As a matter of policy, the Securities and Exchange Commission disclaims responsibility for any private publication or statement of any SEC employee or Commissioner. This speech expresses the author's views and does not necessarily reflect those of the Commission, the Commissioners, or other members of the staff.
The staff has been asked a variety of questions over the past year on a variety of issues. I would like to share with you our consideration of a number of these issues including those on: (1) asset retirement obligations, (2) accounting for income taxes, (3) the settlement of litigation over the purchase price of a business, and (4) the temporary loss of control of a subsidiary.
A lease may require, or permit the lessor to require, a lessee to return leased property to its original condition at the end of the lease term or to perform some other asset retirement activity.
Some believe that any payment a lessee can be obligated to make is within the scope of Statement of Financial Accounting Standard ("Statement) No. 131. Others read the passage in Statement 13 that says "payments that the lessee is obligated to make . . . in connection with the leased property"2 are within the scope of Statement 13 a bit more narrowly, concluding that, as the asset retirement obligation is attributable to leasehold improvements, or other property, and not the leased property itself, the obligation is most appropriately accounted for under Statement 143.3
In its consideration of comment letters, the FASB declined to provide specific guidance as to whether these types of asset retirement obligations are within the scope of Statement 13 or Statement 143. Consistent therewith, we haven't generally objected to either conclusion so long as it is consistently applied.
Where companies have concluded that such an obligation is within the scope of Statement 13, we do not believe that settlement of that obligation is attributable to a contingent event, such as "damage, extraordinary wear and tear, or excessive usage."4 Nor do we believe that the obligation meets the broader definition of a contingent rental as it results solely from the passage of time and settlement is not based upon "factors on which lease payments are based."5 In other words, we don't believe the retirement obligation should be treated as contingent rent. The lessee should accrue the expected settlement costs over the lease term if it is an operating lease or include the present value of the estimated cost as part of the asset if it is a capital lease.
Where companies have concluded that such an obligation is within the scope of Statement 143, some have relied on paragraph A16 as a basis to not recognize a liability. Paragraph A16 states, in part: "Instances may occur in which insufficient information [is available] to estimate the fair value of an asset retirement obligation." Some have argued that it may not be possible for the lessee to reliably estimate both the range of settlement dates and the probability that the lessor would require the lessee to undertake the asset retirement activity.
With regard to the range of settlement dates, we have objected to a conclusion that it is not estimable. That is not to say that there is only one answer; we have not objected to various different analyses. For example, some believe it is appropriate to include expected renewals while others believe that including renewals would be inappropriate. Probability-weighted assessments might also be used; such an approach might adequately address the uncertainty with regard to timing and allow for consideration of possible renewals as well.
With regard to the probability of enforcement, Statement 143 indicates that conditions attached to obligations should be considered in measurement, but that they do not affect recognition. When assessing the likelihood of the lessor either waiving the contractual provisions or not exercising its option, registrants should consider all available evidence, including evidence that is not necessarily company or industry specific. Consistent with Concept Statement No. 76, the evidence is then weighted according to its reliability.
Paragraph A23 states that "in situations in which the conditional aspect has only 2 outcomes and there is no information about which outcome is more probable, a 50 percent likelihood for each outcome shall be used until additional information is available." Consistent therewith, in an instance where the lessor has an option to require the lessee to satisfy certain asset retirement obligations, absent any information about which outcome is more probable, initial measurement might be based upon a 50% probability of option exercise.
In contrast, an unambiguous obligation - an arrangement that contractually and unambiguously obligates the lessee to perform an asset retirement activity - inherently provides evidence about which outcome is likely. The lessor and lessee bargained for an unambiguous obligation. One might consider paragraph A18, which appears to suggest that an unambiguous obligation requires evidence -- a past history of nonenforcement -- in order to estimate the likelihood of the lessor not enforcing that provision of the arrangement. Absent such evidence, I believe measurement of the obligation should assume performance would be required.
Tax-advantaged transactions continue to draw scrutiny from regulators and investors. As Doug Alkema mentioned last year, one of the reasons the financial reporting of tax-advantaged transactions is not always clear is that the benefit of such transactions is not always recognized in the same period that the transaction is entered into.7 I would like to discuss some considerations that affect when deferring the recognition of tax-advantaged transactions is appropriate.
Statement 109 requires that a valuation allowance be established if it is "more-likely-than-not" that a deferred tax benefit will not be realized. That is, Statement 109 provides guidance for assessing the impairment of recognized deferred tax assets by providing for a valuation allowance when it is more likely than not that there will not be sufficient taxable income of the appropriate timing and character in order to realize the tax benefit. It does not, however, provide guidance as to when the recognition of a current tax benefit is appropriate.
In my view, the recognition of the gross amount of a contingent tax asset, whether or not resulting from a tax-advantaged transaction, should be evaluated for initial recognition like any other asset. That is, the company and auditor should conclude that it is at least probable the deduction will be sustained and the temporary difference will truly exist before that asset is recognized in the company's financial statements. Statement 109 is then used to evaluate any deferred tax asset for impairment.
For example, assume a company enters into a tax-advantaged transaction that results in a $100 permanent difference. Further, the tax opinion received by the company states that the deduction is probable of being sustained. In that situation, it is likely that the company would conclude that the $100 deduction is probable and a reduction in the current payable would be appropriate. In contrast, if the tax opinion received by the company indicated that it was something less than probable that the benefit would be sustained, absent other evidence, I do not understand why it would be appropriate to recognize the benefit as a reduction in income tax expense.
Last year Doug Alkema also mentioned that a company might measure temporary income tax differences by comparing the financial reporting bases of assets and liabilities to their corresponding "as filed" tax bases or their probable tax bases.8 One might draw the conclusion from my discussion of deferred income tax assets that, in general, one should measure temporary income tax differences as the difference between the financial reporting bases and the probable tax bases of assets and liabilities. Such an approach is consistent with the one outlined by the FASB staff in Question 17 of their Guide to the Implementation of Statement 109.9 The consequence of such a conclusion is that it would then be necessary to provide for a contingent income tax liability measured as the difference between the probable tax bases and the as-filed tax bases.
For example, assume that for tax purposes a company expenses certain property related costs, but believes it is "probable" the Internal Revenue Service will require that cost to be capitalized. The probable tax basis of property is different from the as-filed tax basis. Consistent with my earlier discussion, the company would measure the temporary difference as the difference between financial reporting and probable tax basis of the property and provide for a contingent tax liability, under Statement 5, measured as the tax effect of the difference between the probable tax basis and as-filed tax basis.
If a provision for contingent income tax liabilities is recognized, it should not be included with deferred income tax liabilities or as part of any valuation allowance.
We understand that, despite the guidance in Question 17 of the Guide to the Implementation of Statement 109, some companies simply measure temporary differences as the entire difference between the financial reporting and as filed tax bases. In this case, there is no additional contingent tax liability to recognize as it would already be part of the deferred income tax liability. While we have not considered specifically the basis for this choice of accounting method, we believe that, if this approach is applied, the disclosure of the components of deferred income tax liabilities should be consistent from period to period.
Paragraph B177 of Statement 14110 indicates that contingencies arising from a business combination, "are the acquiring enterprise's contingencies . . . . Accordingly, Statement 16 applies to those contingencies after the initial purchase allocation." Said differently, contingencies arising from a business combination are not preacquisition contingencies. Accordingly, we have generally concluded that legal claims between an acquirer and the former owners of an acquired business should be reflected in the income statement when settled.
Instances in which we have been persuaded that a settlement of litigation over a purchase is more appropriately reflected as an adjustment to the cost of the acquired business demonstrate a clear and direct link to the purchase price. For example, litigation seeking enforcement of an escrow or escrow-like arrangement, say, specifying a minimum amount of working capital in the acquired business, may establish a clear and direct link to the purchase price.
Frequently, claims seeking enforcement of an escrow or escrow-like arrangement also include claims of misrepresentation or otherwise constitute a mixed claim. While more complicated, I believe that the concept is the same. In order to reflect some or all of the settlement of such a claim as an adjustment of the purchase price of the acquired business, the acquirer should be able to persuasively demonstrate that all or a specifically identified portion of the mixed claim is clearly and directly linked to the purchase price. I would not ordinarily expect that the settlement of litigation that does not indicate that the initial allocation of fair value was incorrect to be accounted for as an adjustment of the purchase price.
Similarly, claims that assert one party mislead the other or that a provision of the agreement is unclear are not unique to business combination agreements and do not generally establish a clear and direct link to the purchase price and, therefore, should be reflected in the income statement. I say "generally fail to establish a clear and direct link" only because I think we would be persuaded that an adjustment of purchase price is appropriate if the acquirer were able to objectively demonstrate that that the purchase price exceeded the acquired business's agreed-upon fair value at the time of the acquisition. For example, assume a purchase agreement explicitly sets forth the understanding that each "acquired customer" is worth $1,000, that not less than one thousand customers will be transferred as of the consummation date, and subsequent litigation determines that the actual number of acquired customers was only nine hundred. The effects of the litigation should properly be reflected as part of the purchase price. In contrast, if the purchase agreement obligates the seller to affect its best efforts to retain customers through the consummation date and litigation subsequently determines that the seller failed to do so, the effects are not clearly and directly linked to the purchase price and, accordingly, should be reflected in the income statement.
Assuming that an adjustment of the purchase price of the acquired business is appropriate, the adjustment should be reflected in a manner that is consistent with the original purchase price allocation and subsequent reporting. For example, if an escrow arrangement specified a minimum net realizable value for acquired accounts receivable, the results of the operation of that escrow arrangement should first affect the accounts receivable balance and, if those receivables have already been written-off subsequent to the acquisition, the adjustment to purchase price should similarly be reflected in the current period income statement.
I have thus far discussed only the effects of litigation over purchase price that involve the former owners of the acquired business. I think it goes without saying that the cost of litigation brought by the acquirer's shareholders should always be reflected in the income statement.
Paragraph 13 of Statement 9411 indicates that "a majority owned subsidiary shall not be consolidated if control does not rest with the majority owner (as, for instance, if the subsidiary is in legal reorganization or in bankruptcy . . .)." I think most would conclude that bankruptcy is indicative of a loss of control and that deconsolidation is appropriate. However, paragraph 32 of SOP 90-712 suggests that there are conditions in which the continued consolidation of a subsidiary in bankruptcy is appropriate.
Recently, we were asked to consider whether the deconsolidation of a majority-owned subsidiary in bankruptcy was appropriate. We were willing to undertake such a consideration because, in part, we believe that, even when a subsidiary is in bankruptcy, there are circumstances where the continued consolidation of a subsidiary is more meaningful. For example, consider an instance where the parent has a negative investment, expects the bankruptcy to be brief, and expects further to regain control of the subsidiary. One might be appropriately concerned about the deconsolidation, recognition of a gain, and reconsolidation of a subsidiary by a parent in a short period of time.
In the fact pattern we considered, the parent was the majority common shareholder, a priority debt holder, and the subsidiary's single largest creditor. Due to its creditor position, the parent was able to negotiate a prepackaged bankruptcy with the subsidiary's other creditors. The parent, pursuant to the terms of the prepackaged bankruptcy, expected to maintain majority-voting control after the bankruptcy. The parent also expected the bankruptcy to be completed in less than one year.
While we are inclined to continue to believe that bankruptcy is indicative of the fact that control does not rest with the majority owner, we did not object to the parent's determination that the continued consolidation of its subsidiary during bankruptcy was more meaningful and that any loss of control would be temporary given the facts and circumstances.
Obviously, a determination that continued consolidation of a subsidiary in bankruptcy is appropriate requires a fairly unique set of facts and is appropriate only in infrequent and uncommon circumstances. It is not a conclusion that a registrant should make without thoroughly consulting with its auditors and one the company should consider discussing with us. In any event, the conclusion and its basis should be adequately disclosed and the company should periodically reassess its facts and circumstances to confirm the appropriateness of such a determination.