Our regular readers know that our shtick is criticizing generally accepted accounting principles, as well as those who create and implement them without fully anticipating their negative impact on the quality of the resulting financial statements. This time we're going after accounting for income tax effects, which has roots dating back to 1967 and APB Opinion 11. Even though the Financial Accounting Standards Board tried to reform tax accounting with SFAS 109 20 years ago, we don't see much actual improvement.
Why raise this topic now? It's simple: Congress is full of people who want to either increase taxes or roll them back. Both think their plan is better, but no one knows, partly because generally accepted accounting principles for taxes are worse than woeful. In short, if tax policy is to be rational (and if the capital markets are to work efficiently), it's essential to know who pays how much tax and when. A total makeover is needed because no one can get those answers from today's financial statements.
We begin by reconsidering three premises behind GAAP for taxes that few accountants or others ever question because the status quo is so well entrenched.
ISSUE 1: ARE INCOME TAXES EXPENSES?
FASB's Conceptual Framework defines expenses as "outflows or other using up of assets or incurrences of liabilities ... from delivering or producing goods, rendering services, or carrying out other activities that constitute the entity's ongoing major or central operations." We suggest income taxes are not expenses under this definition because they aren't essential for these activities. Unlike pre-tax expenses that are incurred to make profits, taxes are incurred only if, and after, the company has already made them. The events flow like this: Expenses ->Revenue -> Profits -> Income Taxes.
In effect, taxes are involuntary dividends to, or profits confiscated by, the government. If this view prevailed, the accounting debate would be over and management would debit retained earnings for taxes actually paid or payable (TAPOP). Although this is consistent with the truth, it hasn't been popular.
ISSUE 2: IF TAXES ARE EXPENSES, WHEN SHOULD THEY BE REPORTED?
If the first issue is ducked and taxes are treated as expenses, then accountants face the issue of when to report them. Because tax expense results from the period's taxable profits without directly creating revenue, it follows that TAPOP should flow through to expense in the period the profits are taxed.
Already we hear anguished cries from status quo defenders because they're conditioned to "normalize" reported earnings. They think they're not merely providers of information but shapers of its content. They falsely assume that they're helping users by smoothing earnings.
In fact, they're doing the users no favor at all because smoothing only soothes managers' volatilaphobia (their fear of volatile reported income). Further, the last thing managers want users to know is exactly how little they actually pay because of deferred tax payments. (Consider the flap created by General Electric's 7.4 percent effective tax rate for 2010 as reported by The New York Times in March.)
The misleading result is artificially smooth reported tax expense that correlates with pre-tax income, not TAPOP.
ISSUE 3: IF TAXES ARE EXPENSES, HOW BIG ARE THEY?
Smoothing uses deferred assets and liabilities to account for temporary tax differences. With disastrous results, FASB concealed the real cost by not discounting these deferred balance sheet items to their present values, thereby ignoring how much is lost by paying early and gained by paying later.
Here's what SFAS 109 says: "Implementation issues include selection of the discount rate(s) and determination of the future years in which amounts will become taxable or deductible. The board decided not to consider those issues at this time."
Translation: "Discounting would be too hard and controversial." It's as if they didn't realize that every tax-planning CPA already knows how to do it.
Not discounting produces profoundly unreliable measures of tax expense and deferred tax assets and liabilities in at least four ways.
CONCEALING THE TRUTH
Two axioms permeate tax planning: Never pay taxes early and always postpone paying them, unless income bracket changes are involved. But what happens without discounting?
If taxes are paid before GAAP profits are reported, management records a deferred tax asset equal to what will eventually be saved when a deductible temporary difference reverses. Without discounting, the same measure is reported whether the reversal is one, 10 or 25 years away. If the savings were discounted, the asset would be initially smaller and the tax expense larger, thus revealing the negative impact of paying taxes early. On the other hand, if tax payments are postponed, management reports a deferred tax liability for the sum of the future payments, even if reversal won't occur until one, 10 or 25 years later. Therefore, the reported expense is grossly overstated. Discounting would reveal the benefit of postponed tax payments.
Clearly, the truth is not faithfully represented.
GOODWILL GONE BAD
When a company is acquired, the buyer assigns goodwill a cost equal to the excess of the price (cash, stock and assumed liabilities) over the acquired assets' fair value. Because the buyer puts the seller's non-discounted deferred tax liabilities on its post-acquisition balance sheet, goodwill gets pumped up because of FASB's reluctance to report the truth. This situation should lead to an immediate impairment but it doesn't.
RATE CHANGES
Assuming only for argument's sake that deferred assets and liabilities exist, we endorse SFAS 109's immediate recognition of changes in their book values caused by newly enacted rates. It makes sense that future rate changes impact taxes to be saved or paid, such that a higher rate increases deferred assets and liabilities, and a lower rate decreases them.
What FASB did with this idea, however, is questionable. Specifically, they decided to immediately modify tax expense in the year that Congress acts.
Suppose a company has a $200 million taxable temporary difference at the end of 2011 when Congress increases the rate for 2012 and beyond by 4 percent. When the difference eventually reverses, the company will pay an additional $8 million. Under SFAS 109, however, the taxpayer's 2011 tax expense is immediately increased by the full non-discounted $8 million, even though the new rate does not apply to 2011 income. This practice clearly produces a flawed measure of operating income in the year that Congress acts.
A far better solution would report rate change effects as gains or losses, instead of folding them into the ongoing expense.
CLASSIFICATION
One reason for FASB's refusal to discount was the difficulty of applying the information technology of the early 1990s to figure out when temporary differences reverse. Besides grossly overstated measurements, this practical compromise led to two non-useful anomalies in balance sheet classifications.
One is that deferred tax assets and liabilities are classified as current (or non-current) if the temporary difference is associated with a current (or non-current) asset or liability, not when the taxes will be saved or paid. A second is that deferred tax assets and liabilities are offset. Thus, the balance sheet doesn't provide useful information about the timing of future cash flows that would allow users to attempt their own discounted estimates.
ONE MORE FLAW: OPERATING CASH FLOWS
FASB produced a defect in the cash-flow statement when SFAS 95 unjustifiably abandoned the concept behind intra-period tax allocation that associates tax consequences with the income categories that produce them. Specifically, all income tax payments are reported as operating cash outflows, even if they were generated by financing and investing activities.
In addition, the indirect method produces an even more abstruse operating section that includes changes in the deferred tax accounts among the myriad adjustments for non-cash expenses.
A BETTER SOLUTION, PART I
We agree that the goal behind deferred tax accounting is explaining the future consequences of differences between GAAP and taxable income. In pursuit of that goal, accountants have vainly tried for 44 years to signal these differences through the accounts. Unfortunately, today's statements hide more than they reveal because the crude and otherwise flawed deferred approach produces indecipherable messages.
We offer a simple and superior solution. Instead of trying (and failing) to use financial statements to convey complex measures and providing footnotes to explain how much tax was actually paid or owed by a company, we propose the opposite approach that makes the income statement understandable by reporting expense equal to TAPOP and invites management to explain the outcome and predict future tax consequences in the notes.
This method will also clearly reveal the actual volatility of after-tax earnings and cash flows, while providing straightforward public information about the real tax rate.
A BETTER SOLUTION, PART II
Going a giant step further, we invite new broad debate on the fundamental issue of whether taxes are expenses, or confiscations of profits that management is fully justified in minimizing.
Here's our thinking: The longstanding accounting solution conveys that it's perfectly suitable for state and federal governments to tax artificial entities, instead of their owners (who end up paying a double tax on dividends).
This system is not only patently unfair but also imposes a huge social cost in writing, interpreting and enforcing the laws, as well as tax planning to minimize them. If there were no corporate tax, those costs would go away and selling prices of corporate-produced goods and services would be forced down, thus benefiting most everyone.
In effect, Congress would stop taxing fictitious persons and impose taxes on real individuals who own corporations. The fact that the corporate tax produces only 6 percent of federal receipts (according to the Times article) shows that it is hardly worth the trouble.
Is repealing this tax far-out, unrealistic thinking? Of course, but it's much more realistic than financial statements produced under today's GAAP. It is clearly time for substantial makeovers, in both accounting and tax policies.
Paul B. W. Miller is a professor at the University of Colorado at Colorado Springs and Paul R. Bahnson is a professor at Boise State University. The authors’ views are not necessarily those of their institutions. Reach them at paulandpaul@qfr.biz.
