Last June, the IASB issued revisions to IAS 19, Employee Benefits. The accompanying press release would have us believe that the five-year project resulting in incremental changes to the accounting for pension costs are a significant improvement; unfortunately, they are just one small step forward followed by a giant step backward.*
The small step forward is the elimination of the alternative approaches for dealing with actuarial gains and losses – i.e., differences between the expected change to the net pension liability (after taking into account contributions and payouts) and the actual changes. The older version of IAS 19 permits a free choice among no less than three treatments:
- Similar to U.S. GAAP (which is also highly deficient in this regard), amortization of 'excess' cumulative gains/losses;
- Immediate recognition of the excess cumulative gains/losses; or
- As per UK accounting standards, immediate recognition of the total gains/losses for each period in other comprehensive income (OCI), without recycling.
The giant step backward is, of course, the new treatment, which gifts issuers with a new set of tools for managing earnings. Under the new accounting, only management's (biased) forecast of the change to the net pension liability is included in net income, and the actuarial gains and losses go to OCI – without recycling.
It's easy to see what is going to happen after companies adopt IAS 19. First, a manager can rest comfortably in the knowledge that, even if the sky falls on the company's pension plan, it won't affect hitting any of their earnings targets – and their earnings-based bonuses – ever. All variability will be forever sequestered in accumulated OCI (AOCI).
Second, and relatedly, management will be much less concerned with actual pension cash flows, especially those cash flows that won't take place until long after current management has departed. A limitation on the amount of income from pension plan assets** could mean that management will care less how those assets are invested; and the plans themselves will become more liberal in ways that wouldn't exist if the total change in the present value of cash flows were fully reflected in net income.
Third, estimates of the future costs that actually find their way into net income will be inappropriately low. Will 'independent' auditors feel any compunction to correct management's naturally occurring bias? I'm not betting on it, and especially not without mandatory audit firm rotation.
OCI Has Become the IASB's Kitchen Sink
The new pension accounting treatment is apparently based on a preposterous premise: that net income for a period is supposed to reflect "the underlying financial performance of the core business" or "an entity's day-to-day operations."
I say preposterous because, although it is certainly an appealing thought, it is perforce not possible for net income to reflect "underlying financial performance" or "an entity's day-to-day operations." To take just a few examples:
- Gains/losses from dispositions of PP&E, as well as impairment charges, largely reflect events that occurred in prior periods. Would the IASB suggest that these items, too, be thrown into the kitchen sink of OCI? Of course not.
- Some events are recognized in net income only after their cash flow effect meets a measurement threshold, like "probable" or "capable of reliable measurement." These terms make me gag, but again, this means that much of the amount recognized in the current period is due to events that occurred in prior periods.
- Some items in accumulated OCI, such as translation adjustments from prior periods, are recycled from AOCI to net income in the current period when certain criteria are met. Yet again, this is a conflation of current and prior period events.
For each of these, and more, IFRS would permit additional line items; and additional subtotals may also be added to the income statement, in order to assist the user in assessing performance for the period. The same could apply to pension costs: if actuarial gain/loss components merit separate disclosure, nothing should prevent a company from separately captioning them within the income statement; and quantitative disclosures as notes accompanied by explanatory paragraphs would also do the trick.
But, none of these treatments would have accomplished the political objective of the IASB, which was to gift managers a new tool for managing their reported net income.
Another Standard for No Good (Stated) Reason
I say apparently based on a preposterous premise because my source for the above quotations is from that little press release I mentioned earlier. The official standard is devoid of any remotely similar language or any other reasonably detailed discourse answering the 'why' questions that must be integral to the "basis for conclusions" of any new standard. This is not new, by the way. See my post on the revisions made to IAS 23, Borrowing Costs.
If you want to see the problem for yourself, go to paragraphs BC3 – BC11 of the revised standard. While actual words are there to be read and understood, they amount to little more than boilerplate, addressing 'what' and 'when' types of questions.**
The unavoidable conclusion is that there is no defensible logic to the new accounting treatment for pension costs. More generally, OCI has become the kitchen sink in which to toss anything that management feels would despoil its precious 'bottom line.'
SEC, Are You Paying Attention?
Whether or not one supports the actual revisions, there can be no denying that a section of an accounting standard labeled "Basis for Conclusions" creates an expectation that the IASB will explain why the accounting treatment chosen is the best among a reasonable set of alternatives. To be as generous as possible, the IASB's explanations are vague and misleading. But, to be brutally honest, the IASB has made a mockery of due process on this and any other occasion where it suits its political agenda.
If the SEC is still thinking about looking to the IASB as its principle accounting standards setter, with the FASB following in lockstep in a so-called "condorsement" role, this is just one more indication of the folly of that strategy. As much as it has tried, the SEC cannot blithely presume that IASB-promulgated standards are high quality, but only that a final standard ultimately emerged out of a five-year political football match.
*On September 9, 2011, Patricia Walters started a discussion of the revisions to IAS 19 on the AECM listserv. Some of the comments to that thread have influenced my own views.
**The new standard also creates a new approach for measuring the expected return on plan assets, which I completely disagree with (see paragraph 125). I may discuss this in a later post.
***To be fair, there is this statement further down in the document (paragraph BC88):
"The Board noted that although changes included in the remeasurements component [i.e., the actuarial gains/losses] may [italics supplied] provide more information about the uncertainty and risk of future cash flows, they provide less information about the likely amount and timing of those cash flows."
I think this is inadequate for reasons that are self evident. Substitute "does not" for the weasel word, "may", in the above quote, and you have a much more realistic assessment.