…Eye of newt, and toe of frog,
Wool of bat, and tongue of dog,
Adder’s fork, and blind-worm’s sting,
Lizard’s leg, and owlet’s wing …
If you have read my blog every now and then, it won’t come as a surprise that this smorgasbord from Macbeth would recall for me a number of accounting topics comprising U.S. GAAP. More on that — as Shakespeare might say — anon.
This is the third installment in a series (first, second) of posts in response to Russell Golden’s recent informal and wide-ranging comments on the “foundational issues” he would like to resolve during his term as FASB chair. Since the FASB has not published the text of Mr. Golden’s remarks, I continue to rely on a Journal of Accountancy article as my sole source for information. Below is what JofA reported about Mr. Golden’s plans for disclosure reform:
“Under current requirements, there may be too many disclosures related to fair value and pensions, and too few disclosures related to income taxes, Golden said. Under FASB’s disclosure framework project, the board’s staff is conducting field-testing to determine whether disclosures can be structured to be more effective in relation to those three topics and others.”
I find it interesting that fair value measurements, pensions and taxes are targeted for change. That’s because they each constitute prime support for the proposition that the more an accounting topic resembles a witch’s brew, “full disclosure” must perforce entail a mind-numbing litany of its ingredients.
Given limitations of time and space, let’s use pension accounting as a case in point. After that, I urge you to contemplate by yourselves the implications for the other topics that Mr. Golden is targeting.
Sans the poetry, here is thumbnail sketch of what the FASB cooked up to be served as pension accounting (for defined benefit plans):
Basis of measurement — The projected benefit obligation requires management to forecast future salaries. This is justified by a questionable notion in the conceptual framework that liabilities should include constructive obligations,* and it unnecessarily complicates measurement of the pension liability.
Discount rate applied to future expected payments — U.S. GAAP requires a risk-adjusted discount rate. But simple logic — which is applied on a regular basis in litigation that I have been associated with as an expert witness — illustrates how a risk-adjusted rate operates to understate the pension obligation. Assume that Jane Doe is illegally terminated by her employer the day before she would have vested in her pension plan. In my experience, the court will reason that if Jane were to be made whole by a single payment, the discount rate used for calculating that payment must be the risk-free rate. That’s because Jane should not have to invest in risky assets to reproduce the series of payments she had been entitled to.
Similarly, the purpose of the discount rate in measuring the pension obligation should be to reduce future expected payments to a single amount that the employer would be obligated to pay if the pension had been terminated on the balance sheet date. Since defined benefit plans require the employer to bear all of the future financial uncertainty, the only appropriate discount rate is the risk-free rate.**
Deferrals of actuarial gains and losses for all kinds of reasons — This is the bulk of the arbitrariness and complexity that makes for pension accounting in accordance with U.S. GAAP. There are earnings deferrals for all sorts of economic events that affect the net position of a pension plan: i.e., changes in interest rates, deviations between actuarial assumptions and actual employee outcomes, deviations from planned investment income.
I hope that the forgoing thumbnail description adequately establishes that pension accounting is a big mess. But, that’s not my primary purpose. The main point is that the right way to fix Mr. Golden’s problems with pension disclosures is to fix the pension accounting problems first. Here’s how:
- Measure the net pension liability/asset as the difference between: (a) the present value (at the risk-free rate) of the expected cash flows based on accumulated vested benefits; and (b) the fair value of pension assets. This number goes on the balance sheet.
- Report the change in the net pension liability/asset for the period on the income statement.
At least to me, it’s self-evident that making pension accounting simple would make the answer to the disclosure question straightforward. First, disaggregate the net pension balance into its asset and liability components. Second, identify all of the significant changes to the pension asset and liability that took place during the period — preferably as a tabular reconciliation of their beginning and ending balances. Third, disclose the measure of the risk-free discount rate used for the current period and for the corresponding period of the previous fiscal year. Fourth, go out for a beer.
Now Comes the Rant
I chose pension accounting pension to illustrate my point, because it is more than just a witch’s brew. It’s a crime.
If FAS 87 (1985) had not been so thoroughly jury rigged to appease the country’s largest employers (and as a by-product to create jobs for accountants and their less gregarious actuary cousins), how many more employees would now be receiving their full pension benefits? How much less dire would be the financial status of hundreds of corporate plans, the Pension Benefits Guarantee Corporation (PBGC), and hundreds of municipal and state pension plans?
One thing I do know is that FAS 87 is one of only a handful of standards to have passed by a one-vote margin. If only one more FASB member had the backbone to push back against the oligarchs and the politicians, how much wealth destruction might have been prevented? I dare say that lousy pension accounting combined with its first cousin, other post-employment benefits, has engendered corporate and governmental value destruction on the order of the cost of the Iraq War.
After the tsunami of corporate pension failures around the time that the dotcom bubble had burst, the FASB promulgated FAS 158 more than 20 years after FAS 87. But, FAS 158 also turned out to be a charade. It was supposed to be the culmination of the first phase of a project to move toward more straightforward, transparent and timely recognition of economic changes to pension plans and other postretirement benefits. Instead, it did little more than to shove the lizard’s legs and all into a stinking cauldron of other comprehensive income. If there ever was a real plan for a second phase of pension accounting reform, it is long forgotten.
Fiddling with pension disclosures is like putting a fresh coat of wax on a car that has a busted cylinder.
Surely, Mr. Golden, you can do better than that.
*Basically, a constructive obligation is defined as a non-legal obligation inferred from business conditions indicating that if a payment would not be made, then the impairment of future business prospects would be greater than the payment saved. While there is some justification to be made on economic grounds, the practical problems of inviting preparers to exercise “judgment” once again trump. By limiting liability recognition to present contractual obligations in general, financial statements will become more reliable and auditable.
**Notice that reducing the measurement of the pension obligation by not accruing future salary increases would be offset to some degree by reducing the discount rate, which would serve to increase the pension obligation.