarly, the cost for a period reflects the cost of the promised benefits earned during that period."

Because future inflation has not happened, the reasoning in paragraph 167 is equally applicable to future cost increases--however they might be measured.

In my opinion, the information will be just as useful if we simply take today's costs and apply them to the data base that would include expected retirement dates and life expectancy. It may even be more useful, since all companies would be using current data, which would avoid the use of different inflation estimates.

There is sound accounting logic to support this position. When inflation occurs, the obligation would be recalculated based on the new cost level. Thus, the year in which the inflation occurred would bear the expense of that inflation.

There are estimates that cannot be avoided, such as retirement dates and life expectancy. These can be actuarially computed with reasonable reliability. Inflation rates, however, cannot be computed with reliability. So why should we add estimates (guesses?) of inflation rates to the measurement process when there are sound reasons not to do so?

To recognize the obligation for postretirement benefits based on current cost levels adjusted annually will provide financial statement users with the information they need. It will be more comparable from company to company and less susceptible to manipulation. It will match the expense with the events giving rise to changes in the obligation. It will match current revenue dollars with current expense dollars rather than recognizing an expense based on estimated future dollar values with revenue based on current dollars.

USEFUL STANDARDS

It is my hope the FASB will find ways to simplify accounting standards to the extent possible, consistent with providing useful financial information. To do otherwise risks the erosion of its constituency.

It is vitally important for the FASB to succeed in its mission. The quest for technical purity makes that mission more difficult and often adds little or nothing to the utility of the resulting accounting standards. Indeed, such a quest can be futile and debilitating.

            But he grew old--
           This knight so bold--
        And o'er his heart a shadow,
             Fell as he found
            No spot of ground
        That looked like Eldorado.

SAMUEL A. DERIEUX, CPA, is a retired partner of Deloitte and Touche, Richmond, Virginia. A member of the AICPA council and chairman of the AICPA task force on accounting standards overload, he is a past president and past chairman of the board of the AICPA.