OPEB: improved reporting or the last straw? The FASB should rethink parts of its proposal on postretirement benefits other than pensions.

The Financial Accounting Standards Board's Exposure Draft (ED) Employers' Accounting for Postretirement Benefits Other Than Pensions, usually referred to as OPEB, has been characterized as a way to improve financial reporting and, alternatively, as a pronouncement that will devastate reported corporate profits needlessly. Simply stated, the FASB is proposing accrual-basis reporting of the cost of, and the obligations for, postretirement benefits, instead of the prevalent current practice of report-as-you-pay (cash-basis) reporting. Though the ED encompasses all postretirement benefits currently owed and expected to be provided to current and future retirees, healthcare will constitute the bulk of the cost. The amounts involved can affect financial statements materially and adversely, and that has the preparers of financial statements worried.

COMPARISON TO PENSION

ACCOUNTING

When companies intially offered postretirement benefits, healthcare costs were relatively low and the ratio of covered retirees to active workers was small. Since cash payments for these costs were generally immaterial, cash-basis expense recognition was acceptable. But a combination of escalating healthcare costs and changing demographics has altered the postretirement landscape.

In deciding how companies should account for postretirement benefits, the board paid particular attention to FASB Statement no. 87, Employers' Accounting for Pensions, because there are similarities between the way pension and other postretirement benefits affect companies.

In some areas, accounting for both pensions and OPEB is inconsistent with the traditional accounting model. These include

* Delayed reporting of events that have occurred.

* Combined current and past compensation costs, financing costs and investment gains and losses reported on a net basis.

* Asset and liability offset without the presence of conditions specified in FASB Technical Bulletin no. 88-2, Definition of a Right of Setoff.

Similarities consistent with traditional accounting principles include

* Relating benefits to years of service.

* Matching costs with revenue.

* Reporting obligations as financial statement liabilities.

Because of similarities between the two models, it is conceptually appealing to pattern the unknown after what is now the accepted methodology of pension accounting. Nevertheless, the accounting implications of differences between pensions and postretirement benefits must be acknowledged. Following are some differences:

* Most postretirement benefits do not increase with years of service.

* Postretirement benefits are payable in services rather than cash.

* Postretirement benefits generally do not vast before retirement--an employee who quits shortly before retirement usually has no claim to them.

* Employers at present have little tax incentive to prefund non-pension benefits. (Some limited opportunities exist for deductible VEBA contributions, deductions to 401(h) accounts, pension benefit enhancements and corporate-owned life insurance.)

Any departures from traditional reporting in postretirement benefit reporting must be justified on their own merits, not just because they follow FASB Statement no. 87's departures.

REPORTING ISSUES

This article considers four key questions:

1. Does a postretirement benefit plan result in an obligation that meets the definition of a liability?

2. If so, how should the obligation be measured?

3. If the obligation can be measured, when should the liability be reported?

4. How should the transition from cash basis to accrual-basis reporting be handled?

ARE THEY LIABILITIES?

The first reporting issue in addressing other postretirement benefits is whether they are liabilities. In Concepts Statement no. 6, Elements of Financial Statements, paragraph 35, the FASB defines liabilities as "probable future sacrifices of economic benefits arising from present obligations of a particular entity to transfer assets or provide services to other entities in the future as a result of past transactions or events."

Pivotal to the board's stance for liability reporting is the view, stated in paragraph 125 of the ED, that postretirement benefits are a form of deferred compensation, not a gratuity. In exchange for the employee's current services, the employer promises the employee both current wages and current and deferred benefits. Critics counter that many of the benefits are discretionary, not meeting the definition's requirement that the liability arise from present obligations.

In defining liabilities, the FASB has implied that the legalistic definition serves as the floor. Thus, if employers are deemed legally liable for postretirement benefits, a liability should be recorded. And the courts generally have upheld the obligation of employers to provide postretirement benefits if there was no unambiguous communication before an employee's retirement that the company reserved the right to reduce or terminate his or her benefits.

As the enterprise continues in existence, some postretirement benefits utimately will be paid, and the "probable future sacrifice" requirement will be met. Payment obviously would not occur without a specified employment period when the employee earned these benefits; thus the "result of past transactions or events" component of the liability definition is satisfied.

Beyond the liability definition. What about the exceptions--for example, companies that decrease the size of their operations? A frequent criticism of recording postretirement benefits as liabilities is that the benefits may never be paid because companies can alter these plans at will. Whether companies decreasing their operations can reduce postretirement benefits to nonretired individuals who are no longer employees is being decided in the courts.

What about businesses that may be found not legally liable for retiree benefits? Two existing components of generally accepted accounting principles can help here:

1. Nothign in GAAP requires certainty for the reporting of an element. Implicit in the probable component of the liability definition is the possibility the liability will not materialize. Although deferred compensation was excluded from the scope of FASB Statement no. 5, Accounting for Contingencies, paragraph 7, an analogy provides further support for this lack of certainty. When describing "the likelihood that the future event or events will confirm the loss or impairment of an asset or the incurrence of a liability" (Statement no. 5, paragraph 3), the board identified only three ranges: probable, reasonably possible and remot. If a future event is probable, which means only that it is likely to occur and can be estimated reasonably (discussed later), the board requires that a liability be reported (Statement no. 5, paragraph 8).

2. Accounting Principles Board Opinion no. 20, Accounting Changes, paragraph 10, recognizes that "accounting estimates change as new events occur, as more experience is acquired, or as additional information is obtained." Such changes could include revised estimates for postretirement benefit costs resulting from a change in any component of cost for these plans, including eliminating the plans and thereby all the costs. The board incorporated methodology for reporting on curtailments and settlements of postretirement benefit obligations in ED paragraphs 85-94. So, even when a company alters its obligation to provide benefits, the existing reporting framework is equipped to handle these changes.

MEASURING THE OBLIGATION

The most difficult task in implementing the conclusion that a postretirement benefit obligation is a liability that should be reported is measuring that obligation. The task is complicated by the lack of historical claims data and the uncertainties surrounding future healthcare costs.

Qualitative issues. In trying to quantify the postretirement benefit obligation, the FASB again confronts the pervasive trade-off between relevance and reliability. Information regarding liabilities is relevant by definition. But critics have suggested that uncertainties inherent in measuring OPEB liabilities render the estimated amounts unreliable.

The use of estimates is an integral part of the reporting framework. Estimation does not necessarily impede reliability, which cannot be equated with precision. As the board notes in Concepts Statement no. 2, Qualitative Characteristics of Accounting Information, paragraph 59, reliability "is hardly ever a question of black or white, but rather of more reliability or less." Nevertheless, the proposed OPEB expense reporting will be based on softer numbers that most other financial statement amounts. The board acknowledges this difference in paragraph 66 of the ED by proposing sensitivity disclosures--the effect on the accumulated postretirement benefit obligation of a 1% change in the healthcare cost trend.

Projecting future costs. ED paragraph 33 would require that estimates include projections of the future cost of providing healthcare benefits. That would make it necessary to project such diverse items as utilization levels, technological advancements in medical knowledge and demographic information. However, cost-sharing programs with other healthcare providers are singled out as an element of OPEB cost requiring a seemingly higher level of evidence to recognize changes. The evidential level in this case requires a change in law, contract or regulation rather than a projection of such future change.

Despite its stated aversion to having a legalistic pespective, the board is emphasizing the legalities of cost-sharing arrangements. Prohibiting projected changes in cost-sharing programs is inconsistent with other specific projection requirements--if some cost elements can be projected to change, why not cost-sharing programs? Concerning occurrence of a future event, FASB Statement no. 5, paragraph 8, requires only that the event be probable for it to be recorded in the financial statements. Based on changes in economic conditions, financial institutions routinely alter their loan loss reserves for anticipated defaults on contractual loan agreements with customers. FASB Statement no. 48, Revenue Recognition When Right of Return Exists, paragraph 6, requires that sellers must be able to reasonably estimate returns of a product before revenue can be reported from the product's sales. In the ED, the board is inconsistent in allowing estimation of changes in only some elements of cost. It should allow projections of all facets of cost.

Measurement analogies. Some items that fit the definition of a given financial statement element are not reported because they are not measurable. Are postretirement benefits in this category or are they measurable enough to be reported as liabilities?

FASB Statement no. 2, Accounting for Research and Development Costs, paragraph 44, implies that assets are not recognized for R&D costs because future economic benefits cannot be measured objectively. The same reasoning applies to advertising expenditures. Even though many advertising campaigns are targeted to enhance the company's image and build long-term sales, those benefits cannot be measured, so advertising costs generally are expensed when incurred.

Even when reporting guidelines specify asset reporting, sometimes practical application is impeded because of measurement difficulties. For example, APB Opinion no. 16, Business Combinations, paragraph 68, specifies that in a business combination accounted for by the purchase method, the purchase price should be allocated to identifiable assets. But even in purchases in which valuable trademarks are acquired, goodwill is frequently reported instead of attributing a portion of the purchase price to the trademarks because of the difficulty of determining their value.

Clearly, the above accounting issues indicate areas where measurement difficulties prevent recognition of an element that otherwise probably qualifies as an asset. But these are more in the nature of cost allocation concerns. Whether an obligation represents a financial statement liability evokes a new set of concerns, best addressed through analysis of FASB Statement no. 5. That statement restrict liability recognition to situations in which the loss is probable and the amount can "reasonably" be estimated. But, given the wide range of contingent loss possibilities (such as risk of loss from future injury to others, threat of expropriation and litigation), it is appropriate to acknowledge that in some cases the items simply are not subject to reasonable estimation.

Postretirement benefits are undeniably difficult to measure, but measuring them is more objective than measuring amounts for the possibilities listed above. At least with postretirement benefits, the current costs for these expenditures provide a starting point for estimation. OPEB costs are different from lawsuit or injury losses, for example; for these there may be no basis for making estimates.

WHEN SHOULD THE LIABILITY

BE REPORTED?

The OPEB liability reported at any point in time ideally should be the sum of the total liability for retirees and an assigned portion of the ultimate total liability for active employees. Attribution is more difficult with postretirement benefits than with pensions because pension benefits generally are specified by plan formulas. And postretirement benefits usually cliff-vest--that is, only on retirement is the employee eligible to receive benefits. The amount ultimately paid usually is dictated by factors unrelated to an employee's years of service in a company.

Probably because most plans give retroactive credit for service with a company before a healthcare plan is initiated or amended, the ED recommends the attribution period begin at the hire date. Specifying the end of the attribution period has been a contentious issue for the FASB; two dates considered were the full eligibility date and the expected retirement date. The board opted in ED paragraph 20 to end the attribution period at the full eligibility date, essentially concluding that all benefits had been earned at that time and no additional benefit could be gained by working past the full eligibility date. Also, the board concluded in ED paragraphs 204 and 206 that using the full eligibility date was consistent with the requirements of FASB Statement no. 87.

Despite differences between the two, the FASB appears to be forcing postretirement benefit reporting into the pension mold, even where it doesn't fit. The FASB should consider allowing companies to use the expected retirement date. Employees do not receive postretirement benefits when they are eligible to retire; benefits are received on retirement. Using the retirement date more clearly reflects the true substance of the exchange.

TRANSITION METHOD

Standard-setting bodies have identified four different transition methods:

* Cumulative effect of the change in reporting principle reported in the period of the accounting change, as specified by APB Opinion no. 20.

* Restating prior period presentations so all presentations conform to the new reporting principle, as specified by Opinion no. 20 and FASB Statement no. 16, Prior Period Adjustments.

* Grandfathering in the transactions previously entered so transactions before a given date are recorded by the old principle and transactions after that date are recorded by the new principle, for example, APB Opinion no. 17, Intangible Assets.

* Phasing in the effects of the new principle on previous transactions so that, for some period of time, financial statements reflect some of the old principle and some of the new principle but, by a later date, all financial data reflect the new principle, for example, FASB Statement no. 87.

ED paragraph 220 opts for the last method and requires the amortization of any unreported transition asset or liability. For the sake of not burdening one income statement and not reflecting an existing unrecorded liability all at one time, the FASB has chosen to phase in the effect of previous promises to employees. Because of similarities in the two kinds of benefits, specifying the same transition method for pensions and other postretirement benefits has merit.

However, the arguments presented in Opinion no. 20 concerning the treatment of a change in reporting principle also should apply to these two kinds of benefits. The FASB recognized in ED paragraph 22 that the OPEB transition is a cash-basis problem. Accordingly, GAAP specify that financial statements be restated. With postretirement benefits, the board concluded in ED paragraph 224 that the cost of either restating financial statements or reporting a cumulative effect in the current period would be prohibitive. The board also said the actuarial techniques for measuring these obligations are still evolving.

The cost of gathering the data to restate prior periods might be prohibitive but, if the obligation will be amortized, an amount representing the obligation at the transition date must be developed. It follows that there is no data-gathering cost differential between reporting a cumulative effect and phasing in through amortization. Because the FASB recognizes that actuarial techniques are still evolving, a more pragmatic alternative may be to delay implementation, as was done for FASB Statement no. 96, Accounting for Income Taxes. The FASB has answered this criticism partially by postponing the effective date of the proposed standard to 1993.

Despite recognizing the conceptual merit of currently accounting for the change in principle, the board seems to have bowed in ED paragraph 120 to the sensitivity of employers that have been able to avoid reporting this liability for so long. But this approach has serious drawbacks:

1. It understates liabilities until the phase-in period expires, to the detriment of the report users.

2. During the phase-in period, the income statement must report expenses of previous promises and previous activities while it is reporting current expenses.

Until conceptual issues to support the phase-in method are raised and found to have merit, the cumulative effect method specified by Opinion no. 20 should be applied to all reporting changes.

POLICY CONCERNS

In addition to sorting out the conceptual issues, which might lead to sound reporting on postretirement benefits, there are myriad social and economic consequences that will result from the reporting choices selected by the FASB. Few financial statement users would suggest that tax consequences should dictate reporting pronouncements or that reporting guidelines are intended to drive employers' OPEB plans. Nor has the FASB set out to promulgate standards for the sole purpose of reducing reported corporate earnings. Nonetheless, these and related concerns arise when considering the proposed new reporting rules in a broader social and economic context.

Earnings reduction vis-a-vis neutrality. Although data-gathering and record-keeping costs are significant, businesses seem most rankled by the proposed statement's effects on reported earnings. As accountants, we are trained to be neutral, and it goes against our grain when a statement that appears to improve financial reporting is criticized because it affects reported earnings negatively. But the political reality of standard setting is that neutrality is an ideal desired by standard setters, not by individual companies.

The board notes in Concepts Statement no. 2, paragraph 98, that "neutrality means that either in formulating or implementing standards, the primary concern should be the relevance and reliability of the information that results, not the effect that the new rule may have on a particular interst." Taking this one step further, not being neutral implies that standard setters should aim for specific policy goals; this first implies that agreement on desirable social and economic policy goals exists. It doesn't. Even if such a consensus did exist, it is not the function of reporting standard setters to implement social and economic policy objectives; such a task should be handled through the public sector.

Social-policy considerations. Some companies may consider the proposed guidelines so onerous that their benefit plans will be reduced or eliminated. While the board acknowledges in ED paragraph 130 that such actions may be a consequence of companies assessing and reporting their obligations, these changes are not the board's objective. The new pronouncement would not cause companies to incur additional costs for retirees; it simply would require reporting costs that the companies themselves have committed to incur.

Other companies will continue their cash outlays for retiree benefits regardless of whether this ED is issued. By continuing the currently prevalent cash basis, the profession would be endorsing the concept that companies have no obligation to provide postretirement benefits. Enactment of the proposed standard would allow society to assess this cost. And if the cost to employers is determined to be too high, perhaps the reporting standard will be the impetus for creating more efficient alternatives for society.

FINANCIAL STATEMENT AND

MARKET CONSIDERATIONS

The Employee Benefit Research Institute has estimated the total unfunded employer liability, including that of the public sector, to be about $280 billion; the private sector's share is estimated to be about $169 billion. The U.S. General Accounting Office has recently estimated the private sector's liability to be $221 billion. The primary difference between the two estimates is Medicare benefits.

Information from companies willing to use the proposed guidelines to compute the reportable cost illustrates the diversity of financial statement effects. A study conducted by the Financial Executives Research Foundation reports that the 26 companies tested would experience decreases in pretax income ranging from 2% to 20%. Two-thirds of these companies predicted an increase in expense of two to six times current costs.

FUNDING

The funding issue is separate from measuring the OPEB obligation. Nonetheless, funding has reporting ramifications because, for companies to minimize reported liabilities, OPEB plans must be funded--a costly alternative in the current tax environment. Although the FASB knows about the practical difficulties posed by existing funding constraints, as stated in ED paragraph 129, it holds how and when to fund retiree benefits is a financing decision influenced by many factors, of which tax is one.

Unlike contributions to pension plans that are deductible when paid, general tax rules prohibit current deductions for deferred compensation. Since no tax deduction is permitted until postretirement benefits are provided to retirees, companies have little tax incentive to prefund the plans.

It is counterproductive to focus on the current tax prohibition for deducting prefunded retiree healthcare. If retiree healthcare is indeed a desirable policy goal, and if employers are deemed the best providers of that service, the ED may serve to recognize that goal and initiate a change in the tax structure that will make prefunding more attractive. Perhaps companies disgruntled by the effects on reported earnings will be the catalyst for that change.

A WORKABLE SOLUTION

Accounting for postretirement benefits is so complex there just isn't an ideal solution to the many problems involved. Therefore, we must move in the direction in which the preponderance of evidence takes us--accrual-basis recognition--and keep working to improve the problem areas. The current cash-basis approach is misleading because of its implication that no liabilities for retiree benefits exist. Although the evidence presented suggests that the board reexamine some specific issues, the ED's primary provisions generally can be supported using the conceptual framework.

PAULA B. THOMAS, CPA, CMA, DBA, is associate professor of accounting at Middle Tennessee State University, Murfreesboro. She is a member of the American Institute of CPAs, the National Association of Accountants and the American Accounting Association. LARRY E. FARMER, CPA, CMA, CIA, DBA, is professor of accounting at Middle Tennessee State University. He is a member of the AICPA, the NAA and the AAA. Drs. Thomas and Farmer are the authors of "Accounting for Stock Options and SARs: The Equality Question," which was published in the June 1984 Journal of Accountancy.