In responding to my previous post on contingent liabilities, a few of my kind and sensitive readers raised the following points:
Why should there be a difference in the accounting for a law requiring sellers to compensate their customers for defective products, and an express warranty to the exact same effect? (Discussed below.)
Are deferred taxes arising from, say, a higher carrying amount for an asset relative to its tax basis a liability? Why do we have deferred tax accounting in the first place? (Discussed below.)
I incorrectly characterized the provisions of IFRS 3(R) pertaining to contingent liabilities assumed in a business combination. (I made the correction.)
A Legal Requirement versus an Express Warranty
I did not explain myself fully on this point. The power of a government to change its laws creates a possibility that an obligation could be arbitrarily and capriciously eliminated, while stiffing the 'receivable' holder in the process. Moreover, the same merely legal obligation (in the absence of a contractual warranty) could be cancelled by the obligor by dissolving itself.
Are these distinctions between a legal obligation and an express warranty too fine, glib, or heaven forfend, theoretical? Perhaps I could be guilty of mere post hoc rationalization to support my subjective point of view: to wit, the futility of harboring any hope that accounting standards can be effectively crafted to result in reliable and consistent measures of contingent liabilities? But, whatever my conscious or unconscious motivations, the point that all must concede is the impossibility of enunciating an accounting policy that will result in the recognition of all of the liabilities of an entity. So, if you have to draw a line somewhere, I am humbly suggesting where that line can be drawn, in a principled manner, no less.
Deferred Taxes: My Conspiracy Theory
Whatever you may think about whether a deferred tax liability is a legitimate liability, you should know of the conspiracy between issuers and auditors that gave birth to the ersatz accounting theory of "interperiod tax allocation." It is the curtain behind which corporate America tries to hide the reality of just how little in the way of taxes a company might be paying in to government coffers. A systematic earnings reduction in the way of an additional tax expense accrual can be a bitter pill to swallow, but so long as everyone has to take the same proportional hit, nobody actually gets hurt.
The political pressure to modify corporate tax laws in the name of some congress person's pet cause/peeve resulted in a tax code replete with accelerations of deductions, deferrals of payments, exemptions, shelters, loopholes, etc. Post WW II, the corporate tax code quickly got to the point that the actual taxes paid by corporations could be miniscule relative to reported GAAP net income before taxes. The public rightly asked why corporations' 'effective' tax rates were so much lower than the advertised 'statutory' rate. In order to evade directly answering that question, deferred tax accounting was invented to make it appear that corporations were paying their 'fair share' – or at minimum, make it appear that corporations would eventually have to pay their fair share.
Which brings us to the question of whether a "deferred tax liability" is really a liability? Once again, and for the same reasons I gave above, my answer is "no." If a government wants to reduce its tax rate, for whatever reason, the liability goes away. And, companies can control their "deferred taxes" as well. Take, for example, the deferred tax liabilities of a company in dire financial straits, like General Motors. After bondholders are paid ten cents on the dollar in some sort of re-organization, and the deferred tax liabilities on the balance sheet melts away, can the government say it was stiffed? No way.
Another Reason to Exclude Contingent Liabilities
If the FASB and IASB are going to keep tweaking their respective contingent liability standards, and they envision a day when balance sheets will be stated at full fair value, they will eventually have to face up to the fact that there is no reasonable way to measure the fair value of a contingent liability that cannot be transferred. The only approach that comports with reality is to measure the amount it would cost the counterparty to replace (as opposed to transfer) a pattern of cash flows identical to its non-contractual, contingent 'receivable.' But, even that can be so problematic and subjective that I say, 'fuggedaboutit.'