Menu
Tax Notes logo

NYSBA Tax Section Suggests Guidance on Contingent Liabilities for COD Income Purposes

NOV. 20, 2012

NYSBA Tax Section Suggests Guidance on Contingent Liabilities for COD Income Purposes

DATED NOV. 20, 2012
DOCUMENT ATTRIBUTES
  • Authors
    Needham, Andrew W.
  • Institutional Authors
    New York State Bar Association Tax Section
  • Cross-Reference
    For the Tax Court opinion in Merkel v. Commissioner, 109 T.C.

    463 (1997), aff'd, 192 F.3d 844 (9th Cir. 1999), see Doc

    1999-30468
    or 1999 TNT 182-7 1999 TNT 182-7: Court Opinions.
  • Code Sections
  • Subject Area/Tax Topics
  • Jurisdictions
  • Language
    English
  • Tax Analysts Document Number
    Doc 2012-23860
  • Tax Analysts Electronic Citation
    2012 TNT 225-17

 

November 20, 2012

 

 

The Honorable Mark Mazur

 

Assistant Secretary (Tax Policy)

 

Department of the Treasury

 

1500 Pennsylvania Avenue, NW

 

Washington, DC 20220

 

 

The Honorable William J. Wilkins

 

Chief Counsel

 

Internal Revenue Service

 

1111 Constitution Avenue, NW

 

Washington, DC 20224

 

 

Steven Miller

 

Acting Commissioner

 

Internal Revenue Service

 

1111 Constitution Avenue, NW

 

Washington, DC 20224

 

Re: Report on Insolvency under Section 108: the Treatment of Contingent Liabilities

 

Dear Messrs. Mazur, Miller and Wilkins:

I am pleased to submit the attached report of the Tax Section of the New York State Bar Association, which provides recommendations for additional guidance regarding the treatment of contingent liabilities in the determination of insolvency for purposes of Section 108(a)(1)(B).

By way of background, when a taxpayer's debt is forgiven in whole or in part, Section 61(a)(12) requires that the taxpayer recognize gross income from cancellation of indebtedness. Under the "insolvency exception" of Section 108(a)(1)(B), however, income from cancellation of indebtedness is excludible from gross income to the extent that a taxpayer is insolvent. Whether a taxpayer is insolvent, and the extent to which a taxpayer is insolvent, is determined by calculating the excess of the taxpayer's liabilities over the fair market value of the taxpayer's assets immediately before the debt discharge.

Neither the Code nor the Treasury Regulations provide any guidance regarding the treatment of contingent liabilities for purposes of the insolvency exception. In Merkel v. Commissioner, the Tax Court concluded that contingent liabilities may be included in the insolvency determination, but only if "it is more probable than not that [the taxpayer] will be called upon to pay that obligation in the amount claimed. . . ." The Ninth Circuit affirmed.

Because we believe this "all-or-nothing" approach is inherently flawed, we do not recommend that Treasury and the IRS adopt the Merkel approach in future guidance. Under Merkel, a taxpayer with a contingent liability that has a 49% probability of being paid is not permitted to include any portion of the contingent liability in the insolvency determination, while a taxpayer with a contingent liability that has a 51% probability of being paid may include the entire liability. In both cases, the result does not accurately reflect the true economic impact of a contingent liability on a taxpayer's net worth.

We therefore recommend that contingent liabilities (without regard to whether they are "more probable than not") should generally be included in the insolvency determination under Section 108(d)(3), but not in excess of their actual fair market values, as determined on the same basis as the valuation of contingent liabilities under Treas. Reg. § 1.752-7(b)(3)(ii). To minimize the complexity associated with any future guidance that adopts this approach and to ensure that that the insolvency exception of Section 108(a)(1)(B) remains a viable alternative to filing for bankruptcy protection, we also recommend that contingent liabilities that are remote (e.g., a less than 20% probability of payment) should be ignored in the insolvency determination.

Finally, we recommend an optional safe harbor for valuing contingent liabilities in the insolvency determination in the case of corporate taxpayers that discharge debt by issuing stock to creditors, which we believe will simplify the determination in many out-of-court debt restructurings.

The report also includes a number of other technical comments and recommendations.

Respectfully Submitted,

 

 

Andrew W. Needham

 

Chair

 

New York State Bar Association

 

Tax Section

 

Enclosures

 

 

cc:

 

William D. Alexander

 

Associate Chief Counsel (Corporate)

 

Internal Revenue Service

 

 

Stephen R. Cleary

 

Office of Associate Chief Counsel (Corporate)

 

Internal Revenue Service

 

 

Erik Corwin

 

Deputy Chief Counsel (Technical)

 

Internal Revenue Service

 

 

Lee A. Kelley

 

Senior Counsel (Tax Policy)

 

Department of the Treasury

 

 

Andrew J. Keyso

 

Associate Chief Counsel (Income & Tax Accounting)

 

Internal Revenue Service

 

 

Emily S. McMahon

 

Deputy Assistant Secretary (Tax Policy)

 

Department of the Treasury

 

 

Rubin B. Ranat

 

Office of Associate Chief Counsel (Corporate)

 

Internal Revenue Service

 

 

Lon B. Smith

 

National Counsel to the Chief Counsel for Special Projects

 

Internal Revenue Service

 

 

Lisa Zarlenga

 

Tax Legislative Counsel (Regulatory Affairs)

 

Department of the Treasury

 

* * * * *

 

 

NEW YORK STATE BAR ASSOCIATION TAX SECTION

 

REPORT ON

 

INSOLVENCY UNDER SECTION 108:

 

THE TREATMENT OF CONTINGENT LIABILITIES

 

 

November 20, 2012

 

 

New York State Bar Association Tax Section

 

Insolvency under Section 108: the Treatment of Contingent

 

Liabilities

 

 

This Report1 provides recommendations for additional guidance with respect to the treatment of contingent liabilities in determining whether a taxpayer is "insolvent" for purposes of Section 108(a)(1)(B) of the Code.2

 

I. Background: COD Income and the Insolvency Exception.

 

When a taxpayer's debt is forgiven in whole or in part, Section 61(a)(12) requires that the taxpayer recognize gross income from cancellation of indebtedness ("COD Income"). Section 61(a)(12) codifies the rule established by the Supreme Court in United States v. Kirby Lumber Co.3 that a taxpayer recognizes taxable income when it discharges its debt at a discount. In Kirby Lumber, the Supreme Court held that a corporation that repurchased some of its outstanding bonds (which had been previously issued at par) for less than their par value "realized . . . an accession to income" because the repurchase "made available . . . assets previously offset by the obligation of bonds now extinct."4

This justification for taxing COD Income came to be known as the "freeing of assets" theory. While the law on COD Income subsequently evolved to include other discharges that did not result in a freeing of assets, this theory became the foundation for the insolvency exception to the recognition of taxable COD Income now codified under Section 108(a)(1)(B) (the "Insolvency Exception").

 

A. The History of the Insolvency Exception.

 

In the wake of Kirby Lumber, courts created a judicial exception to the recognition of COD Income for insolvent debtors outside of bankruptcy. In Dallas Transfer and Terminal Warehouse Co. v. Commissioner,5 an insolvent taxpayer was relieved of debt, but remained insolvent after the discharge. The Fifth Circuit ruled that a debtor who was insolvent before the discharge of a debt, and who remained insolvent after the discharge, should not be taxed on the COD Income. The rationale for this exception was that "[t]axable income is not acquired by a transaction which does not result in the taxpayer getting or having anything he did not have before."6 In other words, no assets had been freed up for the benefit of the taxpayer by the discharge.7

Shortly thereafter, this doctrine was crystallized in Lakeland Grocery Co. v. Commissioner,8 in which the Board of Tax Appeals established the so-called "net assets test". In Lakeland Grocery, an insolvent taxpayer paid $15,473 to its creditors in discharge of $104,710 of debt. While the taxpayer was insolvent before the discharge, it became solvent after the discharge by $39,597. The court ruled that the taxpayer realized COD Income, but only to the extent it became solvent after the discharge.9 The "net assets test" established in Lakeland Grocery therefore required an examination of the debtor's net assets after a discharge of debt -- if the gross value of the assets of the debtor exceeded its liabilities after the discharge, the debtor realized COD Income in an amount equal to the excess value "freed from the claims of creditors";10 if the discharge merely reduced the insolvency of the debtor, the debtor did not realize COD Income at all.

In 1980, Congress codified the Insolvency Exception.11 The Insolvency Exception allows for the exclusion of COD Income from gross income to the extent that the taxpayer is insolvent.12 With respect to any debt discharge, whether the taxpayer is insolvent, and the amount by which the taxpayer is insolvent, is determined by calculating the excess of the taxpayer's liabilities over the fair market value of the taxpayer's assets immediately before the discharge.13 The legislative history to Section 108 explains that the purpose of the Insolvency Exception is to relieve insolvent taxpayers from the burden of an immediate tax liability in respect of COD Income, which would otherwise deny the taxpayer the "fresh start" that the debt discharge was intended to allow.14

There is no guidance in the Code, Treasury Regulations or legislative history that defines "liability" for purposes of the Insolvency Exception. Nor is there any guidance on the impact of contingent liabilities in the insolvency determination.

 

B. Guidance Under the Insolvency Exception Prior to the Merkel Case.

 

In the years between the establishment of the "net assets test" in Lakeland Grocery and the codification of the Insolvency Exception under Section 108, a few court decisions appeared to permit the inclusion of contingent liabilities in the insolvency determination.

In Conestoga Transportation Co. v. Commissioner,15 for example, the taxpayer purchased its own debt obligations at less than face value, resulting in COD Income. The taxpayer claimed that it was insolvent at the time of the purchase, due in part to accrued reserves for contingent claims listed as liabilities on its balance sheet. The IRS did not contest the inclusion of the contingent claims as liabilities, focusing instead on whether the valuation of certain intangible assets was greater than represented.16 The Tax Court ruled that the taxpayer was insolvent and could therefore exclude the COD Income, but without providing any analysis of the treatment of contingent liabilities for this purpose.17

However, the codified Insolvency Exception states that "[e]xcept as otherwise provided in this section, there shall be no insolvency exception" from the general rule of inclusion of COD Income in gross income.18 Because it appears that Congress intended to preempt the common law in this area, it is not entirely clear to what extent pre-1980 guidance has any remaining precedential value.

In a private letter ruling issued after the Insolvency Exception was codified, the IRS took the position that contingent liabilities, in this case contested estate and income taxes, should not be included in the insolvency determination.19 In this ruling, despite citing Conestoga Transportation, the IRS stated that "there is no authority" to support the inclusion of contingent or contested liabilities in the insolvency calculation. In support of its position, the IRS cited a Supreme Court decision, United States v. Consolidated Edison Co.,20 where the Supreme Court held that a taxpayer could not claim a deduction for contested real estate taxes until a final court order determining liability was entered.

In Revenue Ruling 92-53,21 the IRS considered whether the amount by which a nonrecourse debt exceeds the fair market value of the property securing the debt ("Excess Nonrecourse Debt") should be taken into account in the insolvency calculation. The IRS ruled that Excess Nonrecourse Debt should be counted in its entirety as a liability for purposes of measuring insolvency under Section 108, but only if the nonrecourse debt itself was being discharged. If debt other than the nonrecourse debt was being discharged, the nonrecourse debt should be counted only to the extent of the fair market value of the property securing the debt. In distinguishing the two situations, the IRS cited the policy underlying the Insolvency Exception, which is not to impose current tax on a taxpayer that is "unable to pay either the indebtedness or the tax," reasoning that counting Excess Nonrecourse Debt that was itself being discharged was necessary to protect a taxpayer who lacked the ability to pay the tax.

Finally, in a heavily redacted 1997 Field Service Advice, the IRS considered whether a taxpayer's potential obligation under a guarantee should be included in the insolvency determination. The IRS noted that "[t]here is no authority which addresses the issue of whether guarantees should be included or excluded from the insolvency calculation" and concluded that "the better argument is for [taxpayer] to treat the guarantee as a liability only to the extent it is likely to have to pay such debt (or any portion thereof), i.e., only to the extent of its value."22

 

C. The Merkel Case.

 

In Merkel v. Commissioner, the Tax Court, and later the Ninth Circuit, considered the treatment of contingent liabilities for purposes of the Insolvency Exception.23 Dudley and La Donna Merkel and David and Nancy Hepburn were general partners in a partnership, HMH Partners ("HMH"), and each couple owned 25% of HMH. In 1991, a bank forgave a $1,439,000 nonrecourse note of HMH, resulting in COD Income of $359,000 to both the Merkels and the Hepburns. The Merkels and the Hepburns reported the COD Income on their joint tax returns, but excluded it under Section 108(a)(1)(B) based on their claimed insolvency.

Dudley Merkel and David Hepburn were also officers and co-owners of Systems Leasing Corp. ("SLC"). SLC obtained a bank loan that had an unpaid balance in excess of $3.1 million, which Merkel and Hepburn had personally guaranteed. SLC defaulted on the loan and entered into a structured workout agreement with the bank pursuant to which SLC agreed to pay $1.1 million to the bank in exchange for the bank's agreement to discharge the remaining balance and release Merkel and Hepburn as guarantors. The bank accepted the workout agreement on the condition that none of SLC, Merkel or Hepburn filed for bankruptcy within 400 days. After the 400-day period elapsed without a bankruptcy filing, Merkel and Hepburn were relieved of their obligations under the guarantees. As officers of SLC, Merkel and Hepburn were also secondary obligors on a state sales tax claim against SLC of $980,000, although that claim was subsequently abated and no claims were ever asserted against Merkel and Hepburn personally. Prior to the expiration of the 400-day period and the abatement of the sales tax claim, the Merkels and Hepburns realized the COD Income described above. The Merkels and Hepburns claimed that the personal guarantee of SLC's debt and the secondary obligation on the state sales tax claim were contingent liabilities that rendered them insolvent on the date the COD Income was realized.

In Merkel, the IRS took the position that the term "liabilities" under Section 108 includes "only liabilities ripe and in existence on the measurement date" and therefore excludes all contingent liabilities.24 In support of its position, the IRS argued that under GAAP accounting standards "true contingent liabilities" are merely disclosed in the footnotes, rather than accrued in the financial statements as a liability.25 The taxpayers took the position that the plain meaning of the term "liabilities" includes "all liabilities, whether contingent or otherwise" and that contingent liabilities should be counted by multiplying the amount of the liability by the probability of payment, looking to the treatment of contingent liabilities for bankruptcy law purposes.26 After examining the language of the statute, the application of GAAP rules and the legislative history to the Insolvency Exception, the Tax Court rejected the IRS's position but also found that Congress had not "specified the minimum level of certainty" necessary to determine whether a contingent liability should be taken into account in the insolvency determination.27 The court then established a new rule that "any obligation claimed to be a liability" can be included in the insolvency determination only if "it is more probable than not that [the taxpayer] will be called upon to pay that obligation in the amount claimed. . . ."28

In creating this test, the Merkel court was guided by the rules for determining whether to accrue a contingent liability under GAAP. In its examination of these rules, the court acknowledged that GAAP only requires disclosure of contingent liabilities on the financial statements in certain circumstances, but also requires actual accrual of contingent liabilities on the balance sheet in other circumstances.29 Specifically, GAAP requires accrual only for contingent liabilities where information is available indicating that it is "probable" that the liability has been incurred.30 The court stated that the treatment of contingent liabilities under GAAP "is consistent" with the court's own "more probable than not" standard.31

The Merkel decision was upheld by the Ninth Circuit over the vigorous dissent of Judge O'Scannlain, who argued that the plain meaning of the word "liabilities" includes all types of liabilities and that all contingent liabilities, discounted by the probability of occurrence, should therefore be included in the insolvency determination.32 The dissent also argued that from a policy perspective the "all-or-nothing" approach adopted by the majority created an unjustifiably inequitable scenario where a taxpayer with a contingent liability that has a 49% probability of being paid may include none of the contingent liability while a taxpayer with a contingent liability that has a 51% probability of being paid may include the entire contingent liability.33

The actual result in Merkel, where the contingent liabilities were in fact both remote and never realized, may well have been correct based on the facts of that case. Far less clear, however, is whether the "more probable than not" test adopted by the court is the proper test for less remote contingent liabilities.

 

D. The Miller Case.

 

In Miller v. Commissioner,34 the taxpayer had taken out a loan that was guaranteed by a third party. When the taxpayer was unable to pay the loan, the guarantor paid the debt and waived any right of reimbursement from the taxpayer, triggering COD Income to the taxpayer. Relying on Merkel, the IRS argued that given the guarantee and the circumstances immediately before the discharge, it was more probable than not that the taxpayer would not be required to repay the loan, and the loan should therefore not be counted as a liability for purposes of determining whether the taxpayer was insolvent. The IRS appeared to have ignored the fact that the liability was fully recourse to the taxpayer and thus a fixed rather than contingent liability, pointing to the fact that the Millers listed the debt as a contingent liability on a financial statement prepared at the end of 1994 (which the taxpayers claimed was in error).35

The court ruled in favor of the Millers, reasoning that the language in Section 108(d) requiring the insolvency determination to be made "immediately before discharge" meant that Congress intended to count all liabilities for which discharge is imminent, without any discounting from their outstanding amount. The court went on to conclude that a liability simply could not be "contingent" if its discharge gave rise to COD Income:

 

If one argues, as [the IRS] does, that the discharge of the [loan] gives rise to discharge of indebtedness income for petitioner, because the discharge effects a freeing of assets previously offset by the liability arising from that loan, then it necessarily follows that petitioner's liability on the [loan] was not contingent and is to be treated as in existence immediately before the discharge.36

 

The court did apply the Merkel "more probable than not" standard when considering another portion of the taxpayer's liability that was not alleged to give rise to COD Income because it remained outstanding and was later repaid, and found that this portion of the liability satisfied the test.37

 

E. Summary of Recommendations.

 

This Report makes the following recommendations:

1. Contingent liabilities within a specified range of probability should be included in the insolvency determination at their fair market value. Contingent liabilities below this range (e.g., a less than 20% probability of payment) should be ignored in the insolvency determination and contingent liabilities above this range (e.g., a greater than 80% probability of payment) should be included in the insolvency determination without any probability discount. The fair market value of all contingent liabilities within the specified range should be determined on the same basis as the valuation of contingent liabilities under Treas. Reg. § 1.752-7(b)(3)(ii).

2. In the case of corporations that discharge debt by issuing stock to former creditors, Treasury and the IRS should consider allowing the following simplified formula as an optional safe harbor for establishing insolvency, which implicitly includes the value of any contingent liabilities that remain outstanding after the debt workout: Solvency = A + B - C, where A = fair market value of stock outstanding immediately after the debt discharge as determined for purposes of determining the corporation's COD Income, B = amount of cash or other property (other than stock) paid to creditors in partial discharge of the debt, and C = the outstanding amount of the corporation's discharged debt immediately before the discharge. We note that any guidance that incorporates this approach may not require any definition of "contingent liability".

3. Treasury and the IRS should consider adopting the definitions of "obligation," "liability" and "§ 1.752-7 liability" under the Section 752 Regulations as the operative definitions of liabilities, fixed liabilities, and contingent liabilities under Section 108(a), respectively.

4. In the case of contingent liabilities within the same class (e.g., product warranties) or portfolios of contingent assets and liabilities, contingent amounts should be valued in the aggregate on a class-by-class basis rather than on an individual basis for potential inclusion in the insolvency determination.

5. We do not recommend that potential tax liabilities of a debtor attributable to the debt restructuring that gives rise to COD Income or to other future events should be included in the insolvency determination, but recommend that this topic be reserved for further study and guidance.

 

II. Potential Methodologies for Taking Contingent Liabilities into Account.

 

There is a wide spectrum of potential methodologies for taking contingent liabilities into account in measuring insolvency under Section 108. We describe these methodologies and their potential policy implications below.

In assessing the merits of any of these methodologies, it is important not to overlook another statutory exception to the recognition of taxable COD Income: Section 108(a)(1)(A) (the "Bankruptcy Exception"), which excludes all COD Income triggered by a debt discharge in a title 11 case, regardless of whether the taxpayer is insolvent. Since most taxpayers may choose between effecting a debt discharge in an out-of-court restructuring or in a bankruptcy court proceeding, any guidance that adopts an unduly narrow or difficult to administer interpretation of the scope of the Insolvency Exception will force many taxpayers to file for bankruptcy protection merely to ensure that any COD Income does not give rise to an immediate tax liability. The tax law should not encourage taxpayers who will in fact have no ability to pay the tax after the discharge to file for bankruptcy merely because the proper treatment of their contingent liabilities in the insolvency determination is uncertain. We therefore believe that any future guidance in this area should adopt broad and administrable rules to ensure that out-of-court restructurings remain a viable alternative to affected taxpayers in appropriate cases.

We also note that even under the most inclusive of the methodologies described below, most taxpayers who return to profitability after the debt discharge will not be permanently relieved from the obligation to pay tax on the excluded COD Income. Under the attribute reduction rules of Section 108(b), the tax is generally deferred to future periods.

 

A. "All Events Test": Ignore Contingent Liabilities Entirely.

 

One possible approach is to ignore contingent liabilities entirely, based on the theory that only fixed liabilities should be taken into account for tax purposes. (We discuss the issues surrounding the lack of definition of "liability" for purposes of Section 108(d)(3) in Part III below.) This was essentially the IRS's litigating position in Merkel.38

Among the potential policy justifications for such an approach are (i) simplicity; (ii) consistency with the general tax principle that a deduction cannot be accrued for a liability until "all events have occurred which determine the fact of liability and the amount of such liability can be determined with reasonable accuracy;"39 and (iii) a "symmetry" argument that a liability should only be included in the insolvency calculation if the discharge of such liability would give rise to COD Income.40

We believe that the "all events" approach, which was rejected even in Merkel, would be unduly harsh in this context. It would ignore the policy considerations underlying the Insolvency Exception -- ensuring that a taxpayer will only recognize taxable COD Income when it has experienced a "freeing of assets" that inures to the benefit of the taxpayer (rather than its creditors), leaving it with sufficient assets to pay the tax on that income.41 Such an approach would also ignore the economic reality that contingent liabilities reduce the taxpayer's available resources even while they remain contingent. Finally, the symmetry argument (which was also considered, and explicitly rejected, by the Merkel court42) proves too much. Even putting aside contingent liabilities, there are many types of fixed liabilities that would not give rise to COD Income if the taxpayer fails to pay them,43 and nothing in Section 108 requires such fixed liabilities to be excluded from the insolvency calculation.44

Furthermore, the application of the "all-events" test to contingent liabilities will almost inevitably produce anomalies unless the same test also applies to contingent assets. Contingent assets are included at their full fair market value and therefore reduce insolvency. Consider a single lawsuit against a taxpayer who is fully indemnified by a credit-worthy third party. Although the taxpayer's net exposure to the litigation is zero, it will be treated as "less insolvent" than a taxpayer who is not a party to the litigation in any capacity unless both the contingent liability and the related indemnity right are either taken into account at their fair market value or excluded entirely.

For these reasons, we do not recommend an "all events" approach to the treatment of contingent liabilities. We believe that contingent liabilities should be taken into account in measuring a taxpayer's insolvency because it is consistent with the policies underlying the Insolvency Exception as well as other areas of the law in which Congress, Treasury and the IRS have become increasingly cognizant of the need to include contingent liabilities in the analysis of net value or assumed liabilities.45

We believe the more difficult question is exactly how such liabilities should be taken into account.

 

B. The Merkel Approach: All or Nothing.

 

The Merkel decision represents the current state of the law, at least in the Ninth Circuit. As discussed in greater detail above, it stands for the proposition that a contingent liability must be included in the insolvency calculation if it meets a "more probable than not" threshold. If the threshold is met, the entire liability is included; if not, it is completely excluded.

The Merkel approach offers the advantage of a bright-line "probability" test, and may be easier to administer at least for certain types of contingent liabilities. As illustrated in the examples below, however, the Merkel approach produces consistently incorrect results because it does not even attempt to capture the actual impact of the liability on the taxpayer's ability to pay. It instead applies the probability test as a burden of proof, allowing taxpayers who satisfy it to report the full amount of the liability in the insolvency determination without regard to its actual impact on the taxpayer's net worth. For the same reason, the Merkel approach also produces dramatically different results for taxpayers who find themselves on different sides of the bright line. For these reasons, Merkel has been criticized by several commentators.46

 

Example 1. Taxpayer A has a contingent liability with a maximum exposure of $100 that has a 51% likelihood of becoming fixed at $100. Taxpayer B also has a contingent liability with a maximum exposure of $100, but that has only a 49% likelihood of becoming fixed at $100.

 

Taxpayer A gets credit for the entire $100 of his potential liability under Merkel. This result is overly generous, compared to the actual economic burden that the liability is expected to impose on Taxpayer A. By contrast, Taxpayer B gets credit for none of her liability under Merkel. This result is overly harsh. A contingent liability with a 49% likelihood of becoming fixed imposes nearly the same economic burden on Taxpayer B as on Taxpayer A. In fact, any buyer of the business of Taxpayer B who is aware of the potential exposure would most likely reduce the purchase price to reflect the risk of assuming this liability.

In addition, although the court in Merkel stated that the treatment of contingent liabilities under GAAP "is consistent" with the court's own "more probable than not" standard,47 it was apparently unaware that its formulation of a "more probable than not" standard, which would allow all contingent liabilities with a likelihood of payment greater than 50% to be counted for measuring insolvency, differs from the GAAP determination as interpreted by many accountants. We understand that the prevailing view in the accounting profession is that accrual is required only for liabilities with a probability of payment in excess of 75% (or perhaps even 80%).48 The court's formulation, therefore, might allow a taxpayer to include a greater amount of contingent liabilities than the amount that would be accrued for GAAP purposes.49

Finally, as we discuss in greater detail in Part III.B below, Merkel is also an ill-suited approach to a large percentage of the universe of contingent liabilities. While some contingent contingencies, for example the guarantee at issue in Merkel, are basically "whether" contingencies with a binary "yes or no" range of potential outcomes, many others are "how much" contingencies. In a lawsuit involving an undisputed breach of contract, for example, the only issue in contention between the parties is the amount of damages. With respect to "how much" contingencies, therefore, attempting to apply Merkel is tantamount to asking the wrong question.

In summary, under Merkel some taxpayers benefit too much, while others are unduly penalized. And in either case, the true economic impact of a contingent liability on a taxpayer's net worth will rarely (if ever) be captured. Due to these inherent flaws, we do not recommend that future guidance adopt the Merkel approach.

 

C. Modified Merkel Approach: Retain Threshold, But Value the Liability ("Hybrid Methodology").

 

One partial solution to at least mitigate the more egregious flaws of the Merkel approach would be to retain its "more probable than not" standard, but limit the includible amount of any contingent liability that satisfies the standard to its fair market value. For example, in Example 1 above, Taxpayer B would still get no credit for any portion of the $100 potential liability because she failed to meet the probability threshold. Taxpayer A, however, while meeting the threshold, would only be permitted to include the $100 potential exposure in his insolvency calculation at its fair market value. Presumably, the value of the liability would be considerably less than $100, given that the liability has only a 51% likelihood of being realized and may not be immediately payable even if it is.

Although the Hybrid Methodology eliminates Taxpayer A's windfall in Example 1, it offers no help whatsoever to Taxpayer B. At best, therefore, this one-sided improvement to Merkel will properly reflect the insolvency of only some taxpayers while continuing to understate the insolvency of others. The propriety of this approach as a policy matter depends in part on whether one accepts the premise that some probability threshold (whether 50.1%, or some other, perhaps higher, number) is desirable in order to exclude speculative liabilities that are unlikely to materialize. As noted above, the accounting profession appears to have already answered this question affirmatively.50 However, we believe that there are other, more balanced solutions to the flaws of the Merkel approach. Once a decision is made to move away from the simplistic all-or nothing approach of Merkel towards a system that attempts to value the actual contingency, we believe it is fair to make that system apply to all contingent liabilities of a taxpayer other than those that are truly remote.

 

D. Include and Value All Non-Remote Contingent Liabilities ("Complete Valuation").

 

The most comprehensive solution to the flaws of Merkel would be to include all contingent liabilities (whether probable or not) in the insolvency calculation, but limit the amount included to their fair market values. This approach would convey the appropriate amount of credit for a liability to Taxpayer B in Example 1, while still eliminating any windfall to Taxpayer A. This is the methodology supported by the dissenting opinion in Merkel.51 It is also the methodology employed for purposes of measuring insolvency under the federal bankruptcy laws.52

We believe that the Complete Valuation approach would most accurately track the economic burden that contingent liabilities impose on the taxpayers' net worth, and therefore would be most faithful to the policies underlying the Insolvency Exception. It is also the only approach that seems well-suited for assessing both "whether" and "how much" contingencies alike. However, unless it excludes contingent liabilities that do not meet some minimum probability threshold, the Complete Valuation approach is likely to be far more difficult to administer than the Merkel approach because it would require taxpayers to value all contingent liabilities.

In at least one analogous area of the tax law, Treasury has already made this choice with respect to contingent liabilities. Treasury Regulations under Section 752 (the "Section 752 Regulations") take into account both fixed and contingent liabilities assumed by a partnership from a contributing partner in computing that partner's outside basis in connection with certain triggering events. These rules were designed to prevent trafficking in built-in losses, including those represented by contingent liabilities, by reducing the basis of the partnership interest in the hands of the partner who contributed such liabilities to the partnership.

The Section 752 Regulations adopt the Full Valuation approach for contingent liabilities, without any probability threshold. First, they define an "obligation" as:

 

. . . any fixed or contingent obligation to make payment without regard to whether the obligation is otherwise taken into account for purposes of the Internal Revenue Code. Obligations include, but are not limited to, debt obligations, environmental obligations, tort obligations, contract obligations, pension obligations, obligations under a short sale, and obligations under derivative financial instruments such as options, forward contracts, futures contracts, and swaps.53

 

Next, the regulations separate fixed obligations from contingent ones, defining only the former as "liabilities":

 

An obligation is a liability . . . only if, when, and to the extent that incurring the obligation (A) creates or increases the basis of any of the obligor's assets (including cash); (B) gives rise to an immediate deduction to the obligor; or (C) gives rise to an expense that is not deductible in computing the obligor's taxable income and is not properly chargeable to capital.54

 

Items classified as liabilities immediately reduce a partner's basis upon their assumption by the partnership from the partner (to the extent such liabilities are allocated to the other partners of the partnership). Any other obligation that does not meet the definition of liability is governed by Treas. Reg. § 1.752-7 as a "§ 1.752-7 liability"55 which, while not triggering an immediate basis reduction, could trigger such reduction upon certain specified events, such as a sale by the contributing partner of its partnership interest. A partner's share of a § 1.752-7 liability is:

 

the amount of deduction that would be allocated to the partner with respect to the § 1.752-7 liability if the partnership disposed of all of its assets, satisfied all of its liabilities (other than § 1.752-7 liabilities), and paid an unrelated person to assume all of its § 1.752-7 liabilities in a fully taxable arm's-length transaction (assuming such payment would give rise to an immediate deduction to the partnership).56

 

We believe this arm's length test is the appropriate measure of a contingent liability's fair market value, and could be adopted for purposes of valuing contingent liabilities under the Insolvency Exception.57 But we do not believe it should apply to all contingent liabilities of the taxpayer.

Because an out-of-court restructuring will not be a viable alternative to filing for bankruptcy for any taxpayer not clearly insolvent on the basis of its fixed liabilities alone unless the treatment of contingent liabilities under the Insolvency Exception is simple and administrable, we believe any guidance in this area should exclude contingent liabilities that do not satisfy a minimum probability threshold, for example those with a less than 20% likelihood of being realized. If such a threshold were adopted to weed out "remote" contingencies, however, we believe a similar threshold should apply on the opposite side of the spectrum for contingencies that are highly likely to occur (e.g., 80%). Under such a regime, only contingent liabilities in the band between 20% and 80% likelihood would need to be valued, and only to the extent necessary to establish sufficient additional insolvency to absorb any "excess" COD Income.

 

E. Wait and See: Open Transaction Treatment.

 

The final alternative is to keep the insolvency calculation open during the period that the liabilities remain contingent, and then revise the calculation retroactively when and if the contingent liability ever becomes fixed. This is the approach employed by the Treasury Regulations under Section 338 for purposes of determining the buyer's basis in the purchased assets when a Section 338 election is made. Under the Section 338 Regulations, the buyer is not entitled to increase its basis in the acquired assets by any assumed environmental or other liabilities that require "economic performance" under Section 461(h) until the liability is paid.58

In the context of the Insolvency Exception, we believe the application of an open transaction approach would be very difficult to administer. First, one must decide what baseline should be used to compute insolvency during the period that the contingent liabilities remain open. If the baseline is that taxpayers must initially ignore all contingent liabilities, report the maximum amount of COD Income up front, and then claim an ordinary deduction during future periods as and when the contingent liabilities become fixed, taxpayers with no ability to pay the tax at the time they realize the COD Income may be subject to immediate tax. The initial calculation of insolvency under this approach would be identical to the calculation under the "all events" test, discussed above.

On the other hand, allowing taxpayers to include all contingent liabilities in the calculation (subject to possibly reporting COD Income in the future if the liabilities taken into account are never paid) would seem unduly generous, and would still require an initial determination of the size of the contingent liabilities. Furthermore, making the income exclusion covered by the Insolvency Exception subject to potential recapture would frustrate taxpayers' desire for certainty in many out-of-court restructurings, in contrast with the blanket protection offered by the Bankruptcy Exception. Revisiting the value of taxpayer's contingent liabilities would also be inconsistent with the treatment of assets (contingent or otherwise), whose fair market value is not subject to subsequent adjustment under Section 108(a) based on hindsight. It would also run counter to the general notion of the insolvency analysis as a "snapshot" of the taxpayer's net worth on the date of the debt discharge, based on the facts known at that time, since that is the true measure of whether the taxpayer is likely to have the current resources to pay tax on COD Income.

Any redetermination of the excludible amount of COD Income would be further complicated by the corresponding adjustments to the amount of attribute reduction under Section 108(b) during future periods: if the contingent liabilities in question ultimately settle at less than their assumed value, the taxpayer would owe more tax in the year of discharge, offset by less tax in future years due to less attribute reduction. Conversely, if they ultimately settle at more than their assumed value, the taxpayer would owe less tax in the year of discharge, offset by more tax in future years due to more attribute reduction. Meanwhile, many of these attributes in question may have been used (or reported as eliminated) in the interim. In the case of longer term contingent liabilities (e.g., environmental liabilities), the complexity of reporting these redeterminations across multiple taxable years could be staggering.

More importantly, in most cases involving taxpayers who return to profitability, the attribute reduction rules under Section 108(b) already ensure that the tax avoided under the Insolvency Exception will become payable during future periods, as long as sufficient attributes were available for reduction. To superimpose an elaborate redetermination regime on attribute reduction that may at best merely alter what is already a mere timing difference seems very difficult to justify.

In summary, while the simplicity of an open transaction approach in the initial calculation of insolvency may have some superficial appeal, we believe that its initial financial impact (if an "all-events" baseline is used) will be unduly harsh to many taxpayers, and the subsequent adjustments during future periods necessary to account for the "actual" insolvency of the taxpayer would be very difficult to administer.

 

F. A Potential Alternative for Corporations.

 

As a practical matter, we believe that the dilemma of including, excluding, or valuing contingent liabilities is often avoided in determining the insolvency of a corporate debtor when its creditors receive stock of the corporation in a workout that provides no recovery to the historic shareholders. In our experience, for purposes of computing the COD Income under Section 108(e)(8), the debtor first estimates the value of stock issued to creditors, which will reflect the increase in net equity value resulting from the debt discharge. That value is then subtracted from the outstanding amount of the discharged debt to derive both the amount of COD Income under Section 108(e)(8) and the amount of insolvency immediately before the debt discharge under Section 108(a), permitting all of the COD Income to be sheltered under the Insolvency Exception. Under this approach, therefore, if the corporation issues 100% of its stock to its creditors in the workout, the amount of insolvency will match the amount of COD Income.

For example, assume that a corporate debtor issues 100% of its stock with a fair market value of $20 to a creditor in complete satisfaction of $100 of debt. The debtor would be treated as (i) having COD income of $80 and (ii) being insolvent by $80 immediately before the discharge. This analysis would be based on the view that, because the debt discharge freed up $100 of assets but only produced equity worth $20, the net enterprise value of the debtor must have been "under water" by $80 prior to the discharge. If the debtor had instead issued only 95% of its stock (worth $19) to the creditor, with the historic shareholders retaining 5% of the stock (worth $1), the debtor would still be treated as insolvent by $80 but would have COD Income of $81, because the creditor only received stock worth $19 in satisfaction of a $100 of debt.

In both cases, the formula for computing insolvency begins with the value of the stock after the discharge and then subtracts that value from the principal amount of the discharged debt. To the extent any cash or other property of the taxpayer is also used to partially pay down the debt in the restructuring, those assets would be added to the "asset" side of the calculation. Thus, the formula is: Solvency = A + B - C, where A is the fair market value of the stock outstanding immediately after the debt discharge for purposes of computing COD Income, B is the amount of cash or other property (other than stock) paid to creditors in partial discharge of the debt, and C is the outstanding amount of the taxpayer's discharged debt immediately before the discharge. If the resulting amount is negative, the corporation is insolvent to that extent.

Implicitly, the formula above reflects the value of any contingent liabilities that remain outstanding after the restructuring.59 Such liabilities appear in this calculation on the "asset" side, because they reduce the value of the stock after the discharge. By reflecting the fair market value of contingent liabilities, this methodology represents a form of the Complete Valuation approach.

We acknowledge, however, that this simple methodology is not without its shortcomings. First, it is perhaps overly simplistic to assume that equity value increases dollar for dollar with debt relief. Second, the formula may implicitly discount the fixed liabilities of the taxpayer based upon its ability to pay. As we discuss below in Example 2, we believe this type of discounting is inappropriate for fixed liabilities. For these reasons, we would not recommend this alternative as the primary methodology in lieu of the Complete Valuation approach. However, we believe it could serve as an optional safe harbor for corporate taxpayers to establish insolvency in an amount at least equal to the amount of COD Income.

Although this methodology is not available to individuals, it is a far simpler alternative for corporate debtors60 than the Complete Valuation approach and may be even less prone to abuse than any approach that requires taxpayers to separately value all non-remote contingent liabilities. We also believe it is widely used in practice. In some prior guidance, the IRS appeared to endorse this methodology, although some of these precedents involved the old "stock-for-debt" exception.61 We recommend that the IRS and Treasury reaffirm the acceptability of this methodology as an optional alternative for measuring contingent liabilities in appropriate cases.

 

III. Other Issues Relevant to Treatment of Contingent Liabilities.

 

If Treasury and the IRS issue guidance on the treatment of contingent liabilities for purposes of the Insolvency Exception, certain additional technical issues would need to be considered. We discuss them below.

 

A. Definition of "Liability" or "Contingent Liability."

 

As discussed above, the term "liability" is not defined for purposes of Section 108. Furthermore, there may be some confusion over what distinguishes a "contingent" from a "fixed" liability. This distinction was at issue in the Miller case, discussed in Part I.D above, where the IRS appeared to treat a fixed recourse liability of the taxpayer as contingent and speculative.

Although some clarification would be helpful, we do not believe it is absolutely necessary that Treasury and the IRS define these terms. For example, a revenue ruling could be issued that addresses the narrow question of whether a contingent liability (e.g., a guarantee) that is less than 50% likely (but more than 20% likely) to be triggered should be taken into account -- and, if so, to what extent -- without formulating a definition of "liability." Such a ruling would still constitute useful guidance for taxpayers. However, if Treasury and the IRS are inclined to address the issue in a more comprehensive manner, it may be more difficult to do so without defining certain terms.

There are a few definitions in analogous contexts that could be adopted for this purpose. For example, Section 358(h)(3) defines a liability as "any fixed or contingent obligation to make payment, without regard to whether the obligation is otherwise taken into account for purposes of this title" for purposes of Section 358(h). The Section 752 regulations, discussed above, generally define "obligation" in a similar manner, but reserve the term "liability" for fixed liabilities, while relegating all obligations that do not constitute "liabilities," such as contingent liabilities, to the status of "§ 1.752-7 liabilities."

If future guidance provides that either all or certain contingent liabilities are taken into account for purposes of the Insolvency Exception, the definition of "liability" would need to include such liabilities. We believe that the definition of "obligation" under the Section 752 Regulations, which expressly includes contingent obligations and enumerates certain common types of such obligations, is particularly instructive and could be used for this purpose. However, care must be taken to exclude fixed liabilities from any probability or valuation-based approach, which should be reserved for contingent liabilities only. Under current law, full recourse fixed liabilities are taken into account at their face value in computing insolvency,62 as are fixed non-recourse liabilities (except for Excess Nonrecourse Debt that is not the debt that is being discharged in the workout).63 We believe that, absent abusive situations in which the parties create an obligation that is not expected to be repaid,64 fixed debt should be respected as such and included in the calculation at its outstanding principal amount, even if it is trading at a discount.

 

Example 2. Corporation C has assets with a gross fair market value of $70 and publicly traded debt (issued at par) with a principal amount of $100. The debt is currently trading at $70.

 

There should be no question that Corporation C is insolvent by $30. It is liable for $100 and the debt is not contingent. If Corporation C negotiated a debt workout that reduced the principal amount of the debt from $100 to $70, it would not free any of its assets from the claims of creditors, resulting in no benefit to its shareholders. While it is true that Corporation C might be able to buy back the debt in the market for $70, the only reason it could do so is that the creditors know that their claims exceed the value of Corporation C's assets by $30, which is the very definition of insolvency. Taking a fixed recourse debt into account at its current value even though the creditor has a fixed claim on the assets of the debtor in excess of that amount would defeat the very purpose of the Insolvency Exception.65

In summary, we believe that any regulatory guidance defining "liability" for purposes of the Insolvency Exception would need to differentiate between fixed and contingent liabilities, subjecting only the latter to any kind of a valuation and/or probability-based discounting regime. As described above, the distinction could be based on the definitions of "liability" and "§ 1.752-7 liability" in the Section 752 regulations. Fixed liabilities would therefore retain their treatment under current law.

 

B. What Is the "Liability" Being Tested?: Binary vs. Non-Binary Outcomes.

 

Any regime for evaluating contingent liabilities, whether using a probability threshold or a valuation approach or both, would need to frame the universe of potential claims or outcomes that are being tested for probability or valuation. In some cases, where the contingent obligation relates to a single potential claim that is susceptible to a binary "win or lose" evaluation -- e.g., a guarantee of a debt instrument,66 which was at issue in Merkel, or a lawsuit involving a unique breach of contract claim -- it may be easy to define what the potential liability is, and it may be possible to estimate the probability that this particular liability would be triggered with reasonable accuracy. But in other cases, this inquiry may become more challenging. In such cases, or in the case of a "how much" contingency where the existence of a liability is undisputed, a valuation approach is more likely to yield the economically accurate answer than a methodology based purely on some probability threshold.

 

Example 3. D produces barbecue grills. Based on past experience, D knows that 2% of the grills are defective. Some of these defective grills are returned by purchasers during the warranty period, or malfunction, sometimes causing injuries and property damage. For every $10,000 worth of grills manufactured, D estimates a likely exposure for warranty or damage claims of $300, or 3%. Historically, this estimated exposure has proven to be accurate.

Example 4. E is a cigarette manufacturer. E typically litigates 100 new claims by cancer patients every year. Historically, E has lost 5% of such lawsuits, resulting in average damages of $50,000 per claim. Accordingly, for financial statement purposes, E records a reserve of $250,000 for potential cancer litigation losses every year, which is usually accurate.

 

The question in each of these examples would be: what is the "liability" being tested for probability? If the question is framed as the likelihood of any particular grill made by D being defective, or any individual lawsuit against E being successful, the answer would be that the odds are overwhelmingly in favor of D and E not having any liability. Each grill has only a 2% chance of being defective, and each cancer lawsuit has only a 5% chance of prevailing. Using a pure probability approach, but applied on a claim-by-claim basis, would result in none of their contingent liabilities being included in the insolvency calculation if the Merkel approach applies.

We believe the better approach here would be to frame the question as: what is the fair market value of the aggregate potential warranty or damage claims against D, or the aggregate potential cancer claims against E, in any given year? The answer is that D has a highly likely exposure of 3% of the value of its annual output, and E has a highly likely exposure of $250,000 annually.67 This is certainly how a potential buyer of D's or E's business would evaluate these contingent liabilities, or how an insurance company would estimate these exposures when pricing an insurance policy to cover them. Put another way, the proper inquiry is the economic value of a particular class of claims, rather than the mere likelihood of success of any single claim viewed on a stand-alone basis. A Merkel test, focused solely on a binary "win or lose" outcome of each claim, seems inappropriate in these cases.

In the accounting world, claims like this would be tested on a class basis. For example, under IFRS, when a number of similar liabilities exist the aggregate liability is valued by multiplying the possible outcomes by their probabilities and then adding them together.68 The following example, which appears in the IFRS rules, demonstrates this approach.

 

Example 5. An entity sells goods with a warranty under which customers are covered for the cost of repairs of any manufacturing defects that become apparent within the first six months after purchase. If minor defects were detected in all products sold, repair costs of $1 million would result. If major defects were detected in all products sold, repair costs of $4 million would result. The entity's past experience and future expectations indicate that, for the coming year, 75% of the goods sold will have no defects, 20% of the goods sold will have minor defects and 5% of the goods sold will have major defects. The expected value of the cost of repairs is:

(75% of 0) + (20% of $1 million) + (5% of $4 million) = $400,000.69

 

The same issue is raised by any contingent obligation that relates to a class of claims, which need to be evaluated in the aggregate. In many cases, viewed on a class basis, the obligation is not "contingent" at all -- in fact, it is a virtual certainty that some dollar amount will be spent to extinguish the entire class of potential claims, and the only issue is figuring out the correct dollar amount that should be reserved. Thus, on a class basis, it should be viewed as a "how much" contingency rather than a "whether" contingency. Examples include an employer's potential pension liability to a broad population of retired employees, or a financial institution that invests in a wide spectrum of derivatives (none of which individually may be likely to produce a loss, but taken as a whole are very likely to produce both gains and losses).

In the case of pension liabilities, even if they have not yet ripened into a tax deduction, it seems clear that the employer has a future obligation of some amount whose present value should be estimated and reflected in the insolvency calculation. In the case of a trader in derivatives, it would be inappropriate to reflect all of its "winner" positions at their positive fair market value on the asset side of the insolvency calculation, while refusing to reflect the negative fair market value of all of its "loser" positions (which in the aggregate are expected to require the trader to make payments to the counter-parties), by requiring that each loser position must meet some probability threshold.

The example of a portfolio of derivatives further demonstrates that similarly situated contingent assets and contingent liabilities should be treated similarly. We are not aware of any policy reason consistent with the purposes of the Insolvency Exception for taxpayers being required to include a portfolio trading position in the insolvency determination only when its current net value is positive. Such an approach would require a taxpayer to examine each individual position in the portfolio to determine whether it has a positive or negative value rather than value the portfolio as a whole, cherry-picking solely the positive items to reduce its deemed insolvency. Likewise, under a cherry-picking approach any executory contract could either be an includible asset (if it is expected to generate net income) or an excluded liability (e.g., an onerous lease) at any given time. By contrast, the Section 752 Regulations properly treat an onerous contract as a "§ 1.752-7 liability":

 

If the obligation arose under a contract in exchange for rights granted to the obligor under that contract, and those contractual rights are contributed to the partnership in connection with the partnership's assumption of the contractual obligation, then the amount of the § 1.752-7 liability or obligation is the amount of cash, if any, that a willing assignor would pay to a willing assignee to assume the entire contract.70

 

These examples serve as additional illustrations of why the Complete Valuation approach establishes a more accurate measure of a taxpayer's economic net worth than the Merkel all-or-nothing approach.

 

C. Treatment of Potential Tax Liabilities in the Insolvency Calculation.

 

There is virtually no guidance on the treatment of potential future tax liabilities associated with a debt restructuring for purposes of computing insolvency under Section 108(d)(3).71 However, future tax liabilities are well-recognized in the accounting field as a proper line item on the liability side of the ledger.72 They usually appear as deferred tax liabilities on the balance sheet, which are often quite speculative. For example, built-ingain in the assets of a corporate subsidiary that was acquired in a stock acquisition without a Section 338(h)(10) election creates an inherent tax liability that may be triggered in a future sale. Since it is typically not known whether or when this liability would be incurred (e.g., the taxpayer may simply sell the stock), this liability is likely to fail any test based upon a probability threshold. In a world governed by Merkel, for example, it would probably be ignored.

On the other hand, this same contingent tax liability could reduce the assets available to creditors in a restructuring if the taxpayer were to dispose of all of its assets for cash. Is it appropriate to consider these income tax liabilities -- either fixed or contingent -- in assessing insolvency?

 

Example 6. F owes $100 to Bank; F's obligation to Bank is secured by F's assets (which are worth $100 and have a tax basis of $0). Assume that in a bankruptcy, Bank's claim would be senior to the IRS's claim for income taxes. If F sells its assets for cash, he would be able to pay the entire $100 to Bank to extinguish the debt, leaving nothing for the IRS when the tax bill of $35 for the asset sale comes due. To avoid bankruptcy, however, F negotiates a debt reduction of $10 with Bank (reducing F's obligation to Bank to $90).

 

F does not appear to be insolvent. He had $100 of assets available to pay off Bank. He was also subject to a latent tax claim of $35, but that claim would not have been paid in a bankruptcy (due to the lack of assets to pay unsecured creditors) and thus could be viewed as being worth $0. The debt reduction has freed up $10 of F's assets, and F should be able to pay $3.50 in tax.73

 

Example 7. G owes $100 to unsecured Bondholders. Like F in Example 6, G has assets that are worth $100 that have a tax basis of $0. In a bankruptcy, the IRS's claim for taxes would have a higher priority than the Bondholders' claims. G sells the assets for $100, resulting in a tax liability of $35. Shortly thereafter, G pays $65 to the Bondholders and $35 to the IRS.

 

Here, the tax liability is not contingent -- it has in fact been triggered and should count in the insolvency calculation. Immediately prior to settling with the Bondholders, G had $135 of fixed liabilities and $100 of cash. Accordingly, G should be viewed as insolvent in an amount equal to the COD Income on the bonds -- $35 -- resulting in no taxable COD Income.

 

Example 8. Same facts as in Example 7, but now instead of selling assets G negotiates a debt-for-debt exchange with the Bondholders, issuing new bonds with a principal amount of $65 in exchange for the old bonds with $100 principal amount. G retains all of its assets.

 

At first blush, G appears to be solvent. G has retained $100 of assets against only $65 of fixed liabilities. However, G will owe an additional $35 of tax if it sells its assets for $100. It may seem odd to tax G in a debt discharge that provides lenders with economically similar value to the outcome in Example 7 where G was not taxed at all on its COD Income. Still, we believe the contingent income tax liability on a potential asset sale that may never happen, and whose occurrence may be within the control of the taxpayer, is a poor candidate for inclusion in the insolvency calculation here. But there may be other cases where some reserve for contingent future tax exposures would be appropriate.

Should the $12.25 tax on the COD Income itself (which is not a liability "immediately before" the debt discharge, but becomes one as a result of the debt discharge) be factored into the insolvency calculation?74 If so, then the analysis becomes circular, similar to what we saw in Example 2 above.75 G becomes insolvent by $12.25, meaning that only $22.75 of the COD Income is taxable, but this in turn reduces the tax bill and the amount of the insolvency.

Although the "right" answer to this question is beyond the scope of this Report, it could be addressed in separate guidance by Treasury and the IRS. Tax on COD Income does not seem like a contingent liability eligible for inclusion in the insolvency analysis at any point. Prior to the debt discharge, the exposure arguably does not exist -- or, if it exists, the contingent liability may have no value as demonstrated in Example 7 above: if G had effected a liquidating fire sale instead of a negotiated debt reduction, G would not have had any taxable COD Income because G would have qualified for the Insolvency Exception. (Alternatively, G may have been able to file bankruptcy, in which case its COD Income also would have been exempt from tax.) Following the debt discharge, the tax due on any COD Income is a fixed rather than contingent liability, and in any event the statutory language does not seem to permit taking into account a liability that did not exist immediately before the discharge.

We believe this issue deserves further study.

 

FOOTNOTES

 

 

1 The principal drafter of this Report was Vadim Mahmoudov. Substantial contributions were made by Jarrod Shobe. Helpful comments were received from Peter J. Connors, Steven Dean, Larry M. Garrett, Stuart J. Goldring, Stephen B. Land, Steven J. Lorch, William L. McRae, Andrew W. Needham, Michael L. Schler, David R. Sicular, Eric Sloan and Lee Zimet. This report reflects solely the views of the Tax Section of the NYSBA and not those of the NYSBA Executive Committee or the House of Delegates.

2 Unless otherwise indicated, all references in this Report to "Section" and "Sections" are to the Internal Revenue Code of 1986, as amended (the "Code"), and all references to "Treas. Reg. § " are to regulations issued thereunder (the "Treasury Regulations" or "Regulations"). References to the "IRS" are to the Internal Revenue Service, and references to "Treasury" are to the United States Department of the Treasury.

3 284 U.S. 1, 3 (1931).

4Id.

5 70 F.2d 95 (5th Cir. 1934).

6Id.at 96.

7Id. See also Quinn v. Commissioner, 31 B.T.A. 142 (1934) (debtor realized no COD Income if debt discharge did not make any of the debtor's assets available to the debtor).

8 36 B.T.A. 289 (1937).

9Id. at 292.

10Id.

11 The Bankruptcy Tax Act of 1980, Pub. L. 96-589.

12 To the extent COD Income is excluded from gross income, the taxpayer must reduce certain tax attributes, including net operating losses, tax basis in assets, certain tax credits and certain loss and credit carryovers. See Section 108(b). In many cases, therefore, the exclusion from income under the Insolvency Exception merely defers the tax otherwise payable on the COD Income until future periods.

13 Section 108(d)(3).

14 H.R. Rep. No. 833, 96th Cong., 2d Sess. 7, 9 (1980); S. Rep. No. 1035, 96th Cong., 2d Sess. 8, 10 (1980).

15 17 T.C. 506 (1951), acq., 1952-1 C.B. 2.

16Id. at 513.

17See also J.A. Maurer, Inc. v. Commissioner, 30 T.C. 1273 (1949), acq., 1959-2 C.B. 1 (same).

18 Section 108(e)(1).

19 PLR 8348001 (Aug. 18, 1983) (Issue #6).

20 366 U.S. 380 (1961).

21 Rev. Rul. 92-53, 1992-2 C.B. 48; see also Rev. Rul. 2012-14, 2012-24 IRB 1012 (applying Rev. Rul. 92-53 in partnership context for purposes of measuring each partner's insolvency under Section 108(d)(3)).

22 1997 FSA Lexis 144 (June 9, 1997).

23 109 T.C. 463 (1997), aff'd, 192 F.3d 844 (9th Cir. 1999).

24Id. at 467.

25Id. at 478.

26Id. at 467, 482. The taxpayers relied on Covey v. Commercial Natl. Bank, 960 F.2d 657 (7th Cir. 1992), a case decided under section 548 of the Bankruptcy Code (which permits recovery of preferential payments which occurred while the debtor was insolvent, or rendered the debtor insolvent). In Covey, the court required contingent liabilities to be included in the insolvency determination, but discounted by the probability of their occurrence.

27Id. at 474.

28Id. at 476, 484 (emphasis added). The 9th Circuit opinion formulated the test as follows: "taxpayer . . . must prove by a preponderance of the evidence that he or she will be called upon to pay an obligation claimed to be a liability and that the total amount of liabilities so proved exceed the fair market value." Merkel v. Commissioner, 192 F. 3d 844, 850 (9th Cir. 1999). The reasoning of both courts appeared to have been based in part on the taxpayer's burden of proof. Merkel, 109 T.C. at 484. We do not believe that the taxpayer's burden of proof on this issue compels a "more probable than not" test. Under the same burden of proof, the courts could just have easily required the taxpayer to establish a 49% likelihood of payment as a condition to including 49% of the liability in the insolvency determination.

29Id. at 478-79.

30Id. at 478 (citing to FASB Statement of Financial Accounting Standards No. 5, which is now codified in FASB Accounting Standards Codification 450).

31Id. at 479.

32 192 F.3d at 854.

33Id. at 853.

34 T.C. Memo. 2006-125.

35Id.

36Id.

37See id. at fn. 32. Although this bifurcation approach appears to be consistent with the treatment of Excess Nonrecourse Debt under Revenue Ruling 92-53, it is not clear why the court applied either this approach or Merkel to any portion of the taxpayer's fixed and fully recourse liabilities.

38 109 T.C. at 467. See also PLR 8348001 (Issue #6) (August 18, 1983), discussed in Part I.B supra.

39 Section 461(h)(4).

40See Merkel, 109 T.C. at 479-80. For a more detailed discussion of the symmetry argument, see Burgess Raby and William Raby, Do Contingent Liabilities Count for Section 108 Insolvency?, TAX NOTES, Jan. 12, 1998, at 205; Celia Clark, COD Income: New Opportunities for Insolvency Planning After Merkel, 89 J. TAX'N 29 (1998). Generally, a discharge of a contingent liability does not give rise to COD Income. See, e.g., Corporacion de Ventas de Salitre Y Yoda de Chile v. Commissioner, 130 F.2d 141 (2d Cir. 1942); PLR 201027035 (Mar. 31, 2010). Even if the discharge of a contingent liability somehow did give rise to COD Income, the contingent liability would have to be included in the insolvency analysis for the same reason as a fixed liability: to avoid triggering an immediate tax on taxpayers with no ability to pay. See our discussion of the Miller case in Part I.D supra.

41See Part I.A supra.

42See Merkel, 109 T.C. at 480-82.

43 Examples include unpaid interest or some other accrued expenses of a cash-basis taxpayer, whose cancellation would not give rise to COD Income under Section 108(e)(2) because its payment would have triggered a deduction, and debt owed by a subsidiary to its parent, which could be cancelled via contribution to capital without triggering COD Income under Section 108(e)(6).

44 If a contingent liability (or some other liability whose discharge would not give rise to COD Income, e.g., a liability that, if paid, would give rise to a deduction) is discharged as part of a larger restructuring that triggers COD Income on other debt, there is a technical question as to whether such liabilities were outstanding "immediately before the discharge" within the meaning of the statute. In the typical case, however, most contingent liabilities remain outstanding after an out-of-court restructuring of other debt. In such cases, these liabilities are clearly eligible for inclusion in the insolvency determination. Furthermore, to the extent any such liability is actually paid in the restructuring (e.g., a settlement of pension liabilities with the PBGC), the payment itself will constitute the best proof that the liability was in fact "real", in which case an amount no less than such payment should be included in the insolvency determination as a liability outstanding immediately before the discharge.

45See, e.g., Section 358(h) (reducing shareholder's basis in corporation's stock in certain circumstances when a transferee corporation assumes a contingent liability); Treas. Reg. § 1.752-7 (taking contingent liabilities into account in measuring a partner's outside basis in certain circumstances after the partnership has assumed such liabilities); Prop. Treas. Reg. § 1.368-1(f) (proposed regulations that deny tax free reorganization treatment to "no net value" transfers to a corporation). The preamble to the "no net value" proposed regulations states that "a liability should include any obligation of a taxpayer, whether the obligation is debt for federal income tax purposes or whether the obligation is taken into account for the purpose of any other Code provision" and that, generally, "an obligation is something that reduces the net worth of the obligor." See 70 Fed. Reg. 11,903, at 11,905 (Mar. 10, 2005).

46See, e.g., Gordon Henderson and Stuart Goldring, TAX PLANNING FOR TROUBLED CORPORATIONS § 404 (2012 ed.); Richard Lipton, The Tax Court's New Standard for Testing Contingent Liabilities -- Will it Work?, 88 J. TAX'N 150, 154 (1998).

47 109 T.C. at 479.

48See, e.g., IFRS and US GAAP: Similarities and Differences (Oct. 2012), at 120, available at http://www.pwc.com/en_US/us/issues/ifrs-reporting/publications/assets/ifrs-and-us-gaap-similarities-anddifferences-2012.pdf ("While a numeric standard for probably does not exist, practice generally considers an event that has a 75% or greater likelihood of occurrence to be probable."). We understand that various accounting firms have differed in their interpretations. Some firms view "probable" as an event that is at 80% or greater likelihood of occurrence while at least one firm views the threshold to be "more likely than not", i.e. 50.1%. In addition, even after the probability threshold is met, GAAP (unlike Merkel) does not necessarily require accrual of the entire liability. Rather, it requires a measurement of the liability based on a reasonable estimate of the amount payable, discounted as appropriate to reflect the expected date of future payment. Accounting Standards Codification Paragraphs 450-20-25-2 and 450-20-S30.

49 Notably, GAAP is not the only accounting standard applicable to U.S. taxpayers (other than individuals). An increasing number of taxpayers follow the IFRS accounting rules in lieu of (or in addition to) GAAP. Like GAAP, IFRS also employs a "probable" standard for determining when to accrue contingent liabilities; however, IFRS specifically defines "probable" as "more likely than not," which is virtually identical to the Merkel standard. See International Accounting Standard 37, Provisions, Contingent Liabilities and Contingent Assets.

50See Part I.C supra.

51Merkel, 192 F. 3d at 854.

52Covey, 960 F.2d at 659-60 (measuring insolvency for purposes of Section 548 of the Bankruptcy Code).

53 Treas. Reg. § 1.752-1(a)(4)(ii).

54 Treas. Reg. § 1.752-1(a)(4)(i).

55 A "§ 1.752-7 liability" is defined as an obligation described in Treas. Reg. § 1.752-1(a)(4)(ii) to the extent that either the obligation is not described in Treas. Reg. § 1.752-1(a)(4)(i) or the amount of the obligation exceeds the amount taken into account under Treas. Reg. § 1.752-1(a)(4)(ii). Treas. Reg. § 1.752-7(b)(3).

56 Treas. Reg. § 1.752-7(b)(3)(ii) (emphasis added). In addition to having a potential impact on the partner's outside basis, § 1.752-7 liabilities are also generally treated as built-in loss property with a negative value for Section 704(c) purposes. See Treas. Reg. § 1.704-3(a)(12).

57 The Section 752 Regulations are not the only area of tax law that requires a valuation of contingent liabilities. Similar valuations are required for purposes of measuring net unrealized built-in gain or loss ("NUBIG/NUBIL") under Section 382(h), under the "Section 338 approach." See Notice 2003-65, 2003-2 C.B. 747 at IV.A and Example 10. See also Treas. Reg. § 1.1001-1(g)(2)(ii) (requiring seller who receives a debt instrument that provides for contingent payments to value the future contingent payments unless their fair market value is "not reasonably ascertainable"); Treas. Reg. § 1.1275-4(b) ("noncontingent bond method" for certain contingent payment debt instruments, requiring taxpayer to estimate a projected payment schedule for contingent payments under the instrument).

58See Treas. Reg. § 1.338-5(b)(2)(iii), Example 2. Other areas of tax law where an open transaction approach is used or permitted in analogous circumstances include Treas. Reg. § 15A.453-1(d)(2)(iii) (permitting open transaction treatment for installment sales in "rare and extraordinary cases" where the fair market value of a contingent obligation cannot reasonably be ascertained) and Treas. Reg. § 1.1275-4(c)(4) ("wait and see" approach for contingent portions of a contingent payment debt instrument that is not subject to the noncontingent bond method). But see Notice 2003-65 at IV.A and Example 10 (contingent liability's initial estimated value included in NUBIG/NUBIL calculation, but such calculation is not readjusted to reflect a subsequent change in the amount of the liability).

59 To the extent any stock or other property is actually paid to extinguish contingent claims, such claims should be included in the calculation as additional liabilities in an amount at least equal to the fair market value of the property. See our discussion in note 44, supra. This should neutralize the impact under the formula of the use of the debtor's assets to pay contingent claims, which could otherwise result in COD Income in excess of the insolvency amount.

60 This methodology generally would not be applicable to partnerships, because insolvency is tested at the partner level. See Section 108(d)(6). However, it could be useful to a special purpose vehicle whose sole assets consist of a partnership interest in a partnership that realizes COD Income and whose sole liabilities consist of its share of the partnership's liabilities, which can be treated as the partner's liabilities for purposes of the insolvency determination. See Rev. Rul. 2012-14, 2012-24 IRB 1012.

61See, e.g., Rev. Rul. 92-52, 1992-2 C.B. 34; 1995 FSA Lexis 177 (Oct. 17, 1995); 1997 FSA Lexis 144 (June 9, 1997).

62See Merkel, 192 F.3d at 851 (justifying all-or-nothing approach to liabilities because the words "fair market value" in Section 108(d)(3) only modify "assets" but not "liabilities).

63 Rev. Rul. 92-53, 1992-2 C.B. 48.

64 It appears that this concern was the true motivation behind the IRS's litigating position in the Miller case. However, the IRS has other weapons at its disposal to combat such abuse. Arguably, a debt instrument created without any intention to repay or collect is simply not debt for tax purposes. See, e.g., Dixie Dairies Corp v. Commissioner, 74 T.C. 476 (1980). Treasury and the IRS could qualify any future guidance under the Insolvency Exception with an anti-abuse rule that disregards debt instruments or other arrangements created with a principal purpose of qualifying for the Insolvency Exception.

65 This example ignores the potential impact of income tax liabilities. To further complicate matters, suppose Corporation C has gross assets with a value of $70 and a basis of $0 and no net operating loss carryforwards. If the debt were unsecured and junior in bankruptcy to the IRS's claim for unpaid taxes, the most that the creditors may be able to recover would be $45.50 (because a taxable sale of assets would trigger a tax bill of $24.50 at a 35% tax rate). Thus, the debt should really trade at $45.50. But, if the inherent tax liability is not included in the insolvency calculation, and the debt were only included at its trading value, the taxpayer could be viewed as solvent ($70 assets, $45.50 value of debt). In that case, if Corporation C negotiates a debt reduction from $100 to $45.50, it would trigger COD Income of $55.50, producing another $19.43 of tax in addition to the potential future tax of $24.50 on the future asset sale. This potential future tax should reduce the trading value of the debt even further, making the analysis circular.

66 Because the actual liability of a guarantor of debt will often depend upon the extent to which the creditors recover a portion of their guaranteed claims from the assets of the original borrower, even the valuation of a guarantee may not be binary.

67 We are not proposing to include as contingent liabilities any liability that may arise from future events, even if those events are highly likely to occur. The only contingent liabilities that we believe should be eligible for inclusion in the insolvency analysis are those related to exposures arising from past events -- e.g., grills already manufactured and sold by D in Example 3. Contingent liabilities and other losses that may arise from future events should already be reflected on the asset side in the value of taxpayer's goodwill or overall enterprise value, which is normally based on expected future net profits.

68 International Accounting Standard 37, Provisions, Contingent Liabilities and Contingent Assets.

69Id.

70 Treas. Reg. § 1.752-7(b)(3)(ii).

71 With respect to historic tax exposures, e.g., uncertain tax positions taken on previously filed returns, or "nexus" exposures in various jurisdictions for prior years, we see no reason to treat such items differently from any other contingent liabilities.

72 Accounting Standards Codification Paragraph 740-10-25-2b. Deferred tax assets and liabilities are measured using the currently applicable law and tax rates. Deferred tax assets may be reduced if it is more likely than not that some or all of the deferred tax asset will not be realized. Accounting Standards Codification Paragraph 740-10-20. There are certain exceptions whereby a deferred tax liability is not recognized if it becomes apparent that the temporary differences will not reverse in the foreseeable future. Accounting Standards Codification Paragraph 740-10-25-3. Even though deferred tax assets and liabilities represent future income or expense, they are not discounted to present value when included in the current financial statements. Accounting Standards Codification Paragraph 740-10-30-8.

73 Note that, in order to pay the tax, F may need to sell assets worth $3.50 and trigger another $1.23 of tax, leaving F with only $5.27 after taxes, so arguably only $8.77 of assets were freed up by the debt discharge. However, F may be able to finance the payment without selling assets and we do not believe this potential tax on the asset gain should reduce the amount of F's solvency and taxable COD Income. F is clearly able to pay the tax on the whole $10 of debt reduction, so the policies of the Insolvency Exception do not make this a good case in favor of including latent tax liabilities in the analysis.

74 Some commentators believe this is the right answer. See, e.g., Robert Livsey, Determining if a Taxpayer Is Insolvent for Purposes of the COD Income Exclusion, 76 J. TAX'N 224, 227 (1992); Richard Lipton, Planning for Noncorporate Debt Workouts Outside of Bankruptcy, 70 TAXES 275, 290 (1992).

75 See our discussion in note 65, supra.

 

END OF FOOTNOTES
DOCUMENT ATTRIBUTES
  • Authors
    Needham, Andrew W.
  • Institutional Authors
    New York State Bar Association Tax Section
  • Cross-Reference
    For the Tax Court opinion in Merkel v. Commissioner, 109 T.C.

    463 (1997), aff'd, 192 F.3d 844 (9th Cir. 1999), see Doc

    1999-30468
    or 1999 TNT 182-7 1999 TNT 182-7: Court Opinions.
  • Code Sections
  • Subject Area/Tax Topics
  • Jurisdictions
  • Language
    English
  • Tax Analysts Document Number
    Doc 2012-23860
  • Tax Analysts Electronic Citation
    2012 TNT 225-17
Copy RID