CCA 200024058: Prior interest allocation method generally cannot be changed retroactively

Plain-English summary

Chief Counsel technical assistance addressed how a taxpayer values assets when apportioning interest expense for foreign tax credit purposes. A taxpayer that chose the tax book value method on its original return generally could not amend an earlier year's return to switch retroactively to the fair market value method. The advice applied the doctrine of election because both methods were available and the taxpayer had affirmatively chosen one. An amended return filed before the original return's due date could qualify for a timely filing exception, and the office said case specific facts still had to be examined. The regulation permitting a later change from tax book value without consent concerns future elections, not retroactive changes.

Ruling snapshot

• Question: May a taxpayer retroactively switch from tax book value to fair market value when apportioning interest expense?
• Outcome: Advice given: generally no, except for a timely amended return.
• Key authorities: IRC §§ 861, 864(e), and 904(a); Treas. Reg. §§ 1.861-8T(c)(2) and 1.861-9T(g).

Full text (IRS public release)

DEPARTMENT OF THE TREASURY
I N T E R N AL R E V E N U E S E R V I C E WASHINGTON, -1- D.C. 20224

Number: 200024058
Release Date: 6/16/2000

UILC: 861.09-07; 864.04-00

Date: May 31, 2000

CC:INTL:Br3 WTA-N-115390-99

INTERNAL REVENUE SERVICE NATIONAL OFFICE TECHNICAL ASSISTANCE

MEMORANDUM FOR GAYE SCHROEDER
IFASP–Foreign Tax Credits

FROM: Irwin Halpern
Senior Technical Reviewer CC:INTL:Br3

SUBJECT: Application of the Doctrine of Election to Retroactive
Elections to Apportion Interest Expense According to the Fair Market Value Method

This Technical Assistance responds to your request for advice received by our office on September 20, 1999. Technical Assistance does not relate to a specific case and is not binding on Examination or Appeals. This document is not to be used or cited as precedent.

ISSUE:

Whether the doctrine of election precludes a taxpayer from amending past years’ returns to retroactively elect to value its assets according to their fair market value for purposes of apportioning interest expense under Treasury Regulation section 1.861-9T(g), when the taxpayer originally elected to value the assets under the tax book value method for those years.
 -2-

DISCUSSION:

I. Background

Section 901 allows a credit for foreign income, war profits, and excess profits taxes paid or deemed paid by qualifying taxpayers that elect the foreign tax credit in lieu of a deduction under section 164(a)(3). Section 904(a) limits a taxpayer’s foreign tax credit to an amount equal to the precredit U.S. tax on the taxpayer’s foreign source taxable income.

Sections 861(b), 862(b), and 863(a) provide that taxable income attributable to gross income from domestic or foreign sources shall be determined by deducting the expenses, losses, and other deductions properly apportioned or allocated thereto, and a ratable part of any expenses, losses and other deductions that cannot be definitely allocated to some item or class of gross income. Treasury Regulation sections 1.861-8 through 1.861-17 provide specific guidance regarding the allocation and apportionment of deductions. Generally stated, deductions are allocated to classes of gross income and, as required by the operative section of the Code, apportioned between statutory and residual groupings of gross income.

The allocation and apportionment regulations emphasize the factual relationship between deductions and gross income. A deduction is considered to be definitely related to a class of gross income, and therefore allocable to such class, if the deduction is incurred as a result of, or incident to, an activity or in connection with property from which such class of gross income is derived. If a deduction is not definitely related to a class of gross income constituting less than all gross income, it is generally treated as allocable to all gross income.

In the case of interest expense, the regulations apply the factual relationship principle in a manner that emphasizes the fungibility of money. As Treasury Regulation section 1.861-9T(a) explains:

The method of allocation and apportionment for interest set forth in this section is based on the approach that, in general, money is fungible and that interest expense is attributable to all activities and property regardless of any specific purpose for incurring an obligation on which interest is paid. Exceptions to the fungibility rule are set forth in § 1.861-10T. The fungibility approach recognizes that all activities and property require funds and that management has a great deal of flexibility as to the source and use of funds. When money is borrowed for a specific purpose, such borrowing will generally free other funds for other purposes, and it is reasonable under this approach to attribute part of the cost of borrowing to such other purposes.
Consistent with the principles of fungibility, except as otherwise -3-
provided, the aggregate of deductions for interest in all cases shall be considered related to all income producing activities and assets of the taxpayer and, thus, allocable to all the gross income which the assets of the taxpayer generate, have generated, or could reasonably have been expected to generate.

Prior to the Tax Reform Act of 1986, taxpayers were generally allowed to apportion interest expense using either a “gross income” method or “asset” method. Treas. Reg. § 1.861-8(e)(2) (as in effect prior to September 9, 1988). However, use of the gross income method was restricted. As Treasury Regulations section 1.861- 8(e)(2)(v) explained: “[n]ormally, the deduction for interest expense relates more closely to the amount of capital utilized or invested in an activity or property than to the gross income generated therefrom, and therefore the deduction for interest should normally be apportioned on the basis of asset values.”

The Tax Reform Act of 1986 revised the treatment of interest expense in several respects. Under section 864(e)(2), all allocations and apportionments of interest expense must be made on the basis of assets rather than gross income. Under section 864(e)(1), the taxable income of each member of an affiliated group is determined by allocating and apportioning interest expense of each member as if all members of the group were a single corporation.

II. The Asset Method of Apportionment

Under the asset method, interest expense is apportioned between statutory and residual groupings of gross income (or among statutory groupings) in proportion to the average total values of the assets within each such grouping for the taxable year. Treas. Reg. § 1.861-9T(g)(1)(i). This average is generally computed on the basis of values at the beginning and end of the year; however, under certain circumstances, a different method must be used. Treas. Reg. § 1.861-9T(g)(2).

Taxpayers may elect to value their assets based on their tax book value or fair market value. Treas. Reg. §§ 1.861-8T(c)(2); 1.861-9T(g)(1)(ii). This election is generally indicated by marking a box on Schedule H of Form 1118, however, a taxpayer can also communicate the election by allocating its interest expense pursuant to a particular valuation method. Once a taxpayer elects to use the fair market value method, the taxpayer and all related persons must continue to use such method unless expressly authorized by the Commissioner to change methods. Treas. Reg. § 1.861-8T(c)(2). Treasury Regulation section 1.861-9T(g)(1)(ii) states a limited exception to this rule for certain taxpayers’ taxable years that commenced in 1987 and 1988.

A taxpayer that elects to apportion interest expense on the fair market value method must establish the fair market value of its assets to the satisfaction of the Commissioner. Treas. Reg. § 1.861-9T(g)(1)(iii). If a taxpayer fails to establish the
 -4-

fair market value of an asset to the satisfaction of the Commissioner, the Commissioner may determine the appropriate asset value. If a taxpayer fails to establish the value of a substantial portion of its assets to the satisfaction of the Commissioner, the Commissioner may require the taxpayer to use the tax book value method of apportionment. Id. See also Treas. Reg. § 1.861-8(f)(5) (stating taxpayers’ obligation to furnish information to verify their allocations and apportionments under the section 861 regulations) .

III. Application of the Asset Method

A taxpayer’s election to apportion interest expense under either the fair market value or tax book value method will directly affect several computations on its return. Most notably, the section 904 foreign tax credit limitation requires taxpayers to distinguish between their domestic and foreign source income, and to calculate separate foreign tax credit limitations for various categories of foreign source income. Because the section 904 limitation is based on foreign source taxable income, taxpayers must allocate and apportion their deductions, including interest expense, in order to compute their section 904 limitations. See also Treas. Reg. § 1.861-8(f)(1) (describing other computations affected by the allocation and apportionment of deductions).

A taxpayer’s choice of valuation methodology will also affect several computations related to its controlled foreign corporations (CFCs). For example, Treasury Regulation section 1.861-9T(f)(3)(i) provides that the interest expense of CFCs may be apportioned on either the asset method or a modified gross income method (described in Treasury Regulation section 1.861-9T(j)). This apportionment applies for purposes of computing subpart F income and computing earnings and profits for all other federal tax purposes. However, a CFC may not use the modified gross income method if the CFC is controlled by United States shareholders whose affiliated group elects the fair market value method of apportionment (rather than the tax book value method). Treas. Reg. § 1.861-9T(f)(3)(i). Thus, any election by an affiliated group to change its valuation method to the fair market value method effectively disqualifies its controlled CFCs from utilizing the modified gross income method of apportionment.

A CFC’s method of apportioning interest expense can directly affect its U.S. shareholders’ returns. For example, a shift in valuation methods at the CFC level could alter the amount of interest expense allocated to subpart F income and non- subpart F income, thereby affecting a U.S. shareholder’s inclusion under the subpart F regime. Similarly, for purposes of applying the section 904(d)(3) “look- through” rules, a shift in valuation methods at the CFC level will likely alter the distribution of the CFC’s interest expense between its various categories of income. See Treas. Reg. § 1.904-5(c)(2) (stating specific rules for the allocation and apportionment of interest expense at the CFC level). In short, any time a U.S. parent uses a net figure provided by a CFC to calculate its tax liability, as long as
 -5-

that net figure is derived by subtracting interest expense, a change of asset valuation methods at the CFC level will likely impact the U.S. parent’s return.

IV. The Doctrine of Election

The doctrine of election generally binds a person to their initial choice, where the person had an equal right to choose one or more alternatives or inconsistent rights. J. Mertens, Law of Federal Income Taxation § 60.27 (1989). “[A] viable, healthy doctrine applicable to elections made under the tax laws,” the doctrine of election has enjoyed widespread application. Grynberg v. Commissioner, 83 T.C. 255, 261 (1984).

The doctrine of election as it applies to federal tax law consists of the following two elements: (1) There must be a free choice between two or more alternatives; and (2) there must be an overt act by the taxpayer communicating the choice to the Commissioner, i.e., a manifestation of the choice. See Grynberg, 83 T.C. at 261. See also Bayley v. Commissioner, 35 T.C. 288, 298 (1960), acq., 1961-2 C.B. 4; Burke & Herbert Bank & Trust Co. v. Commissioner, 10 T.C. 1007, 1009 (1948). Pursuant to the doctrine of election, a taxpayer that makes a conscious election under the tax laws may not, without the consent of the Commissioner, revoke or amend its election merely because events do not unfold as planned. See, e.g., J.E. Riley Inv. Co. v. Commissioner, 311 U.S. 55 (1940); Pacific Nat’l Co. v. Welch, 304 U.S. 191 (1938).

Pacific Nat’l Co. v. Welch, 304 U.S. 191 (1938), is “often regarded as the fundamental authority for the development” of the doctrine of election. Estate of Stamos v. Commissioner, 55 T.C. 468, 473 (1970). In Pacific National, the taxpayer had the option to treat certain income under the deferred payment or installment method. The taxpayer reported the income using one method and later sought a refund based on a computation under the other method. Among the reasons articulated by the Supreme Court for its refusal to allow the taxpayer's change from one method to the other is that such changes would impose burdensome uncertainties upon the administration of the revenue laws and would require recomputation and readjustment of tax liability for subsequent years. Id. at

194. Pacific National established the general rule that a taxpayer that elects a
     proper method on a return may not later revoke or change that election and substitute another method.

The doctrine of election has been applied in a widespread manner under a variety of Code provisions. In Grynberg, the Tax Court applied the doctrine of election to prohibit taxpayer from changing its method of calculating its charitable contribution deduction to another permissible method. Id. See also, e.g., Rose v. Grant, 39 F.2d 340 (5th Cir. 1930) (husband and wife prohibited from filing separate returns after making valid election to file joint returns); Burke & Herbert Bank & Trust Co., 10 T.C. 1007 (1948) (taxpayer prohibited from changing election to include tax-free
 -6-

interest in excess profits tax net income upon realization that election was disadvantageous once Commissioner determined that taxpayer erroneously omitted certain tax-free interest.)

Courts have articulated several rationales supporting the general principle that elections are considered binding. These rationales include: (1) preventing administrative burdens and inconvenience in administering the tax laws, particularly when the new method chosen requires a recalculation of tax liability for several taxable years or for other taxpayers; (2) protecting against loss of revenues by preventing taxpayers from using the benefit of hindsight to choose the most advantageous method of reporting; (3) promoting consistent accounting practice (i.e., foreclosing adjustments based on hindsight), thereby securing uniformity in the collection of revenue; and (4) providing an equitable and fair tax system by treating similarly situated taxpayers consistently. J.E. Riley Inv. Co. v. Commissioner, 311 U.S. at 59; Pacific Nat’l Co. v. Welch, 304 U.S. at 194; Mamula v. Commissioner, 346 F.2d 1016, 1018-19 (9th Cir. 1965); Barber v. Commissioner, 64 T.C. 314 (1975); Thorrez v. Commissioner, 31 T.C. 655, 668 (1958), affd. per curiam 272 F.2d 945 (6th Cir. 1959); Pacific Vegetable Oil Corp. v. Commissioner, 26 T.C. 1, 16 (1956), revd. on another issue 251 F.2d 682 (9 th Cir. 1957); Estate of Curtis v. Commissioner, 36 B.T.A. 899, 906-07 (1937). While these points support application of the doctrine of election, they are properly construed as underlying explanations of the doctrine, and not required elements for its application. Accordingly, the absence of one or more of these considerations in a particular case does not render the doctrine of election inapplicable.

A limited number of exceptions to the doctrine of election have been recognized and may permit taxpayers to change affirmative elections made on their federal tax returns: (1) The amended return was filed prior to the date prescribed for filing the original return; (2) the taxpayer's treatment of the contested item in the amended return was not inconsistent with his treatment of that item in his original return; or (3) the taxpayer's treatment of the item in the original return was improper and the taxpayer elected one of several allowable alternatives in the amended return. See Grynberg, 83 T.C. at 262; Goldstone v. Commissioner, 65 T.C. 113, 116 (1975). It should be noted that, outside of these limited circumstances (as well as analogous circumstances not involving elections), amended returns are regarded as “creature[s] of administrative origin and grace,” and acceptance of amended returns is solely within the discretion of the Commissioner. Badaracco v. Commissioner, 464 U.S. 386, 393 (1984). See also Cloutier v. United States, 709 F. 2d 480, 484 (7th Cir. 1983); Koch v. Alexander, 561 F.2d 1115, 1117 (4th Cir. 1977); Miskovsky v. United States, 414 F.2d 954, 955 (3d Cir. 1969). Cf. Woodbury v. Commissioner, 900 F.2d 1457, 1461 (10th Cir. 1990); Keeler v. Commissioner, 180 F.2d 707, 710 (10th Cir. 1950).

Some courts have discussed further exceptions to the doctrine of election for certain taxpayer “mistakes.” See, e.g., Grynberg, 83 T.C. at 261; Shull v.
 -7-

Commissioner, 30 T.C. 821, 828 (1958). While the scope and application of these exceptions in the context of the doctrine of election is far from clear, it is apparent that taxpayers cannot retroactively revoke an election on the sole basis that they later realize, through hindsight, that an elected choice failed to maximize their tax benefits. Such a “mistake” exception would virtually swallow the doctrine of election and thereby contradict the well-established principle that an exception to a rule cannot be construed so broadly as to render the rule meaningless. See, e.g., Camps Newfound/Owatonna, Inc. v. Town of Harrison, 520 U.S. 564, 594 (1997); Digital Equipment Corp. v. Desktop Direct, 511 U.S. 863, 868 (1994); Guidry v. Sheet Metal Workers Nat’l Pension Fund, 493 U.S. 365, 377 (1990). Thus, a taxpayer’s failure to ascertain the fair market value of its assets for purposes of electing the fair market value method on its original return cannot be considered a “mistake” that entitles it to revoke its original tax book value method election. As the Tax Court observed, “Oversight, poor judgment, ignorance of the law, misunderstanding of the law, unawareness of the tax consequences for making the election, miscalculation, and unexpected subsequent events have all been held insufficient to mitigate the binding effect of elections made under a variety of provisions of the Code.” Grynberg, 83 T.C. at 262 (quoting Estate of Stamos, 55 T.C. 468, 474 (1970)).

V. Analysis and Conclusion

While it is imperative that you examine the specific facts present in each case, the following general statements can be made. Taxpayers have a free choice between utilizing the tax book value and fair market value methods. Taxpayers also affirmatively manifest this choice on their returns. Thus, the two elements required for application of the doctrine of election would be present in a case where a taxpayer attempts to amend past years’ returns to retroactively elect the fair market value method of apportionment.

The doctrine of election will not apply if taxpayers amend their returns prior to the due date of such original return. However, a request to utilize the fair market value method is inconsistent with the prior use of the tax book value method, which is a permissible method for apportioning interest expense. Thus, the “timely filing” exception to the doctrine of election could apply; the latter two exceptions should not.

In conclusion, unless a taxpayer falls within the timely filing exception, the doctrine of election applies to prevent a taxpayer from amending past years’ returns to retroactively elect the fair market value method of apportionment.

It should be noted that, in reaching our conclusion, we have examined the requirement in Treasury Regulation section 1.861-8T(c)(2) that a taxpayer that uses the fair market value method must continue to use that method unless expressly authorized to change methods. While this rule can reasonably be construed to
 -8-

imply that a taxpayer does not have to secure the Commissioner’s consent to discontinue the tax book method, both of these aspects of Treasury Regulation section 1.861-8T(c)(2) apply solely with regard to prospective elections; that is, these rules describe the extent to which a past year’s election affects a subsequent election made on a timely filed, current year return. It is for this reason that Treasury Regulation section 1.861-8T(c)(2) requires that a taxpayer “continue to use [the fair market value] method” (emphasis added), clearly indicating that the rule applies on a going forward basis. The section 861 regulations do not address the permissibility of retroactive elections, which are properly considered under the doctrine of election.

If you have any questions, please call Melissa D. Arndt at (202) 622-3850.

______
IRWIN HALPERN
Senior Technical Reviewer
Branch 3
Office of Associate Chief Counsel
(International)