Reset A A Font size: Print

United States Court of Appeals, Federal Circuit.

RUSSIAN RECOVERY FUND LIMITED, Plaintiff-Appellant v. UNITED STATES, Defendant-Appellee

2016-1718, 2016-1719

Decided: March 14, 2017

Before O'MALLEY, BRYSON, and WALLACH, Circuit Judges. DOUGLAS HALLWARD-DRIEMEIER, Ropes & Gray LLP, Washington, DC, argued for plaintiff-appellant. Also represented by COURTNEY M. COX, JUSTIN FLORENCE, KATHLEEN SAUNDERS GREGOR, LORETTA R. RICHARD, Boston, MA; BRITTANY CVETANOVICH, Chicago, IL. ANDREW M. WEINER, Tax Division, United States Department of Justice, Washington, DC, argued for defendant-appellee. Also represented by RICHARD FARBER, GILBERT STEVEN ROTHENBERG, CAROLINE D. CIRAOLO, DIANA L. ERBSEN.

Appellant Russian Recovery Fund Limited (“RRF”), acting through its tax matters partners Russian Recovery Advisers, L.L.C. (“RRA”) and Bracebridge Capital, L.L.C. (“Bracebridge”), sued the United States (“the Government”) in the U.S. Court of Federal Claims, seeking readjustment of partnership items pursuant to the Tax Equity and Fiscal Responsibility Act (“TEFRA”), I.R.C. §§ 6221–6233 (2000). RRF alleges that the Internal Revenue Service's (“the IRS”) October 14, 2005 Notice of Final Partnership Administrative Adjustment (“2005 FPAA”) improperly disallowed approximately $50 million of losses that RRF had claimed for fiscal year 2000 and imposed a 40% penalty on any underpayment. The parties filed cross-motions for summary judgment on timeliness grounds, and the Court of Federal Claims held that the limitations period for assessing taxes against RRF's indirect partners had expired as to some, but not all, indirect partners. See Russian Recovery Fund Ltd. v. United States (RRF I), 101 Fed. Cl. 498, 510–11 (2011) (granting-in-part and denying-in-part the parties' motions for summary judgment). Following trial on the claims not resolved at summary judgment, the Court of Federal Claims entered judgment for the Government, sustaining the IRS's disallowance of the losses and imposition of penalties. See Russian Recovery Fund Ltd. v. United States (RRF II), 122 Fed. Cl. 600, 601–02 (2015).

RRF appeals. We have jurisdiction pursuant to 28 U.S.C. § 1295(a)(3) (2012). We affirm.


The Court of Federal Claims's factual findings are extensive and clearly presented. See RRF II, 122 Fed. Cl. at 602–14; RRF I, 101 Fed. Cl. at 500–01. Because these factual findings are largely undisputed, we recite only those facts necessary to resolve this appeal.

There are several players of interest. Nancy Zimmerman co-founded Bracebridge, a management company. RRF II, 122 Fed. Cl. at 602. Bracebridge created RRF, a hedge fund. Id. Bracebridge also manages FFIP, L.P. (“FFIP”), another fund. Id. All three—Bracebridge, RRF, and FFIP—are partnerships. RRF I, 101 Fed. Cl. at 500. Relevant to this appeal, Ms. Zimmerman is a direct partner of FFIP, and FFIP is a direct partner of RRF. Id. In this context, Ms. Zimmerman is an indirect partner of RRF and represents similarly situated indirect partners of RRF (direct partners of FFIP). Id.

In 1998, Russian sovereign debt was traded exclusively on the Moscow Interbank Currency Exchange (“MICEX”). RRF II, 122 Fed. Cl. at 603–04. Non-Russian investors could not invest directly in Russian sovereign debt on the MICEX; however, they could invest in derivative instruments known as credit-linked notes (“CLNs”) sold by certain authorized banks. Id. at 603. When Russia defaulted on its sovereign debt in August 1998, the Russian ruble collapsed, and CLNs lost nearly all of their value. Id. These assets also became extremely illiquid: the Russian Central Bank imposed currency exchange limitations that prevented the ruble from being freely traded, and the Russian government only allowed the authorized banks to access the debt and trade in rubles. Id.

These events had serious consequences for Tiger Management, LLC (“Tiger”), one of the world's largest managers of hedge funds. Id. at 604. Two of Tiger's funds, foreign partnerships that do not pay U.S. taxes, had purchased CLNs through Deutsche Bank for more than $230 million. Id. After the collapse, those CLNs were worth less than 10% of their original value. Id. And Tiger overall was in bad straits: in 1998, Tiger managed $22 billion; but by 2000, that amount had dropped to $6 billion as a result of heavy losses in Russian debt, Asian debt, and an investment in US Airways. Id. at 613. During that period, Tiger needed cash to redeem the shares of investors who wanted out, but the capital controls on Russian debt hampered Tiger's ability to sell its devalued CLNs. Id.; see id. at 604 & n.9.

Ms. Zimmerman “believed that she could make money for herself and investors by obtaining devalued Russian debt at pennies on the dollar in anticipation of a recovery of the ruble and hence something approaching face value of debt instruments.” Id. at 603. As a result, Bracebridge established RRF and sought holders of Russian securities to contribute CLNs or cash in exchange for shares of RRF. Id. at 603−04; see J.A. 1758. Bracebridge also established RRA, a separate management company to advise RRF and collect management fees. RRF II, 122 Fed. Cl. at 602.

Despite earnest marketing efforts by Bracebridge during the first several months, RRF largely failed to obtain investors and still had no assets in March 1999. Id. at 605. An internal Bracebridge email on March 9, 1999 discussed a potential contribution of CLNs from an entity through Deutsche Bank. Id. Given concern that RRF needed partners to attract the potential investor, the email proposed having Bracebridge-controlled entities become RRF partners. Id. A telephone list circulated the next day contained the contact information of players from Bracebridge, Deutsche Bank, and Tiger. Id.

In April 1999, FFIP “contribute[d] the first assets to RRF.” Id.; see id. at 602. Then, on April 30, 1999, Bracebridge's James DiBiase emailed Ms. Zimmerman about the “need[ ]” to represent that a “high” percentage “of RRF (i.e., FFIP) is owned by individuals” to attract Deutsche Bank's investors. Id. at 606 (internal quotation marks and citation omitted). In a second email on May 14, 1999, he advised Ms. Zimmerman that RRF should not allow corporations to join because “it could possibly impair one of our most valuable assets,” i.e., “the built-in losses in Russian depreciated assets that might end up in RRF.” Id. (internal quotation marks and citation omitted). As explained in a later email by Mr. DiBiase, the presence of corporations could preclude later resale “since people interested in buying tax losses don't want to transact with corporations.” Id. (internal quotation marks and citation omitted).

A series of transactions followed, each of which was orchestrated by Deutsche Bank. Id. at 604, 620. First, in late May 1999, RRF's first two substantial outside investors—both funds operated by Tiger—transferred CLNs to RRF in exchange for an ownership interest in RRF. Id. at 607. Prior to investing, however, Tiger requested certain changes to the “standard RRF offering memorandum” and refused to execute the standard subscription agreement representing that “the Shares subscribed for hereby are being acquired by the undersigned for investment purposes only, for the account of the undersigned [,] and not with a view to any sale or distribution thereof.” Id. (paraphrasing J.A. 8178); see J.A. 5898–99. In response, RRF reduced the three-year lock-up period to “allow[ ] Tiger to redeem its shares on or after July 1, 1999, in exchange for cash or assets ‘in kind,’ ” and excluded the representation that Tiger was purchasing the shares “for investment purposes only” from the subscription agreement. RRF II, 122 Fed. Cl. at 607; see J.A. 8853, 8900, 8945–48. Second, on June 3, 1999 (i.e., approximately two weeks after the first transaction between RRF and Tiger), “Tiger sold all of its RRF partnership shares to FFIP” for approximately $14.1 million, a discount of $800,000. RRF II, 122 Fed. Cl. at 609; see J.A. 9069. Notably, during the two weeks between Tiger's acquisition of its ownership interest in RRF and its sale of that interest to FFIP, the value of the shares had in fact increased. RRF II, 122 Fed. Cl. at 609. And a fax from Deutsche Bank to Mr. DiBiase during this period makes clear that “it was RRF, not Tiger, that would have had an interest in an entity like FFIP purchasing [Tiger's] shares” and acquiring the built-in losses. Id. at 608. Third, on June 22, 1999, RRF sold 77.18% of the Tiger CLNs to General Cigar Corporation (“General Cigar”) for cash and shares. Id. at 609; see J.A. 4992–95. Finally, in 2000, RRF sold the remaining 22.82% of the Tiger CLNs on the open market. RRF II, 122 Fed. Cl. at 609–10.

Following these transactions, Mr. DiBiase began working with Ernst & Young to “provide[ ] the documents and facts that would collectively lay the foundation upon which the accountants would prepare RRF's [tax] returns.” Id. at 622. On August 14, 2001, RRF filed its 2000 tax return, allocating a loss to FFIP, which included a loss of $49,786,826 from the sale of the 22.82% of the Tiger CLNs. Id. at 609−10; RRF I, 101 Fed. Cl. at 500; see J.A. 1621–23, 9496.1 FFIP then reported losses for the 2000 and 2001 tax years, much of which were attributable to the loss claimed by RRF in 2000. RRF I, 101 Fed. Cl. at 500. FFIP's 2001 losses flowed through FFIP to Ms. Zimmerman, who filed her 2001 individual tax return on October 15, 2002. Id. On her 2001 individual tax return, Ms. Zimmerman reported a “substantial amount” of RRF's loss. Id. “In other words, the bulk of the losses RRF allocated to FFIP in 2000 were not passed through in 2000, but were retained by FFIP until 2001, at which point the losses impacted Ms. Zimmerman's 2001 return.” Id.

In 2005, the IRS performed an audit of FFIP's 2001 partnership return, which ultimately resulted in the issuance of a “no adjustments letter” to FFIP. Id. at 501; see J.A. 201. However, in October 2005, the IRS issued the 2005 FPAA to RRF for its 2000 tax year, which disallowed the loss RRF claimed for the sale of the Tiger CLNs and imposed a 40% penalty. RRF I, 101 Fed. Cl. at 501; RRF II, 122 Fed. Cl. at 621.


RRF argues that the Court of Federal Claims erred by (1) denying its cross-motion for summary judgment in RRF I because “the proposed assessments were time-barred,” Appellant's Br. 22 (capitalization omitted); (2) “holding that Tiger's contributions to RRF were not valid partnership contributions,” id. at 33 (capitalization modified); and (3) “upholding a massive penalty based on its new partnership requirements,” id. at 55 (capitalization omitted). After articulating the relevant standard of review, we address these arguments in turn.

I. Standard of Review

The present appeal involves factual findings and legal conclusions reached on summary judgment and following trial. “We review the Court of Federal Claims' [s] grant of summary judgment under a de novo standard of review, with justifiable factual inferences being drawn in favor of the party opposing summary judgment.” Winstar Corp. v. United States, 64 F.3d 1531, 1539 (Fed. Cir. 1995) (en banc) (citation omitted), aff'd, 518 U.S. 839 (1996). In appeals following a trial, we review the Court of Federal Claims's legal conclusions de novo and its factual findings for clear error. See John R. Sand & Gravel Co. v. United States, 457 F.3d 1345, 1353 (Fed. Cir. 2006).

The present appeal also raises issues of statutory and regulatory construction, the characterization of transactions for tax purposes, and the reasonable cause exception to tax penalties. “We ․ review questions of statutory and regulatory construction without deference.” SRA Int'l, Inc. v. United States, 766 F.3d 1409, 1412 (Fed. Cir. 2014). “We review the characterization of transactions for tax purposes de novo, based on underlying findings of fact, which we review for clear error.” Wells Fargo & Co. v. United States, 641 F.3d 1319, 1325 (Fed. Cir. 2011) (citation omitted). Finally, as to the reasonable cause exception to tax penalties, “[w]hether the elements that constitute reasonable cause are present in a given situation is a question of fact, but what elements must be present to constitute reasonable cause is a question of law.” United States v. Boyle, 469 U.S. 241, 249 n.8 (1985) (internal quotation marks and citations omitted).

II. The Court of Federal Claims Did Not Err by Denying RRF's Cross-Motion for Summary Judgment in RRF I

In RRF I, the parties filed cross-motions for summary judgment disputing whether the IRS timely issued the 2005 FPAA and whether it suspended the limitations period for adjustment and assessment of RRF's indirect partners' (FFIP's direct partners') individual tax returns. 101 Fed. Cl. at 499. With the agreement of the parties, the Court of Federal Claims selected Ms. Zimmerman as representative of the RRF indirect partners (who also are FFIP direct partners) whose tax returns were filed less than three years prior to the issuance of the 2005 FPAA. Id. at 499, 504. The Court of Federal Claims determined that “if it is demonstrated that the loss[ ] from RRF's 2000 tax return can be traced to Ms. Zimmerman's 2001 tax return then [the IRS] may assess additional taxes.” Id. at 509. It then “h[e]ld that the [2005] FPAA ․ validly suspended the limitation[s] period for assessing Ms. Zimmerman's 2001 individual tax return.” Id.

On appeal, RRF argues that “[a]ny attempt by the IRS to collect tax from FFIP partners in 2001 and later years based on FFIP partnership items is time-barred because the IRS failed to issue an FPAA to FFIP for those years.” Appellant's Br. 22. According to RRF, the 2005 FPAA “toll[ed] the period for assessing tax ‘attributable to’ RRF's 2000 partnership items, not FFIP's 2001 partnership items,” id. at 23 (capitalization modified), because the 2005 FPAA cannot apply to either two partnerships (i.e., RRF and FFIP) or two tax years (i.e., 2000 and 2001), see id. at 24–32. We hold the Court of Federal Claims correctly determined that the losses claimed on Ms. Zimmerman's 2001 tax return are “attributable to” the loss claimed in RRF's 2000 tax return, the limitations period for which was suspended by the 2005 FPAA.

A. Legal Framework

Pursuant to the TEFRA, the “[g]eneral rule” is that “the period for assessing any tax imposed by subtitle A [i.e., income tax] with respect to any person which is attributable to any partnership item (or affected item) for a partnership taxable year shall not expire before the date which is [three] years after the later of” either filing of the partnership's return or the return's due date. I.R.C. § 6229(a) (emphases added). If an FPAA “with respect to any taxable year is mailed to the tax matters partner,” the limitations period in § 6229(a) “shall be suspended—(1) for the period during which an action may be brought under [§] 6226 (and, if a petition is filed under [§] 6226 with respect to such administrative adjustment, until the decision of the court becomes final), and (2) for [one] year thereafter.” Id. § 6229(d). Taken together, I.R.C. § 6229(a) and (d) provide that the issuance of an FPAA for a given year “suspend[s]” the limitations period for assessing “any tax” of “any person” that is “attributable to” “any partnership item” for that year.

B. “Attributable to” Means Due to, Caused by, or Generated By

The central issue here is whether the losses that FFIP allocated in 2001 to Ms. Zimmerman were “attributable to” the loss reported by RRF in 2000 under § 6229(a). This is a question of statutory interpretation and, thus, “our inquiry begins with the statutory text.” BedRoc Ltd., LLC v. United States, 541 U.S. 176, 183 (2004) (citations omitted). “The plain meaning of legislation should be conclusive, except in the rare cases in which the literal application of a statute will produce a result demonstrably at odds with the intentions of its drafters.” United States v. Ron Pair Enters., Inc., 489 U.S. 235, 242 (1989) (internal quotation marks, brackets, and citation omitted). When interpreting another provision of the Internal Revenue Code, we explained that “attributable to” “is not defined anywhere in the [Internal Revenue] Code and has no special technical meaning under the tax laws.” Electrolux Holdings, Inc. v. United States, 491 F.3d 1327, 1330 (Fed. Cir. 2007) (citation omitted). We noted that, in tax cases, various other courts “have construed the phrase according to its plain meaning, which is understood to be ‘due to, caused by, or generated by.’ ” Id. at 1330–31 (citations omitted) (collecting cases); see Keener v. United States, 551 F.3d 1358, 1365 (Fed. Cir. 2009) (same).

Other principles of statutory construction reinforce this interpretation. First, “term[s] should be construed, if possible, to give [them] a consistent meaning throughout” the relevant statutory scheme. Gustafson v. Alloyd Co., 513 U.S. 561, 568 (1995). Interpreting “attributable to” in § 6229(a) differently from how it is interpreted in other Internal Revenue Code provisions, i.e., “due to, caused by, or generated by,” would violate this principle. Second, “limitations statutes barring the collection of taxes otherwise due and unpaid are strictly construed in favor of the Government.” Badaracco v. Comm'r, 464 U.S. 386, 392 (1984) (internal quotation marks and citation omitted); see Bufferd v. Comm'r, 506 U.S. 523, 527 n.6 (1993) (stating that, even where the statute of limitations for assessments is ambiguous, if “the Commissioner's construction ․ is a reasonable one ․ [courts] should accept it absent convincing grounds for rejecting it”). As such, § 6229(a) should be interpreted broadly and the IRS's interpretation, if reasonable, should be given deference. Defining “attributable to” in § 6229(a) to mean “due to, caused by, or generated by” preserves the phrase's plain meaning, maintains consistency with the phrase's interpretation elsewhere in the Internal Revenue Code, and follows the Supreme Court's precedent for affording the IRS deference in interpreting the Internal Revenue Code. Therefore, we see no reason why that same definition should not apply here.

C. The Court of Federal Claims Did Not Err in Determining that the Losses Claimed on Ms. Zimmerman's 2001 Individual Tax Return Are “Attributable to” the Loss Claimed on RRF's 2000 Tax Return

Applying that definition of “attributable to” here, the 2005 FPAA suspended the limitations period for assessing any tax against Ms. Zimmerman that was “due to, caused by, or generated by” any partnership item on her 2001 individual tax return. The parties do not dispute that the IRS issued the 2005 FPAA to RRF less than three years after Ms. Zimmerman filed her 2001 individual tax return. RRF I, 101 Fed. Cl. at 500; see J.A. 151. And, as explained above, RRF allocated the loss claimed in its 2000 tax return to FFIP, much of which FFIP passed through in its 2000 and 2001 tax returns to Ms. Zimmerman, who claimed these losses in her 2001 individual tax return. RRF I, 101 Fed. Cl. at 500. Thus, the losses Ms. Zimmerman claimed on her 2001 tax return were “generated by” the loss claimed on RRF's 2000 tax return, and the 2005 FPAA suspended the limitations period for assessing taxes on these losses.

This interpretation of “attributable to” also comports with the Internal Revenue Code's reasonable policy of treating partnership items at their source. Generally, “the tax treatment of any partnership item (and the applicability of any penalty, addition to tax, or additional amount which relates to an adjustment to a partnership item) shall be determined at the partnership level.” I.R.C. § 6221. Pursuant to this principle, the tax liability of an indirect partner 2 depends upon the partnership items, and “[a]ll adjustments required to apply the results of a proceeding with respect to a partnership ․ to an indirect partner shall be treated as computational adjustments.” I.R.C. § 6231(a)(6); see Sente Inv. Club P'ship v. Comm'r, 95 T.C. 243, 249 (1990) (applying § 6231(a)(6)). Computational adjustments are “change[s] in the tax liability of a partner which properly reflects the treatment ․ of a partnership item.” I.R.C. § 6231(a)(6) (emphasis added). The IRS's actions here fall squarely within the definition of a computational adjustment because the IRS “change[d] ․ the tax liability of one [indirect] partner,” i.e., Ms. Zimmerman, “to properly reflect[ ] the treatment ․ of a partnership item,” i.e., the loss claimed in RRF's 2000 tax return. Id. As a result, the IRS properly adjusted the partnership item at its source.3

D. RRF's Counterarguments Are Unpersuasive

RRF presents two counterarguments, neither of which is persuasive. First, RRF argues that “an item can only be a partnership item of a single partnership.” Appellant's Br. 25. According to RRF, “the [G]overnment conceded that the assessment at issue was attributable to ‘a 2001 FFIP partnership item.’ ” Id. (quoting J.A. 973).4 RRF contends that “the partnership item at issue is, and can be, a partnership item of FFIP and only FFIP.” Id. In support, RRF avers that permitting a “partnership item” to be attributable to multiple partnerships would disregard Congress's intent “ ‘to simplify the procedures' for partnership tax proceedings.” Id. (brackets omitted) (quoting Transpac Drilling Venture v. United States, 16 F.3d 383, 387 (Fed. Cir. 1994)). Vague references to the objective of simplifying partnership tax proceedings are insufficient to demonstrate that the plain meaning of “attributable to” “will produce a result demonstrably at odds with the intentions of its drafters,” and, thus, the plain language is “conclusive.” Ron Pair Enters., 489 U.S. at 242.

Second, RRF contends that our interpretation of § 6229 would “violate [ ] the tax system's bedrock annual accounting principle,” i.e., that “taxes are to be determined on an annual basis.” Appellant's Br. 29 (citing Burnet v. Sanford & Brooks Co., 282 U.S. 359, 364–65 (1931)). Specifically, RRF alleges that our interpretation ignores the Supreme Court's instruction that, “[a]bsent other specific directions from Congress, [Internal Revenue] Code provisions must be interpreted so as to conform to the basic premise of annual tax accounting.” Comm'r v. Gordon, 391 U.S. 83, 96 (1968) (footnote omitted); see Appellant's Br. 30. The annual tax accounting principle concerns the annual calculation of tax liabilities based on receipts and deductions, not the limitations period to assess a tax. See United States v. Skelly Oil Co., 394 U.S. 678, 680–81 (1969) (explaining the procedures for calculating annual tax liabilities under the annual accounting principle). Indeed, “it is well settled that the IRS and the courts may recompute taxable income in a closed year in order to determine the tax liability in an open year.” Barenholtz v. United States, 784 F.2d 375, 380–81 (Fed. Cir. 1986) (footnote omitted). That is precisely what has occurred here—the IRS's disallowance of the loss claimed on RRF's 2000 tax return will result in Ms. Zimmerman owing tax for losses claimed in her individual tax returns for 2001 and any later years in which she claimed losses attributable to the 2000 RRF loss, whether she or FFIP carried them over.5

III. The Court of Federal Claims Did Not Err in Determining that Tiger Was Never a Bona Fide Partner in RRF

In RRF II, the Court of Federal Claims concluded that “Tiger had no real intention of becoming a partner in RRF[ ] and that RRF had reason to know that.” 122 Fed. Cl. at 617. Instead, the Court of Federal Claims found “[a] review of the evidence demonstrates that ․ their transaction was a sham, that the transaction lacked economic substance, that the contribution can be ignored, and that the transaction should be characterized as a sale.” Id. RRF argues that the Court of Federal Claims erred by dismissing the relevant provisions of the Internal Revenue Code, focusing on Tiger's subjective intent rather than objective indicia, and ignoring precedent permitting parties to structure transactions to achieve tax advantages. Appellant's Br. 33; see id. at 33−55. Because there was no bona fide partnership between RRF and Tiger, we hold that the Court of Federal Claims did not err.

A. RRF and Tiger Did Not Intend to Form a Bona Fide Partnership

1. The Legal Framework

When a party acquires an economic interest in a partnership, they are only treated as a partner for tax purposes if the partnership “interest is acquired in a bona fide transaction, not a mere sham for tax avoidance or evasion purposes.” Treas. Reg. § 1.704-1(e)(1)(iii) (2015). In determining whether a bona fide partnership exists, the Supreme Court requires that courts evaluate “whether the partners really and truly intended to join together for the purpose of carrying on the business and sharing in the profits and losses or both.” Comm'r v. Culbertson, 337 U.S. 733, 741 (1949) (internal quotation marks and citation omitted). More specifically, the Supreme Court explained that

[t]he question is not whether the services or capital contributed by a partner are of sufficient importance to meet some objective standard ․ , but whether, considering all facts— ․ [including] any ․ facts throwing light on their true intent—the parties in good faith and acting with a business purpose intended to join together in the present conduct of the enterprise.

Id. at 742 (emphases added) (footnote omitted). Contrary to RRF's repeated assertions, the focus of the Culbertson test is “not ․ objective”; it is the parties' “true intent.” Id. The parties' “true intent” is evaluated by “considering all facts,” id., and “[t]riers of fact [who] are constantly called upon to determine the intent with which a person acted” are best able to make these determinations, id. at 743 (footnote omitted).

2. The Court of Federal Claims Did Not Err in its Factual Findings as to RRF's and Tiger's Intent

The Court of Federal Claims “consider[ed] all facts,” id., and determined that it was “clear” that “Tiger had no real intention of becoming a partner in RRF[ ] and that RRF had reason to know that,” RRF II, 122 Fed. Cl. at 617 (emphasis added). But the court did not merely find that RRF “had reason to know” that Tiger intended a sale, not a partnership. The court also found that, as early as April 1999, Tiger's contribution was “part of a plan (of which [RRF principals] were fully aware) to move highly depreciated assets to RRF via Deutsche Bank in a way that preserved their tax characteristics.” Id. at 621. In other words, both players knew before the first transaction that Tiger would sell its CLNs for cash and that RRF would obtain CLNs with massive built-in losses. As the court stated, “Tiger and RRF thus collaborated in a scheme to use the tax laws to their advantage.” Id. We discern no clear error in these findings by the Court of Federal Claims.

Indeed, the Court of Federal Claims's factual findings are thorough, established by the record, and supportive of its conclusion that RRF and Tiger did not form a bona fide partnership. For example, both RRF and FFIP were Bracebridge-managed funds, and Deutsche Bank worked closely with both RRF and FFIP to orchestrate each of the relevant transactions. As the Court of Federal Claims found,

[t]he quickest means of seeing the events in focus is to step back and look for the actions of the common denominator, Deutsche Bank. It was the broker who helped Tiger acquire its Russian assets. It linked Tiger with the Bracebridge funds [i.e., RRF and FFIP]. It helped arrange the transfer of the Tiger assets to RRF. It brokered the sale of Tiger's partnership interest in RRF to FFIP, in the process making certain that the form of the sale did not jeopardize the subsequent transfer of the built-in losses to a third party. It then obtained an option to sell the [CLNs] from RRF and finally arranged a sale to General Cigar. The evidence clearly indicates that RRF was a knowing and willing participant in these activities, at least as of April 1999.

Id. at 620 (emphasis added); see id. at 607–10 (explaining the relevant transactions, including Deutsche Bank's involvement, and providing supporting citations).

This is particularly telling when Tiger (i.e., the only party to the transactions that was not managed by Bracebridge) retained its interest in the partnership for approximately two weeks. See id. at 608–09 (explaining that Tiger retained its interest from either May 20 or 25, 1999 to June 3, 1999); see also J.A. 8792, 8949, 9069. The Court of Federal Claims found that “the evidence is clear that Tiger was interested in the spring of 1999 in selling its position in [the CLNs].” RRF II, 122 Fed. Cl. at 618. And at the time of Tiger's contribution to RRF, Tiger employees were already emailing about the next step: sale. Id. Tiger was at all times interested in liquidity (i.e., a sale), not a partnership. Because Culbertson requires that the parties “act [ ] with a business purpose [and] intend[ ] to join together in the present conduct of the enterprise,” 337 U.S. at 742, it is highly significant that Tiger refused to sign the standard subscription agreement stating that “the Shares subscribed for hereby are being acquired ․ for investment purposes only, ․ and not with the view that any resale or distribution thereof ․,” RRF II, 122 Fed. Cl. at 607 (emphases added) (citing J.A. 8178); see J.A. 8853, 8900, 8945–48.

Moreover, relying on the Government's experts, the Court of Federal Claims found “that Tiger's entry into RRF made no sense as an investment, and its exit made no sense in terms of timing.” RRF II, 122 Fed. Cl. at 618–19. The transaction neither diversified Tiger's investment portfolio (RRF II, 122 Fed. Cl. at 611; see J.A. 3680–81, 5725–26) nor provided Tiger with any additional expertise (RRF II, 122 Fed. Cl. at 614, 619; see J.A. 3678–80, 5724–26). As one of the world's largest management companies, Tiger was already paying its own experts and would have been better off managing its own CLNs rather than “paying for nothing”—i.e., paying management fees to RRA—and committing to any kind of lockup period. Id. at 619. In addition, Tiger did not perform basic due diligence prior to the acquisition (RRF II, 122 Fed. Cl. at 619; see J.A. 3655–56, 3741–43, 5541, 5637), and it sold the CLNs to FFIP two weeks later at a discount even though the value had increased during that short period. Id. at 609.

As for RRF, the Court of Federal Claims found that “there is a massive amount of circumstantial evidence that RRF was aware early on that Tiger had no real interest in becoming a partner,” and it concluded that RRF “was a willing participant at some point [at least as of April 1999] in facilitating the transfer of assets through the sham partnership.” Id. at 620. For example, emails between RRF principals in April and May 1999, before the first transaction, revealed RRF's knowledge that Tiger intended to engage in a sale and that it would be important to preserve the tax basis of Tiger's contribution for that future sale. Id. at 606, 621.

These factual findings are sufficient to demonstrate that neither RRF nor Tiger intended to form a bona fide partnership under the Culbertson standard.

3. The Court of Federal Claims Did Not Err in Its Legal Conclusion that RRF and Tiger Did Not Form a Bona Fide Partnership

Lacking any basis to challenge the Court of Federal Claims's factual findings, RRF argues that the Court of Federal Claims erred in its selection of the appropriate legal standard and the legal conclusions it drew from its underlying factual findings. See Appellant's Br. 33–55. RRF's primary argument is that the Court of Federal Claims improperly ignored sections of the Internal Revenue Code that dictate that “Tiger was a partner[ ] and [that] the built-in losses on the property contributed by Tiger properly transferred to the partnership.” Appellant's Br. 34; see id. 34–40 (citing to I.R.C. §§ 704(c), (e)(1), 721(a), 761(b)). By ignoring these provisions, RRF alleges that the Court of Federal Claims “eschewed the time-tested and congressionally mandated standard for determining partnership formation in favor of its own test, under which objective indicia of partnership intent are disregarded as mere ‘formalities' and one party's unilateral intent can invalidate the partnership.” Id. at 40; see id. at 40–50.

The Court of Federal Claims did not apply “its own test,” id. at 40; it applied the Supreme Court's. Under Culbertson, the focus of the inquiry is the parties' “true intent,” 337 U.S. at 742, which is determined by “considering all the facts,” id. Contrary to RRF's assertions, the Court of Federal Claims considered the totality of the circumstances, see, e.g., RRF II, 122 Fed. Cl. at 607–08 (discussing the revisions to the subscription agreement mandated by Tiger), 612 (discussing RRF's expert's estimate that RRF's rate of return was 225% for 1999 and 105% for 2000), 614 (discussing Tiger's “ability to do its own market and asset analysis”), and determined that RRF's and Tiger's actions were mere “formalities,” id. at 619. The Court of Federal Claims weighed all of the relevant factors, i.e., made underlying factual findings, and applied the appropriate legal standard to these findings to determine that RRF and Tiger did not enter into a bona fide partnership, i.e., reached a legal conclusion. This is exactly what is required by both Culbertson and our precedent. See 733 U.S. at 742 (requiring courts to “consider[ ] all facts” to determine the parties' “true intent,” i.e., that “the parties in good faith and acting with a business purpose intended to join together in the present conduct of the enterprise”); Wells Fargo, 641 F.3d at 1325 (stating that “the characterization of transactions for tax purposes” is a legal issue that is “based on underlying findings of fact” (citation omitted)). We find no error in the Court of Federal Claims's factual findings and agree with its legal conclusion.

RRF's arguments to the contrary are unpersuasive. First, RRF contends that Culbertson does not apply here because Congress has provided the standard governing partnership formation. Appellant's Br. 40–46. In support, RRF cites to a general statement from the D.C. Circuit that Culbertson does not supersede clear Congressional intent. Id. at 41 (citing Horn v. Comm'r, 968 F.2d 1229, 1231 (D.C. Cir. 1992) (“Although useful in determining congressional intent and in avoiding results unintended by tax code provisions, the [Culbertson] doctrine cannot trump the plainly expressed intent of the legislature.”)). However, Culbertson clearly articulates the standard for determining whether a partnership is bona fide, and we are bound by Culbertson until either the Supreme Court or Congress overrules it. Accord Hohn v. United States, 524 U.S. 236, 252–53 (1998) (“Our decisions remain binding precedent until we see fit to reconsider them, regardless of whether subsequent cases have raised doubts about their continuing vitality.” (citation omitted)); Dickerson v. United States, 530 U.S. 428, 437 (2000) (“Congress retains the ultimate authority to modify or set aside any judicially created rules of evidence and procedure that are not required by the Constitution.” (citations omitted)). In addition, even if we were bound by Horn, which we are not, see Int'l Custom Prods., Inc. v. United States, 843 F.3d 1355, 1360 (Fed. Cir. 2016) (“When our precedent is silent on a particular question, we may look to another circuit for guidance and may be persuaded by its analysis, though decisions from other circuits are not binding on this court.” (internal quotation marks and citation omitted)), Horn does not provide that objective indicia should serve as the foundation of a court's analysis of whether a partnership is bona fide. In fact, Horn does not mention “partnership” at all. See generally 968 F.2d 1229. Horn is simply inapposite.

Second, RRF argues that the Court of Federal Claims incorrectly focused on Tiger's unilateral intent. Appellant's Br. 46–50. For example, RRF states that the Court of Federal Claims “dwelled on its finding that Tiger had no real interest in being a long term investor. ․ But the [Court of Federal Claims] could never explain why this mattered to partnership formation.” Id. at 47. RRF overlooks that Culbertson explicitly counsels that both parties must intend to form a partnership, meaning that both RRF's and Tiger's intent were relevant. See 337 U.S. at 742 (repeatedly referring to the intent of the “parties” (emphasis added)). In addition, contrary to RRF's assertions, the Court of Federal Claims did not look to Tiger's unilateral intent; instead, it found that RRF both knew of and shared in Tiger's intention. See, e.g., RRF II, 122 Fed. Cl. at 609 n.15 (“The balance of the evidence of RRF's knowledge of what was really happening is so overwhelming ․”), 621 (“Tiger and RRF ․ collaborated in a scheme to use the tax laws to their advantage. ․ [W]e are not obligated to give them effect when their sole intent was to avoid treating the ․ transaction as what it was, a sale.”).

Finally, RRF asserts that the Court of Federal Claims incorrectly relied on its finding that the parties had “use[d] the tax laws to their advantage.” Appellant's Br. 50 (quoting RRF II, 122 Fed. Cl. at 621). It is true that a “taxpayer has an unquestioned right to decrease or avoid his taxes by means which the law permits.” Coltec Indus., Inc. v. United States, 454 F.3d 1340, 1355 (Fed. Cir. 2006) (citation omitted). However, this was not the foundation of the Court of Federal Claims's holding. Instead, it determined that RRF's and Tiger's “sole intent” was manipulating the tax code, RRF II, 122 Fed. Cl. at 621, and, thus, that they lacked the “true intent” to form a bona fide partnership, Culbertson, 733 U.S. at 742.

B. RRF's Transaction with Tiger Lacked Economic Substance

Even had RRF and Tiger intended to form a bona fide partnership, the Court of Federal Claims correctly determined that RRF's transaction with Tiger fails under the economic substance doctrine, see RRF II, 122 Fed. Cl. at 617, which “prevent[s] taxpayers from subverting the legislative purpose of the tax code by engaging in transactions that are fictitious or lack economic reality simply to reap a tax benefit,” Coltec, 454 F.3d at 1353–54. We have articulated five principles guiding our analysis as to the economic substance doctrine, four of which are relevant here.

“First, although the taxpayer has an unquestioned right to decrease or avoid his taxes by means which the law permits, ․ the law does not permit the taxpayer to reap tax benefits from a transaction that lacks economic reality.” Id. at 1355 (citation omitted). RRF argues that the Court of Federal Claims “did not find tax avoidance was RRF's sole motive.” Appellant's Br. 51. That is untrue. The Court of Federal Claims found that RRF's and Tiger's “sole intent was to avoid treating the ․ transaction as what it was, a sale,” by “collaborat[ing] in a scheme to use the tax laws to their advantage.” RRF II, 122 Fed. Cl. at 621 (emphasis added). Although RRF claims that this finding is “unsupported and contradicted,” Appellant's Br. 52, we disagree. RRF has not demonstrated that the Court of Federal Claims's factual findings were unsupported by the record, as explained above. See supra Section III.A.2. Nor has RRF shown that these findings are contradicted, as the purportedly contradictory findings primarily relate to the formation of RRF, not RRF's transaction with Tiger. See Appellant's Br. 51–53.

“Second, when the taxpayer claims a deduction, it is the taxpayer who bears the burden of proving that the transaction has economic substance.” Coltec, 454 F.3d at 1355. RRF claimed a deduction for a loss that was passed through to FFIP and then to Ms. Zimmerman. See RRF I, 101 Fed. Cl. at 500. Because RRF claimed the deduction, it “bears the burden of proving that the transaction has economic substance.” Coltec, 454 F.3d at 1355. RRF has not met that burden.

“Third, the economic substance of a transaction must be viewed objectively rather than subjectively.” Id. at 1356. There are some objective indicators of the economic reality of the transaction, such as RRF's expert's testimony that, under RRF's business model, RRF had a legitimate reason to provide shares instead of paying cash for the CLNs. RRF II, 122 Fed. Cl. at 611. However, the great bulk of the objective evidence indicates that the Tiger transaction lacked economic substance, including Tiger's quick sale of its RRF shares to FFIP for “approximately $800,000 less than the sales price of the shares roughly one to two weeks earlier,” when the value of the shares had increased during that period. Id. at 609 (footnote omitted). What could have been accomplished via a direct sale of CLNs from Tiger to FFIP was instead carried out via Tiger's contribution of CLNs to RRF and subsequent sale of its partnership interest to FFIP. The former would result in no transfer of Tiger's $230 million in built-in losses, while the latter transferred the built-in losses to U.S. tax-paying entities. Tellingly, RRF arranged for these losses to go entirely to FFIP. See RRF II, 122 Fed. Cl. at 622 (“Mr. DiBiase ended with a ‘challenge’ to the accountants: ‘Get tax losses from [CLNs] to FFIP. Don't want any of such losses to be allocated to other entities [i.e., RRF partners] which will get no benefit from them.’ ” (citation omitted)). This principle supports the Government.

“Fourth, the transaction to be analyzed is the one that gave rise to the alleged tax benefit.” Coltec, 454 F.3d at 1356. Here, the transaction that gave rise to the tax benefit is RRF's exchange of shares for Tiger CLNs and, relying on expert testimony, the Court of Federal Claims found that Tiger “was gaining nothing” from this transaction. RRF II, 122 Fed. Cl. at 619. RRF has not demonstrated any reason to disturb this finding. See Appellant's Br. 51–53.

“Finally, arrangements with subsidiaries that do not affect the economic interest of independent third parties deserve particularly close scrutiny.” Coltec, 454 F.3d at 1357. Because this transaction is not between subsidiaries, this factor is not relevant to our analysis. In sum, four of the five factors indicate that RRF's transaction with Tiger lacked economic substance.

IV. The Court of Federal Claims Did Not Err in Determining that Penalties Applied

The Court of Federal Claims upheld the 40% penalty that the IRS imposed because RRF “did not reasonably rely on objective advice from a tax professional based on all of the pertinent laws, facts, and circumstances.” RRF II, 122 Fed. Cl. at 623–24. RRF argues that imposing penalties (1) violates the Internal Revenue Code's “basic principle ․ that no penalty can be imposed when a taxpayer's view is reasonable and in good faith, ․ even if a court ultimately disagrees,” Appellant's Br. 56, and (2) ignores “that RRF reasonably relied on its tax experts' advice,” id. at 58 (citation omitted). We disagree.

Although partnerships do not pay income tax, I.R.C. § 701, “the applicability of any penalty ․ which relates to an adjustment to a partnership item” is determined at the partnership level, id. § 6221. When a taxpayer underpays, the IRS “shall ․ add[ ] to the tax an amount equal to 20 [%] of the portion of the underpayment,” id. § 6662(a), and this penalty “shall” be increased to 40% for “gross valuation misstatements,” id. § 6662(h)(1). However, “[n]o penalty shall be imposed ․ with respect to any portion of an underpayment if it is shown that there was a reasonable cause ․ and that the taxpayer acted in good faith ․” Id. § 6664(c)(1). Section 6664(c)(1) is a “narrow defense,” and “[t]he taxpayer bears the burden of showing this exception applies.” Stobie Creek Invs. LLC v. United States, 608 F.3d 1366, 1381 (Fed. Cir. 2010). Reliance on a professional tax advisor's advice may provide such a defense if, inter alia, the advice is “based upon all pertinent facts and circumstances and the law as it relates to those facts and circumstances,” Treas. Reg. § 1.6664-4(c)(1)(i) (2003), and is “not ․ based on unreasonable factual or legal assumptions” or “unreasonably rel[iant] on the representations ․ of the taxpayer,” id. § 1.6664-4(c)(1)(ii).

We agree with the Court of Federal Claims that RRF cannot meet its burden. As to RRF's first argument, the Court of Federal Claims did not apply novel reasoning based on a new legal standard. Instead, it applied longstanding Supreme Court precedent, i.e., Culbertson.

As to RRF's second argument, the Court of Federal Claims found that Mr. DiBiase supplied all of the information upon which Ernst & Young relied. See, e.g., RRF II, 122 Fed. Cl. at 622 (stating that “the list of working ‘facts' behind E[rnst] & Y[oung]'s preparation of RRF's tax return were orchestrated by Mr. DiBiase to achieve a desired result and were not critically evaluated by” Ernst & Young's representatives), 623 (Ernst & Young “simply took at face value Mr. DiBiase's self-interested summary and utilized these ‘facts' to prepare the tax forms.”). These conclusions are well-supported by the record. See, e.g., J.A. 2576–77 (confirming that the tax group at Ernst & Young “accepted the information that was supplied by Mr. DiBiase as correct” and that “the tax group at Ernst & Young did no independent investigation into the factual accuracy of the information that Mr. DiBiase supplied”), 9350 (fax from Ernst & Young raising concerns about ACM P'ship v. Comm'r, 157 F.3d 231 (3d Cir. 1998), a case involving the economic substance doctrine, that Ernst & Young did not address elsewhere in the record). This constitutes “unreasonabl[e] rel [iance] on the representations ․ of the taxpayer,” Treas. Reg. § 1.6664-4(c)(1)(ii), which does not satisfy the requirements of I.R.C. § 6664(c)(1).

Indeed, the only evidence that RRF offered the Court of Federal Claims of any “advice” that Ernst & Young provided is the tax returns themselves. See RRF II, 122 Fed. Cl. at 623 (“[T]he only record [RRF] offers of ‘advice’ given to RRF concerning the propriety of taking the losses is the returns themselves. There are no backup memos or records of conversations concerning the propriety of claiming built-in losses. We are simply asked to accept that, by signing off on the returns for 1999 and 2000, E[rnst] & Y[oung] was giving its considered advice on whether it was appropriate to take the loss deduction.”). The same is true on appeal. See Appellant's Br. 60–62 (arguing that tax returns are advice). However, tax returns are insufficient to demonstrate reliance on professional tax preparer advice for the reasonable cause exception. See Richardson v. Comm'r, 125 F.3d 551, 558 (7th Cir. 1997) (finding no reasonable cause where, “other than the fact that a tax preparer signed [the taxpayer's] returns, there [was] no evidence in the record that [the taxpayer] received any advice from professionals”); Neonatology Assocs., P.A. v. Comm'r, 115 T.C. 43, 100 (2000) (“The mere fact that a certified public accountant has prepared a tax return does not mean that he or she has opined on any or all of the items reported therein.”), aff'd, 299 F.3d 221 (3d Cir. 2002).


We have considered RRF's remaining arguments and find them unpersuasive. For these reasons, the final decision of the U.S. Court of Federal Claims is



1.   RRF claimed the balance of the Tiger built in losses—approximately $171 million—on its 2004 return upon redeeming its preferred stock in General Cigar in 2004. RRF II, 122 Fed. Cl. at 610.

2.   An indirect partner is “a person holding an interest in a partnership through [one] or more pass-thru partners,” I.R.C. § 6231(a)(10), and a pass-thru partner is “a partnership ․ or other similar person through whom other persons hold an interest in [another] partnership,” id. § 6231(a)(9). The Court of Federal Claims explained that FFIP is a direct partner of RRF and that Ms. Zimmerman is a direct partner of FFIP and an indirect partner of RRF. RRF I, 101 Fed. Cl. at 500.

3.   RRF concedes that if FFIP had simply passed through all of RRF's loss in 2000, the losses reported by the indirect partners would be “attributable to” RRF's loss. Appellant's Br. 28 n.7. RRF's position founders on the shoals of that concession. In Sente, the IRS issued an FPAA to a pass-thru partner rather than to the partnership that was the source of the reported losses. See 95 T.C. at 245. The Tax Court determined that it lacked jurisdiction, requiring that the flow-through losses be addressed in proceedings directed at the source partnerships instead of the pass-thru partner. Id. at 248. Here, the IRS issued the 2005 FPAA to RRF (the source partnership), not FFIP (the pass-thru partner), as required by Sente. See RRF II, 122 Fed. Cl. at 621. The only evidence that RRF has identified to demonstrate that FFIP's role was materially different from the pass-thru partner's role in Sente is that FFIP carried over some of the 2000 RRF loss to 2001. However, that action changed the year of the pass through, not the character of the losses, which are still “attributable to” the 2000 RRF loss.

4.   The Government, however, did not state that the Tiger losses are a 2001 FFIP partnership item and, more importantly, did not state that Ms. Zimmerman's underpayment was “attributable to” a 2001 FFIP partnership item. See J.A. 973 (the Government clarifying that the losses are a 2000 RRF partnership item and a 2001 “affected partnership item”); compare I.R.C. § 6231(a)(3) (defining “partnership item”), with id. § 6231(a)(5) (defining “affected item”).

5.   RRF also contends that the Court of Federal Claims improperly “traced” the losses “back through to items from different partnerships,” contrary to “Electrolux's instruction to focus on the direct cause, i.e., the partnership item.” Appellant's Br. 27. However, in Electrolux, we found that the carryover to 1995 was “attributable to” the 1994 capital loss, which was the “original source,” 491 F.3d at 1331; it was not “attributable to” the 1993 carryback, which was an intermediate step rather than the “direct[ ] cause” of the 1995 carryover, id. at 1332. Similarly, Ms. Zimmerman's 2001 losses are “attributable to” the original 2000 RRF loss, not the intermediate carryover by FFIP.

WALLACH, Circuit Judge.

Copied to clipboard