Should deferred tax be abandoned, basically?
It’s not uncommon of course to come across complaints that the concepts underlying deferred tax balances are hard to follow, that the resulting assets and liabilities don’t mean much to users, and so forth. I suppose I’ve always taken a fairly standard view towards that issue, as I put it here:
- Indeed, it seems unlikely that many users can obtain much specific value from the numbers attaching to deferred tax or from what’s disclosed about them: the main value of the exercise, it seems to me, is really in avoiding the distortion that would result if the accounting for taxes were based solely on what’s payable for any particular period (whereby, for example, an entity’s reported tax rate might fluctuate materially based on what proportion of its expenditures from one period to the next is subject to accelerated treatment for tax purposes, regardless that all profits are ultimately taxable at the same rate). To that perhaps necessary but not inherently very exciting end, I expect the majority of us should be truly grateful to those brave practitioners who spend their lives immersing themselves in the technicalities of IAS 12, so that the rest of us don’t have to.
I’ve started to wonder lately though whether the whole mechanism is indeed necessary. As one of the big four firms has long pointed out in its guide to IFRS:
- (Some commentators) argue that the tax authorities impose a single annual tax assessment on the entity based on its profits as determined for tax purposes, not on accounting profits. That assessment is the entity’s only liability to tax for that period, and any tax to be assessed in future years is not a present liability as defined in the IASB’s Conceptual Framework. Supporters of (this view) acknowledge the distortive effect of (transactions for which the pattern of recognition for tax purposes significantly differs from that for accounting purposes) but argue that this is better remedied by disclosure than by creating what they see as an ‘imaginary’ liability for deferred tax.
The virtues of the “imaginary liability” are perhaps best appreciated on a micro-level. If an entity had only a single asset for which the tax impact all occurred in year 1, but the accounting depreciation went on for a few years beyond that, it might seem worthwhile going through the deferred tax hoops to avoid any prospect of investor over-reaction to the initial low effective tax rate. But of course, deferred tax balances are an aggregate of a potentially vast number of transactions, a fact which blurs their comprehensibility and, perhaps as importantly, reduces the likelihood of an overall reversal ever occurring. This is Tom Selling in The Accounting Onion (writing in the context of US GAAP):
- Twenty years ago, I conducted, but did not attempt to publish, an informal test of the proposition that reported current rates would be persistently lower than effective rates. I analyzed the public companies in the Compustat PCPlus database for the years 1977 to 1996 that met the following criteria:
- Calendar fiscal year-end
- S. company in an industry other than financial services
- Current and deferred income taxes are reported for all 20 years.
- I calculated the effective and current tax rates by year for each of the 546 companies that met these criteria….If the effective tax rate were a good long-run measure of taxes paid, we would expect to see that median current tax rates would exceed effective rates roughly 10 times of the 20 years studied. It didn’t happen even once.
“Yet,” observes Selling, “interperiod tax allocation appears to have one practical, albeit cynical, use: to obscure how little income taxes corporations are paying relative to statutory rates”.
If one focuses on the deferred tax liability itself, it’s plain how it fails to reflect the concepts and methods that underlie other balance sheet liabilities. Most prominently, deferred tax balances aren’t discounted in large part because, as IAS 12.54 puts it, the work required to do that reliably would in many cases be “impracticable or highly complex.” Yet, since the whole thing’s highly complex anyway, you might think even a highly estimation-dependent approach to applying the time value of money would be preferable to ignoring it. A few years ago, Dave Walters on the PWC blog made this same observation, and went on:
- Should we stop recording book/tax differences in the balance sheet? Why not require clear disclosure of the expected future cash tax rates over a reasonable time horizon and significant judgments or uncertainties around those assumptions rather than an incomprehensible number in the balance sheet?
- Companies could disclose their predicted sustainable cash tax rate based on current tax rates and profits and the sensitivities in that estimate. This might help investors assess performance and value with less complexity. I know it’s not an optimal solution – but in the world of deferred taxes I believe we are looking for the “least worst” solution. What we have today doesn’t quite meet that test in my view.
Of course, companies could disclose all those things without any changes to the current accounting model (I wrote on some related matters here), and one hopes more of them will start doing so. Despite all the problems with IAS 12, there doesn’t appear to be an active clamour for change, and it’s a bit hard to imagine the IASB embracing something for which “least worst” might be as buoyant as the labeling gets. On the other hand, if the topic was ever opened for discussion, would the current approach have more than a handful of truly passionate, persuasive defenders…?
The opinions expressed are solely those of the author