FASB standard on accounting for other postretirement benefits takes shape. (Financial Accounting Standards Board) (Accounting) (column)by Carmichael, Douglas R.
FASB deliberations on employers' accounting for postretirement benefits other than pensions (OPEBs) have progressed to the point where the broad outlines of the statement can now be discerned. Absent some unexpected FASB reversal, the final statement will be substantially the same as the 1989 exposure draft, with two important exceptions:
* If certain conditions are met, employers' cost computations can reflect anticipated plan changes; and
* Employers need not recognize a minimum liability in excess of the accrued benefit cost.
The Board also removed one controversial disclosure requirement. However, it has rejected most other suggested changes, including--so far at least--the hoped-for change in the period over which the cost will be accrued. At this writing, the Board has not yet taken up the difficult topic of transition to the new requirements.
Anticipating Plan Changes--The
Since January 1990, when the FASB began to consider comments on its proposed OPEB statement, one question has dominated the Board's discussion: When, if ever, in computing an employer's benefit cost can plan changes be anticipated?
On its face, the question may seem odd. A fundamental tenet of accounting is that financial statements portray the results of transactions and events that have already happened. The suggestion that an employer's OPEB cost might reflect future events, specifically future changes in the governing plan instrument, seems like one the FASB would reject out of hand. In fact, the exposure draft (paragraph 167) did reject it: "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." And, when I suggested the possibility of anticipating plan changes at a meeting of the FASB's OPEB task force, I came as close as the Board's polite proceedings allow to being hooted out of the room.
Yet the anticipation of plan changes now seems virtually certain to be in the OPEB standard: the Board has tentatively agreed to it. The evolution of the notion from derision to respectability provides insight into some of the special problems of accounting for OPEBs.
That Which We Call a Rose...
Part of the reason the FASB became reconciled to the need for allowing future plan changes to enter into the computation of OPEB cost was the Board's realization that an OPEB plan is not just a pension plan with a different name. It is in some ways a fundamentally different arrangement. On paper, the differences are not apparent. Both pension plans and OPEB plans promise retirement benefits in exchange for employee service before retirement. In theory at least, most of the same actuarial techniques can be used to compute the cost of benefits earned by each year of service. The differences between the two plans show up in the way the plans are managed, and they arise from the different kind of promise each plan makes.
The pension plan promise is largely self-limiting. A plan promises benefits in the form of fixed or determinable dollar payments, thereby limiting the employer's obligation to a reasonably determinable amount. The obligation can of course be increased by a plan amendment. However, absent a medical breakthrough greatly extending the human life span, or some other equally unexpected development, little outside the employer's control can cause the pension obligation to greatly exceed the approximate amount forecast when the plan was adopted.
In contrast, no such limits confine the typical OPEB obligation. While the nature of OPEB plans is evolving, the typical plan still is a cost reimbursement plan. It promises benefits in the form of services purchased when needed at prevailing prices.
For a retiree medical benefit plan, this promise ties the employer's obligation to medical cost inflation, a component of the Consumer Price Index that is growing faster than any other. Anyone who has experienced recent injury or illness can find in his or her checkbook ample evidence of the steep increase in medical costs. Despite intensive cost containment efforts, annual percentage increases in medical costs remain in double digits, and annual increases in medical insurance premiums have recently exceeded 20%.
Some Control Costs By Changing
Employers tied to retiree medical plans with no built-in cost limits have turned to the only effective cost-control technique available to them: changes in plan terms; and they have used the technique extensively. Shell Oil, for example, testified at the FASB's public hearing on OPEBs that it had changed its retiree medical plan in each of the preceding 10 years and intended to continue annual changes as long as necessary to keep the plan's cost within reasonable bounds. Such changes are widespread, as shown by a number of surveys. The 1989 employee benefits survey of the U.S. Chamber of Commerce found, for example, that one third of the 932 respondents had during the preceding two years either added a deductible or raised an existing deductible on their health insurance plans. An equal number had increased the percentage of health insurance premiums paid by employees.
Unless the FASB allowed the accounting for OPEBs to anticipate such changes, many companies argued that the results would make no sense. For example, while Shell projected that by the year 2058 medical costs under its plan would reach $420,000 per person, the exposure draft would have limited the per person deductible to its current level of $250!
If necessary to get sensible accounting results, Shell said it would write cost-containment provisions into its plan. But such provisions would limit the company's ability to tailor its response to fit each year's circumstances, and it urged the FASB to find a better alternative.
While the FASB readily recognized the merits of Shell's argument, the Board initially balked at the suggestion that employers be allowed to recognize the effects of future plan changes.
Anticipating Plan Changes--The
In the end, however, the Board overcame its doubts and agreed that in certain circumstances anticipated plan changes can be recognized in the computation of OPEB cost. Specifically, the Board agreed first that the basis for OPEB accounting should be the "substantive plan," that is, the plan as it is operated. As explained in the FASB's Action Alert for April 11, 1990, substantive plan provisions may be established by "an employer's pattern of increasing or reducing participants' contributions or certain other employer and retiree cost-sharing provisions." If so, these substantive plan provisions should be the basis for the employer's accounting, even if they are not part of the written plan.
The Board further agreed that an established pattern of changing cost- sharing provisions generally should be assumed to continue in the future. However, the pattern may be changed if the employer has evidence of its ability and intent to change the pattern and has communicated that intent to plan participants.
Even if an employer has not established a pattern of plan changes, the Board agreed to allow anticipation of a change if four conditions are met.
* The employer has the ability to change the plan.
* The employer has the intent to change the plan.
* The employer has communicated to plan participants its intent to change the plan.
* The employer has communicated to plan participants the expected timing of the change.
It is important to note that the anticipation of plan changes is limited to changes in cost-sharing provisions--deductibles, coinsurance, employer "caps," employee out-of-pocket limits, employee contributions, and the like. It does not extend to changes in eligible charges (for example, a change that would exclude certain types of hospital charges) or changes in basic coverage (for example, a change that would eliminate dental coverage). Anticipating either of these changes, the Board concluded, would overly complicate the problems of measurement.
By allowing employers to anticipate plan changes directly, the Board was able to pass over a number of suggestions for accomplishing the same thing indirectly. For example, some commentators had suggested that employers project benefit cost increases at the general inflation rate instead of at the usually much higher health care cost trend rate. In the long run, these commentators reasoned, an employer will try to manage its medical plan, by regular plan changes or other means, to keep the rate of cost increase reasonably close to the general inflation rate. The FASB rejected this and other suggestions for a standard cost inflation rate, preferring instead a rate that reflects each employer's particular circumstances.
Eliminating the Minimum
The FASB's tentative decision to eliminate the minimum liability reporting requirement caught a lot of people by surprise. While two Board members (Brown and Northrop) had objected to the proposed requirement in the exposure draft, the other five members seemed firm in their determination to carry it forward into the final statement. After all, the proposal largely paralleled a pension accounting requirement of SFAS 87 and it had not been a particularly contentious issue at the Board's public hearings.
However, at the April 11, 1990 meeting, the Board overturned the proposal. The reasons for the decision varied. Brown and Northrop had argued in the exposure draft that the notion of a minimum liability is inconsistent with the Board's decisions for delayed recognition of gains and losses, prior service cost, and the transition obligation. Other Board members were reportedly won over by the fact that the minimum liability would have been based on the obligation to "fully eligible plan participants" (retirees or employees who could retire with full benefits), while the benefit cost is based on the obligation to all participants. These members questioned whether financial statement users would recognize the difference or be able to comprehend its purpose and significance.
Footnote disclosure of the obligation, they finally concluded, would be more comprehensible and just as useful as balance sheet recognition.
Because the effective date of the minimum liability reporting requirement would have been deferred five years anyway, eliminating the requirement will make little difference to many companies. Of the companies participating in the Financial Executives Research Foundation's field test of the FASB proposals, over one-quarter would have already accrued a liability at least equal to the minimum by the end of the sixth year. For them, the requirement was of no consequence. Most of the remaining companies in the field test would have accrued the minimum by the end of the tenth year, so the effect on them would have been temporary.
The FASB also agreed to drop the unpopular proposal for an employer to disclose its vested benefit obligation. This "walk-away liability," which assumes that all employees entitled to benefits will quit tomorrow, was roundly criticized as meaningless.
Board Still Likes Eligibility Date
Not Retirement Date
However, the Board has decided to retain the one provision of the exposure draft that probably drew more criticism than any other, both in the letters of comment and at the public hearings. That is the requirement for the OPEB cost to be accrued over the period of employee service up to the date of full eligibility for benefits. Many argued that the period should end at the employee's expected retirement date, contending that management intends postretirement benefits as a reward for the entire period of employee service. Thus, they argued, the proposed shorter "attribution period" would not match the benefit cost with the period of employee service earning the benefits.
Those who made this argument thought it had a good chance of prevailing, since the Board had flip-flopped on the length of the accrual period when preparing the exposure draft. But when the final vote was taken, the proposal in the exposure draft was sustained, albeit by a slim 4:3 vote.
The majority was persuaded to retain the proposal by three arguments:
* The terms of the plan establishing a date for full eligibility for benefits clearly define the end of the period of employee service exchanged for the benefits.
* Accounting is either based on plan terms or it is not; any attempt to distinguish those plan terms that are relevant for accounting purposes from those that are not relevant will fail.
* Accrual to the full eligibility date is consistent with the accounting for other forms of deferred compensation.
The Board then proceeded to strengthen the last of the three reasons by amending APB Opinion 12, as proposed in the exposure draft, to make a comparable change in the requirements for accruing the cost of individual deferred compensation contracts. Unless the contract defines the years of service exchanged for the compensation, the amendment will require the cost to be accrued over the period ending at the date of full eligibility. Currently, APB 12 is not clear as to when the accrual period ends, but in practice it has usually ended at the employee's expected retirement date.
What Else Was Rejected?
Most other FASB decisions to date have also reaffirmed provisions of the exposure draft. Suggested changes that were rejected include the following:
* Changing the discount rate either to a company-specific rate (such as a company's incremental borrowing rate, its internal rate of return, or the rate of return on plan assets) or to a "normalized" long-term rate. The Board found suggestions for a company-specific rate hard to implement and the use of a "normalized" rate inconsistent with the objective of measuring the benefit cost and obligation as of a point in time. Accordingly, it reaffirmed the exposure draft's proposal for benefit projections to be discounted at an external market rate derived from the yields currently available on high-quality, fixed-income investments with maturities matching the timing of expected benefit payments. That is consistent with the discount rate for pension accounting required by SFAS 87.
* Eliminating the proposed requirement for benefit projections to be based on explicit assumptions for gross eligible charges, Medicare reimbursements, deductibles, coinsurance, retiree contributions, and so forth. Many had argued against this proposal, pointing out that insurance administrators often have reliable information only for net incurred claims costs. Howerver, the actuarial profession is working to find ways to approximate the amount of each components, and the Board tentatively decided to reaffirm its earlier proposal.
* Widening the corridor in the so-called corridor method of amortizing gains and losses. The corridor will remain at 10% of the greater of the accumulated benefit obligation or the market-related value of plan assets. The Board considered requests for a 20% corridor as a means of reducing year-to-year cost fluctuations. However, when an FASB- sponsored study revealed that widening the corridor would have only a slight dampening effect, the Board decided to keep the corridor at 10%. This too is consistent with the pension accounting requirements of SFAS 87.
* Changing the required attribution method or allowing employers to choose from alternative methods. The Board was persuaded that there was no reason to require or allow an attribution method for OPEBs that differed from the benefits/years-to-service method that SFAS 87 requires for pensions.
Transition Requirements and
The most troublesome topic still awaiting an FASB decision is the transition to the new requirements. Letters of comment and oral presentations at the public hearings suggested two principal changes to the transition proposals of the exposure draft.
The first suggestion is for the FASB to allow employers the option of taking a one-time charge to earnings for the full amount of the unrecognized prior service obligation at transition. The exposure draft would require this "transition obligation" to be deferred and amortized only. Those who suggest an alternative argue that employers should not be required to burden future years' operations with the cost of prior years' benefits. While not every employer can stand to absorb the entire burden in one year, many feel that those who can should be allowed to do so.
The second suggestion is for the FASB to extend the optional period over which employers can amortize the transition obligation. The exposure draft generally calls for amortization over the average remaining service life of active employees, but if the resulting amortization period is less than 15 years, would allow employers the option of a 15-year amorization period.
While the transition proposals of the exposure draft parallel the requirements of SFAS 87 for pension plans, those who advocate extending the optional period believe that the pension requirements are not apt precedents. When SFAS 87 was adopted, so the argument goes, pension plans were generally so well funded that large transition obligations were rare. In contrast, OPEDB plans are generally not funded, so large transition obligations will be the rule. Extending the optional amortization period to 20 or 30 years would ease the burden of absorbing large obligations.
The FASB still expects to get its OPEB standard out by the end of the year and so far is moving at a pace that makes the expectation realistic.
Dale L. Gerboth, CPA, Partner, Ernst & Young
The CPA Journal is broadly recognized as an outstanding, technical-refereed publication aimed at public practitioners, management, educators, and other accounting professionals. It is edited by CPAs for CPAs. Our goal is to provide CPAs and other accounting professionals with the information and news to enable them to be successful accountants, managers, and executives in today's practice environments.
©2009 The New York State Society of CPAs. Legal Notices
Visit the new cpajournal.com.