Alpha I, L.P. Ex Rel. Sands v. United States

682 F.3d 1009, 2012 WL 2161297, 109 A.F.T.R.2d (RIA) 2548, 2012 U.S. App. LEXIS 12216

• Court: Court of Appeals for the Federal Circuit
• Decided: June 15, 2012
• Docket: 2011-5024, 2011-5030
• Status: Published

Opinion

O’MALLEY, Circuit Judge.

The U.S. Court of Federal Claims dismissed the Internal Revenue Service’s determination that certain taxpayers’ transfers of their partnership interests to trusts were shams because the court believed it lacked jurisdiction to address the IRS’s determination at the partnership level. The United States appeals that ruling in Case No. 2011-5024. We reverse the Court of Federal Claims’s dismissal. The identity of the true partners in the partnership at issue appropriately is determined at the partnership level because, on the particular facts of this case, partnership identity could affect the distributive shares reported to the partners.

After this action commenced, the taxpayers conceded certain capital gain and loss adjustments imposed by the IRS and moved for summary judgment on the valuation misstatement penalties that the IRS sought as a result of those adjustments. The taxpayers argued that their concession of the IRS’s adjustments rendered the valuation misstatement penalties moot. The Court of Federal Claims agreed, granted summary judgment to the taxpayers, and declined to impose the valuation misstatement penalties. The United States appeals that ruling in Case No. 2011-5024. We vacate that judgment because the Court of Federal Claims failed to determine whether the taxpayers’ underpayment of tax was attributable to the alleged valuation misstatement.

Finally, the Court of Federal Claims granted the government’s motion for summary judgment with respect to additional penalties for negligence, substantial understatement, and failure to act reasonably and in good faith, and imposed a twenty-percent penalty on the taxpayers. The taxpayers appeal that ruling in Case No.2011-5080. We dismiss the taxpayers’ appeal as premature. If the Court of Federal Claims concludes on remand that the forty-percent valuation misstatement penalty applies, that valuation misstatement penalty could render moot the propriety of the twenty-percent penalty that is the subject of the taxpayers’ cross appeal.

I

This case arises from two so-called “Son-of-BOSS” transactions, as well as a transaction involving charitable remainder unitrusts (“CRUTs”), conducted by the heirs of the late Marvin Sands, the founder of Constellation Brands, Inc. In a Son-of-BOSS transaction, a taxpayer attempts to realize tax benefits by transferring assets encumbered by significant liabilities to a partnership in an attempt to increase the partner’s basis in the partnership. See Stobie Creek Invs. LLC v. United States, 608 F.3d 1366, 1368-71 (Fed.Cir.2010); Kornman & Assocs. v. United States, 527 F.3d 443, 446 n. 2 (5th Cir.2008). Normally, when a partner contributes property to a partnership, the partner’s basis in the partnership increases, and when the partnership assumes a partner’s liability, the partner’s basis decreases. See I.R.C. §§ 722, 733, 752, 754. A Son-of-BOSS transaction recognizes a partnership’s acquisition of a partner’s asset (here, short-sale proceeds), and disregards the partnership’s acquisition of an essentially offsetting liability (here, the obligation to close out the short-sale position). See Stobie Creek Invs., 608 F.3d at 1368-69; Kornman & Assocs., 527 F.3d at 446 n. 2. By employing this strategy, the taxpayer attempts to generate a tax loss or reduce the gain that would otherwise result from the sale of an asset. Id.

In this ease, Marvin Sands’s heirs owned stock in Constellation. In the first Son-of-BOSS transaction, they used several partnerships to convert approximately $66 million in taxable gain that they anticipated receiving from the sale of their stock into large capital losses. The heirs also prearranged for their partnership interests to be held temporarily by tax-exempt CRUTs at the time of the sale so that, as the government alleges, any gain that might be recognized from the sale would escape taxation. The CRUTs were terminated shortly after they were formed, and the assets of the CRUTs, including the sale proceeds, were distributed to the heirs, purportedly tax free. In the second Son-of-BOSS transaction, the heirs sought to generate significant capital losses to offset other income, again through the use of various partnerships.

In notices of final partnership administrative adjustment (“FPAAs”) issued to the partnerships involved in the Son-of-BOSS transactions, the IRS determined that the transactions should be disregarded and that the transfers of the partnership interests to the CRUTs were shams. The IRS also asserted various basis and capital gain and loss adjustments, as well as several alternative penalties, including a forty-percent penalty for gross valuation misstatement, a twenty-percent penalty for substantial understatement of tax, and a twenty-percent penalty for negligence. The IRS also asserted that the transactions did not increase the partners’ amounts at risk under I.R.C. § 465.

The partnerships involved in the Son-of-BOSS transactions initially challenged the IRS’s adjustments to the basis, capital gain, and capital loss calculations. In an amended complaint, however, the partnerships conceded the capital gain and loss adjustments on the purported basis of I.R.C. § 465. The Court of Federal Claims later agreed with the partnerships that, because the adjustments had been conceded on the basis of I.R.C. § 465, the forty-percent gross valuation misstatement penalty sought by the IRS because of the adjustments was inapplicable. The Court of Federal Claims also held that the identity of a partnership’s partners is a non-partnership item that cannot be addressed in a partnership proceeding, such that it could not consider whether the transfers of the partnership interests to the CRUTs were shams. Finally, the Court of Federal Claims determined that the twenty-percent penalties for substantial understatement of tax and negligence applied and imposed such a penalty on the taxpayers.

A

During 2001 and 2002, the years at issue, Constellation was a leading producer and marketer of alcoholic beverages in North America and the United Kingdom, with gross sales exceeding $3 billion in fiscal year 2001. Constellation was founded and owned by the late Marvin Sands. After his death, the following family members held a controlling interest in Constellation through stock ownership: Marvin’s sons, Robert and Richard Sands; his wife, Marilyn Sands; and two trusts established for the benefit of his grandchildren, Abigail Stern and Zachary Stern (the “Children’s Trusts”). The government claims that, in 2001, the heirs’ stock was worth more than $75 million and had a tax basis of approximately $9 million.

The heirs received tax advice from The Heritage Organization, LLC, which designed and directed the implementation of the two Son-of-BOSS transactions. The heirs implemented the first Son-of-BOSS transaction as follows. Between August 21 and 23, 2001, they established brokerage accounts with Paine Webber. Through those accounts, they collectively sold short approximately $85.6 million of U.S. Treasury Notes. 1 On August 27, 2001, the Children’s Trusts assigned their portion of the short-sale proceeds and the obligation to close out the short sale to CWC Partnership I (“CWC”), a preexisting family investment partnership. On August 28, 2001, the heirs and CWC contributed (i) the proceeds from the short sale, (ii) the obligation to close out the short sale, and (iii) a total of 2,000,000 shares of Constellation stock to R, R, M & C Group, L.P. (“RRMC Group”), a new partnership created at the direction of Heritage. The general partner of RRMC Group was R, R, M & C Management Corporation (“RRMC Corp.”), an entity created by Richard and Robert on August 23, 2001. For its claimed 0.1% interest in RRMC Group, RRMC Corp. contributed 2,002 shares of Constellation stock to the partnership.

On August 31, 2001, RRMC Group, in turn, contributed the Constellation stock, the short-sale proceeds, and the obligation to close out the short sale to R, R, M & C Partners, L.L.C. (“RRMC Partners”), another new entity established at the direction of Heritage. RRMC Group held a 99.7168% interest in RRMC Partners. The remaining interest was held by Gloria Robinson, the mother of the heirs’ accountant.

On September 6, 2001, RRMC Partners closed out the short-sale position at a net loss of $425,565. On September 10, 2001, RRMC Group purchased Robinson’s interest in RRMC Partners, thereby effecting a termination of RRMC Partners. The government alleges that, as a result of this transaction, RRMC Group claimed that its basis in the Constellation stock increased by approximately $85.6 million, from $9 million to $94.7 million.

In connection with the first Son-of-BOSS transaction, the heirs prearranged to channel the sale of the Constellation stock through tax-exempt CRUTs. A CRUT is a charitable trust that provides an income beneficiary, often the grantor, annual payments for a fixed term. At the end of the fixed term, a charity receives the remainder interest. See I.R.C. § 664. A CRUT generally is exempt from tax, but the income of a CRUT is taxable to its income beneficiaries upon distribution. I.R.C. § 664(b), (c)(1).

On September 21, 2001 — eleven days after RRMC Partners had terminated and distributed the Constellation stock back to RRMC Group with the increased basis— Richard, Robert, Marilyn, and CWC each created a CRUT with a twenty-year term. Richard and Robert were designated the trustees of the CRUTs, and the heirs and CWC were named as the income beneficiaries. On the same day, Richard, Robert, Marilyn, and CWC transferred their respective interests in RRMC Group, which held the Constellation stock, to the CRUTs. At that time, appraisals were obtained valuing each partnership interest at $5,198,897. Based on these appraisals, each of the four transferors claimed a $519,935 charitable contribution deduction with respect to the remainder interests purportedly transferred.

On October 1, 2001, RRMC Group sold the Constellation stock for approximately $75 million. The government claims that the actual basis of the stock was approximately $9 million and that RRMC Group should have realized a $66 million gain. As the government further claims, however, RRMC Group claimed a $20 million loss because the heirs inflated the basis through the Son-of-BOSS transaction.

On January 28, 2002, the Sands Supporting Foundation was designated as the charitable beneficiary of the CRUTs. On February 22, 2002, that charity designation was revoked, and the Educational and Health Support Fund, an entity created by Robert and Richard on the same day, was named as the charitable beneficiary. Updated appraisals valued each CRUT’s partnership interest at $5,482,334. Based upon the updated appraisals, on February 27, 2002, Richard, Robert, Marilyn, and CWC purchased from the Educational and Health Support Fund the remainder interests in the CRUTs for $550,000 each. This had the effect of terminating each CRUT (because the income and remainder interests were merged), and the partnership interests in RRMC Group, which held the $75 million from the sale of the Constellation stock, were distributed back to the heirs and CWC. The heirs and CWC claimed that these distributions were tax-free distributions of the CRUTs’ assets.

In the second Son-of-BOSS transaction, on December 3, 2001, the heirs and CWC formed Alpha I, L.P. (“Alpha”), a limited partnership in which they held a cumulative 99.9% interest. The remaining 0.1% interest was held by RRMC Corp. as general partner. By agreement dated December 10, 2001, Alpha and Gloria Robinson, who was involved in the first Son-of-BOSS transaction, formed Beta Partners L.L.C. (“Beta”), with Alpha holding a 99.0043% interest and Robinson holding the remaining 0.9957% interest. On December 12, 2001, the heirs and CWC collectively transferred approximately $1.1 million to their respective Paine Webber accounts. On the same day, the heirs and CWC sold short approximately $44 million of U.S. Treasury Notes. On December 13, 2001, the heirs and CWC contributed the short-sale proceeds, the short-sale obligations, and $1.24 million to Alpha. Shortly thereafter, RRMC Corp. contributed $2,582 to Alpha.

On December 17, 2001, Alpha purchased 67,525 shares of Corning, Inc., common stock for $595,570.50 and 33,400 shares of Yahoo, Inc., common stock for $599,530. On December 20, 2001, Alpha contributed its assets, including the short-sale proceeds, the short-sale obligations, and the Corning and Yahoo stock, to Beta. On December 27, 2001, Beta closed out the short-sale position, realizing a net gain of $90,018. On the same date, Alpha purchased Robinson’s 0.9957% interest in Beta, causing the termination of Beta. As a result of these transactions, Alpha claimed it had a basis in the Corning stock of $23,230,361 and a basis in the Yahoo stock of $22,262,094.

Alpha distributed most of the Corning and Yahoo stock to the heirs and CWC, who, in February 2002, transferred the stock to two other family partnerships: Mickey Management L.P. and M, L, R & R. Each time the stock was transferred, the transferee claimed a carryover basis in the stock equal to Alpha’s allegedly inflated basis. In December 2002, Alpha, Mickey Management, and M, L, R & R each sold a portion of its Corning and Yahoo stock, claiming losses of approximately $9 million total due to the allegedly inflated basis of the stock.

B

The IRS audited the partnerships’ 2001 and 2002 partnership returns. In 2005 and 2006, the agency issued FPAAs denying the losses from the transactions. With respect to the first Son-of-BOSS transaction, the IRS adjusted the basis of the Constellation stock and determined that the sale of the stock resulted in capital gains rather than capital losses. The IRS listed several alternative theories supporting the adjustments, including I.R.C. § 752 and related regulations, the sham-partnership doctrine, and the economic-substance doctrine. The IRS issued similar FPAAs to the partnerships involved in the second Son-of-BOSS transaction, adjusting the basis of the Corning and Yahoo stock and eliminating the loss claimed on the sales of that stock. The IRS also asserted in each FPAA a forty-percent penalty for gross valuation misstatement, or, alternatively, a twenty-percent penalty for substantial valuation misstatement; a twenty-percent penalty for substantial understatement of tax; and a twenty-percent penalty for negligence.

In a section titled “I.R.C. § 465 At Risk Rules,” the IRS further stated in each FPAA that “[i]t is determined that none of the transactions of the Partnership increases the amount considered at-risk for an activity under I.R.C. § 465(b)(1),” such that “the amount for which a partner is considered to be at risk for an activity is not increased by any transactions with the Partnership.”

Finally, in the FPAAs issued to the partnerships involved in the first Son-of-BOSS/CRUTs transaction, the IRS also asserted that the transfers of the RRMC Group partnership interests to the CRUTs should be disregarded as shams. Consequently, the IRS asserted, the proceeds from RRMC Group’s sale of the Constellation stock should not flow through to the CRUTs, but instead should flow through to the heirs and CWC as partners in RRMC Group.

C

In their complaints, the partnerships challenged each of the determinations contained in the FPAAs. The partnerships engaged in the first Son-of-BOSS/CRUTs transaction filed a motion to dismiss the determination that the transfers to the CRUTs were shams. They argued that the identity of RRMC Group’s partners is not a “partnership item,” such that the court lacked jurisdiction under I.R.C. § 6226(f) to decide whether the transfers should be disregarded. The Court of Federal Claims granted the motion to dismiss, holding that the identity of RRMC Group’s partners is not a partnership item and that, as a result, it did not possess jurisdiction to consider the issue.

The partnerships involved in both Son-of-BOSS transactions challenged the forty-percent gross valuation misstatement penalty. With respect to the FPAAs’ I.R.C. § 465 determination, the partnerships asserted that the at-risk amounts were properly computed, and they argued that, even if incorrectly computed, the “at-risk” rules of I.R.C. § 465 could be used only to disallow the claimed losses and would not require a partnership or its partners to recognize any gain.

The partnerships, however, later filed a motion to amend their complaints in which they conceded the capital gain and loss adjustments on the purported basis of I.R.C. § 465. They stated that “[plaintiffs have determined that it would be more economical to narrow the issues before the Court by conceding the defendant’s capital gains adjustments on one of the several alternative grounds asserted by defendants. Plaintiffs believe that such concession also eliminates the possibility of incurring a 40 percent penalty.” They conceded that “none of the transactions of the partnerships increases the amount considered at-risk for an activity under I.R.C. § 465(b)(1) and that the at-risk rules would disallow losses and require the partnerships and their partners to recognize gain on the transactions as described in the adjustments set forth in [the FPAAs].” The partnerships further stated that “[pjlaintiffs do not concede defendant’s capital gains and losses adjustments on any other ground, nor do they concede any other determination set forth in the FPAAs issued to them. However, plaintiffs’ concession of the § 465(b)(1) issue and defendant’s capital gain adjustments eliminates the need for the Court to decide whether defendant’s alternative grounds for such capital gain adjustments ... support the adjustments.” The Court of Federal Claims granted the partnerships’ motion to amend their complaints.

The partnerships then moved for summary judgment on the valuation misstatement penalty. The Court of Federal Claims granted the motion, holding that the partnerships’ concession of the capital gain and loss adjustments on the basis of 1.R.C. § 465 obviated the need to make any valuation determinations and, therefore, rendered the valuation misstatement penalty inapplicable. The government filed a motion for reconsideration, which the court denied, concluding that the government failed to identify error in the court’s summary judgment opinion.

The parties also filed cross-motions for summary judgment on the government’s remaining penalty claims: the twenty-percent negligence penalty and the twenty-percent substantial-understatement-of-tax penalty. The Court of Federal Claims held that both penalties applied. The court found that the partnerships had engaged in a tax shelter based on “the transfer of short sale proceeds to RRMC Group without the transfer of the related contingent obligations....” 2

On September 16, 2010, the Court of Federal Claims entered final judgment consistent with its various opinions.

II

The Court of Federal Claims erred when it dismissed the IRS’s determination that the transfers of the partners’ interests in RRMC Group to the CRUTs were shams. The court’s ruling was on a question of law, which we review de novo. Keener v. United States, 551 F.3d 1358, 1361 (Fed.Cir.2009) (“The Court of Federal Claims’ decision to grant [a] ... motion to dismiss for lack of jurisdiction is a matter of law, which this court reviews de novo.”).

In determining whether the transfers were shams, the Court of Federal Claims was asked to determine the identity of the true partners of RRMC Group. The Court of Federal Claims erred in finding that it lacked jurisdiction to determine partner identity because it incorrectly found that the determination of the identity of the partners in RRMC Group would not affect the allocation of the partnership items among the partners. On the facts of this case, partner identity could, in fact, affect allocation of the partnership items. As such, partner identity is properly resolved at the partnership level in this proceeding.

Whether the Court of Federal Claims may exercise jurisdiction to determine the identities of the partners in RRMC Group turns on whether partner identity must be determined at the partnership or partner level. The statutory framework governing partnership- and partner-level determina tions provides the foundation on which that issue must be resolved.

Partnerships are pass-through entities, meaning they do not pay federal income tax. Rather, all income, deductions, and credits are allocated among the individual partners. I.R.C. §§ 701, 702; Keener, 551 F.3d at 1361. While partnerships are not taxpayers, they are required to file annual information returns reporting the partners’ distributive shares of income, gain, deductions, or credits. I.R.C. § 6031. The individual partners then report their distributive shares on their federal income tax returns. I.R.C. §§ 701-04.

In the past, the differing tax treatment of partnerships and their partners resulted in duplicative audits and litigation, and, often, inconsistent treatment of partners in the same partnership. See Anisa Afshar, The Statute of Limitations for the TEFRA Partnership Proceedings: The Interplay Between Section 6229 and Section 6501, 64 Tax Law. 701 (2011). In 1982, Congress enacted legislation with the goal of establishing coordinated procedures for determining the proper treatment of “partnership items” at the partnership level in a single, unified audit and judicial proceeding. See id.; Tax Equity and Fiscal Responsibility Act of 1982 (“TEFRA”), Pub.L. No. 97-248, § 402 et seq., 96 Stat. 648; and Keener, 551 F.3d at 1361. These procedures are commonly referred to as “TEFRA” procedures. “Whether a tax item is a ‘partnership item’ governs how the TEFRA procedures apply.” Keener, 551 F.3d at 1361.

When the IRS disagrees with a partnership’s reporting of any partnership item, it must issue an FPAA before making any assessments attributable to that item against the partners. I.R.C. §§ 6223(a)(2), (d)(2), 6225(a). The tax-matters partner has ninety days from the date of the FPAA to file a petition contesting the adjustments in the FPAA in the Tax Court, a federal district court, or the Court of Federal Claims. I.R.C. § 6226(a). If no such petition is filed, any other partner entitled to notice of partnership proceedings may file a petition within the following sixty days. I.R.C. § 6226(b)(1).

If a petition contesting the FPAA is filed, the reviewing court’s jurisdiction at the partnership-level proceeding is limited to certain categories. I.R.C. § 6226(f). In this action, the parties dispute whether the identity of RRMC Group’s partners may be categorized as a reviewable item.

I.R.C. § 6226(f) specifies the categories that fall within a reviewing court’s jurisdiction as follows:

A court with which a petition is filed in accordance with this section shall have jurisdiction to determine all partnership items of the partnership for the partnership taxable year to which the notice of final partnership administrative adjustment relates, the proper allocation of such items among the partners, and the applicability of any penalty, addition to tax, or additional amount which relates to an adjustment to a partnership item.

I.R.C. § 6226(f) (emphasis added). As relevant to this action, the Court of Federal Claims may exercise jurisdiction to determine the identity of RRMC Group’s partners if partnership identity is a partnership item for the relevant partnership tax year, and may then determine the proper allocation of such items among the partners. Id. The statute defines a partnership item, in relevant part, as follows:

[A]ny item required to be taken into account for the partnership’s taxable year under any provision of subtitle A [of the tax code] to the extent regulations prescribed by the Secretary pro vide that, for purposes of this subtitle, such item is more appropriately determined at the partnership level than at the partner level.

I.R.C. § 6231(a)(3). The statutory definition has two prongs: (1) a requirement that the item is one which must be “taken into account for the partnership’s taxable year under any provision of subtitle A”; and (2) a requirement that the Secretary has concluded that the item is “more appropriately determined at the partnership level than at the partner level.” Id. We must look to IRS implementing regulations for guidance regarding both prongs of this inquiry. This is so for two reasons: one governed by our case law and the other governed by the statutory text.

First, we concluded in Keener that the reference in Section 6231(a)(3) to Subtitle A of the tax code is ambiguous, requiring that we give deference to any reasonable agency interpretation of that phrase. 551 F.3d at 1363 (citing Chevron U.S.A., Inc. v. Natural Res. Def. Council, Inc., 467 U.S. 837, 843, 104 S.Ct. 2778, 81 L.Ed.2d 694 (1984)). 3 Next, the statute expressly requires that a “partnership item” be designated by the Secretary via its regulatory authority as more appropriately determined at the partnership level than at the partner level. I.R.C. § 6231(a)(3).

Thus, because the statute expressly refers to the regulations with respect to the second prong, and our precedent requires Chevron deference to those regulations governing the first, we look to the regulations as the primary source for the definition of a “partnership item.”

Treasury Regulation § 301.6231(a)(3)-1(a) implements the statutory definition of “partnership item.” Subsection (a) of that provision defines a partnership item, in relevant part, as “[t]he partnership aggregate and each partner’s share of ... [i]tems of income, gain, loss, deduction, or credit of the partnership[.]” 26 C.F.R. § 301.6231(a)(3)-1(a), (a)(1), (a)(1)(i). 4 The U.S. Tax Court determines whether partner identity would affect the “partnership aggregate” or “each partner’s share of ... income, gain, loss, deductions, or credits of the partnership” by examining the particular facts of the case before it. Grigoraci v. Comm’r, 84 T.C.M. (CCH) 186 (2002), 2002 WL 1835711, at *6, 2002 Tax Ct. Memo LEXIS 207, at *17-18. If partner identity would affect the distributive shares reported to the partners, it is a partnership item for purposes of that case. Id. at *4-5, 2002 Tax Ct. Memo LEXIS 207 at *13. If, under the particular facts of the case, partner identity would affect the distributive shares of the partnership, the item must be determined at the partnership level. See, e.g., Blonien v. Comm’r, 118 T.C. 541 (2002), 2002 WL 1298749, at *7 n. 6, 2002 U.S. Tax Ct. LEXIS 33, at *20 n. 6.

The statutory requirement that a partnership file an information return establishes a close relationship between allocation and partner identity. See I.R.C. § 6031(a). That requirement mandates that a partnership’s information return identify “the individuals who would be entitled to share in the taxable income if distributed and the amount of the distributive share of each individual.” Id. Here, the parties dispute which partners are entitled to share in the taxable income and the distributive share of each party. The Court of Federal Claims cannot determine the parties who would be entitled to share in the taxable income and the distributive share of each party unless it properly identifies the true partners of the partnership.

One regional circuit that has considered this issue and the Tax Court have come to the same conclusion. In Katz v. Commissioner, the Tenth Circuit recognized the relationship between allocation and partner identity. 335 F.3d 1121, 1128-29 (10th Cir.2003). There, the taxpayer declared bankruptcy and attempted to allocate his share of losses incurred by partnerships in which he was a partner between his bankruptcy estate and himself. Id. at 1123. The IRS determined that the losses should have been allocated entirely to the bankruptcy estate and attempted to challenge the allocation in a partner-level deficiency proceeding. Id. at 1124-25. The Tenth Circuit rejected the partner-level challenge. The court observed that the partnership was required to specify a partner’s income and losses on its partnership return. Id. at 1128-29. That task could not be accomplished unless allocation were conducted at the partnership level, the court observed. “What is important is that the debtor was a partner during part of the partnership year, so the partnership returns must set forth the debtor’s share of income, loss, etc.” Id. at 1128.

The Tax Court, similarly, has found that partner identity can be necessary to properly allocate the partnership items. In Blonien, the Tax Court found that partner identity was more appropriately determined at the partnership level because it could affect allocation of partnership items among the partners in that action. 2002 WL 1298749, at *7 n. 6, 2002 U.S. Tax Ct. LEXIS 33, at *20 n. 6. Blonien was a partner-level proceeding. The government argued that the court lacked jurisdiction to revisit a determination made in a prior, partnership-level proceeding that the taxpayer, Blonien, was a partner in the partnership. Id. at *6, 2002 U.S. Tax Ct. LEXIS 33 at *18. The Tax Court agreed with the government because, if Blonien successfully argued that he was not a partner, “the share of [the partnership’s] COD income wrongly allocated to Mr. Blonien would have to be reallocated among the other partners.” Id. at *7 n. 6, 2002 U.S. Tax Ct. LEXIS 33 at *20 n. 6. The Tax Court recognized that the determination of partner identity could be a partner-level determination “where resolution of the issue would not affect the allocation of partnership items to the other partners,” but found that resolution of that issue in Blo-nien’s case would affect allocation. Id.

This case is similar to Blonien and Katz. Here, it is possible that the distributive shares reported to the partners of RRMC Group would change if one or more of the CRUTs were disregarded. During the 2001 tax year, four heirs, four CRUTs, and RRMC Management Corp. were listed as the partners of record on RRMC Group’s partnership return. If the Court of Federal Claims were to determine that some of the heirs’ transfers to the CRUTs were shams, each remaining partner’s share of the partnership items could change. If the court were to disregard one, two, or three of the CRUTs, for example, the distributive shares would be different than if the court were to disregard all of the CRUTs. In such a scenario, the court would be required to decide who should report the disregarded CRUT’s share. Thus, while the validity of the CRUTs would not impact the partnership’s aggregate income, it could affect the remaining parties’ individual shares of that income.

Even in cases in which the Tax Court determined that partner identity was not a partnership-level determination, it recognized that the inquiry turns on the facts of the particular case and the effect that partner identity would have on the distributive shares. In Grigoraci, the IRS asserted in its FPAAs that certain corporate partners were shams and sought to reallocate shares to certain individuals. 2002 WL 1835711, at *1, 2002 Tax Ct. Memo LEXIS 207, at *3-4. The taxpayers challenged the FPAAs in a partnership-level proceeding in the Tax Court. Id. at *1-2, 2002 Tax Ct. Memo LEXIS 207 at *1-4. The government moved to dismiss on the ground that the identity of the partners was a partner-level issue over which the Tax Court lacked jurisdiction in the partnership-level proceeding. Id. at *1, 2002 Tax Ct. Memo LEXIS 207 at *2-3. The Tax Court began its analysis by stating that “the hallmark of a partnership item is that it affects the distributive shares reported to the other partners.” Id. at *5, 2002 Tax Ct. Memo LEXIS 207 at *13. To that end, the Tax Court inquired whether identifying individuals rather than corporations as the true partners would affect the distributive shares. Id. at *5, 2002 Tax Ct. Memo LEXIS 207 at *15.

The Tax Court noted that no dispute existed concerning the aggregate income, gain, loss, deductions, and credits of the partnership. Id. at *5-6, 2002 Tax Ct. Memo LEXIS 207 at *16. It further noted that, even if the income were reallocated to the individual partners, “there [would be] no dispute about the amount of the allocations made to the partners.... ” Id. at *5-6, 2002 Tax Ct. Memo LEXIS 207 at *16-17. Because the allocations would have no effect on “either the partnership’s aggregate or each partner’s share of income, gain, loss, deductions, or credits of the partnership,” the court concluded that partnership identity was not a partnership item on the facts of that case. Id. at *6, 2002 Tax Ct. Memo LEXIS 207 at *18. While the Tax Court found that the partner identification question at issue before it was a partner-level determination, the court made clear that it made that determination based on the particular facts of that case: “Under the circumstances of this case, we hold that a determination that the partners of record were not the true and actual partners is not a ‘partnership item....’” Id. at *7, 2002 Tax Ct. Memo LEXIS 207 at *21 (emphasis added).

The facts of this case are distinct from those in Grigoraci. RRMC Group was required to report its gain or loss from the sale of the Constellation stock on its partnership return. RRMC Group also was required to assign that gain or loss to its partners. The IRS’s FPAAs challenged both the losses claimed by RRMC Group and its assignment of certain of those losses to the CRUTs. As discussed above, whether those losses were properly as signed to the CRUTs could affect the distributive shares reported to the remaining partners. This is an exercise that must be conducted at the partnership level.

' The Court of Federal Claims relied on Grigoraci and another Tax Court case, Hang v. Commissioner, 95 T.C. 74 (1990), to conclude that RRMC Group’s partners’ identities cannot be determined at the partnership level. Specifically, relying on Grigoraci, the Court of Federal Claims concluded that identification of the partners in this case would not affect the distributive shares. The court stated that, “[w]here ‘[t]here is ... no dispute about the amount of the allocations made to the partners])]’ ... as is the case before the court[,] ... ‘[a]n item with ‘no effect on either the partnership’s aggregate or each partner’s share of income, gain, loss, deductions, or credits of the partnership’ is not a partnership item.’ ” Alpha I, L.P. v. United States, 84 Fed.Cl. 209, 221 (2008) (quoting Grigoraci, 2002 WL 1835711, at *6, 2002 Tax Ct. Memo LEXIS 207, at *17-18, and Russian Recovery Fund Ltd. v. United States, 81 Fed.Cl. 793, 800 (2008)). While the trial court was correct to rely on the general rule set forth in Grigoraci, for the reasons noted above, its application of the facts at issue here to that rule was erroneous.

The Court of Federal Claims’s reliance on Hang also is misplaced. In Hang, the Tax Court held that the identity of shareholders in an S corporation was more appropriately determined at the shareholder level, but based that conclusion on somewhat different reasoning than it employed in Grigoraci. In Hang, the IRS attempted to reallocate an S corporation’s income from the two owners of record, who were minor children, to the children’s father, who was not an owner of record. 95 T.C. at 75. The statute and regulations defining “S corporation items” in Hang are analogous to the TEFRA provisions at issue here. See id. at 78. Like the implementing regulation defining a partnership item, the regulation at issue in Hang defined a “Subchapter S” item, in relevant part, as “[t]he S corporation aggregate and each shareholder’s share of, and any factor necessary to determine ... items of income, gain, loss, deduction, or credit of the corporation.” Id. at 79 (quoting 26 C.F.R. § 301.6245-1T).

Because the father was not a shareholder of record, the Tax Court found that the IRS’s proposed reallocation could not expressly fall within the scope of “the S corporation aggregate,” “each shareholder’s share of,” or “any factor necessary to determine” the “income, gain, loss, deduction, or credit of the corporation.” Id. at 80. While the government argued that the Tax Court should find that the regulation nevertheless encompassed the proposed reallocation because the father was allegedly the beneficial owner of the corporation’s stock and could therefore qualify as a shareholder for purposes of the regulation, the Tax Court declined to adopt that argument. Id. The court found that insufficient information existed at the corporate level to determine beneficial ownership: “[a]s a practical matter, there is no way for a corporation to determine who the beneficial owners of its stock are because the information necessary to make the determination would not be available at the corporate level where the beneficial owner of stock is not a shareholder of record.” Id.

Here, the Court of Federal Claims equated the relationship of the heirs to the CRUTs to that of the sons to the father in Hang. It concluded that “the [RRMC] Group partnership is not in a position to go behind the transactions between a partner and its successor of record. Both are questions of succession to interests that appear to the court to require determination at the individual taxpayer level.” That analysis was incorrect, however. In Hang, the scenario expressly fell outside the scope of the regulatory definition of an S corporation item because the father was not a shareholder of record. The Tax Court was required, consequently, to address the IRS’s alternative theory of beneficial ownership by inquiring whether all information was available at the partnership level to determine whether the father was the beneficial shareholder. Here, however, the scenario expressly falls within the regulation’s definition of a partnership item because all parties who could be identified as true partners — the heirs, the CRUTs, and RRMC Corp. — are listed as partners of record on the partnership’s 2001 information return, and the determination of the true partners in the partnership could affect the distributive shares attributed to one or more of those named partners. 5 The fact that the distributive shares could be affected by the determination of the partners’ identity is sufficient to sweep the issue within the definition of a partnership item, because the effect on the distributive shares is the “hallmark” of a partnership item. Grigoraci, 2002 WL 1835711, at *4-5, 2002 Tax Ct. Memo LEXIS 207, at *13.

Because partner identity in this case falls within the regulation’s definition of a partnership item, the Court of Federal Claims erred in adding an additional layer to the analysis by requiring that “all information” necessary to determine the identity of RRMC Group’s partners be available at the commencement of the partnership-level proceedings. The taxpayers defend the trial court’s analysis, arguing that a court cannot make findings in a partnership-level proceeding on the intentions of the individual Sands family members and the trustees of their charitable funds when they engaged in the CRUT transactions. The taxpayers are wrong. The Court of Federal Claims can allow discovery and hear testimony in a partnership-level proceeding as long as the inquiry regards a partnership-level item,' which we hold it does on these facts. Indeed, adjudicating this issue in a partnership-level proceeding is consistent with the congressional policy favoring that such issues be resolved in unified judicial proceedings. See Afshar, 64 Tax Law. 701; TEFRA, Pub.L. No. 97-248, § 402 et seq., 96 Stat. 648; and Keener, 551 F.3d at 1361.

The taxpayers next argue that the allocation among them will not change regardless of the identity of RRMC Group’s true partners. They point out that the IRS proposed disregarding the four CRUTs and reallocating the items reported by RRMC Group among the four heirs in the same proportion they were allocated among the CRUTs. Each heir, in other words, would merely stand in the shoes of its CRUT and be allocated the same share that its CRUT would have been allocated. The taxpayers claim that, because the partnership items would be allocated in the same manner among either four CRUTs or four heirs, the distributive shares would not change.

While the IRS did propose such an allocation, its proposal does not deprive the Court of Federal Claims of jurisdiction. The Court of Federal Claims exercises de novo review of an FPAA and is not bound to follow the IRS’s proposal. See Jade Trading, LLC v. United States, 80 Fed.Cl. 11, 43 (2007). The Court of Federal Claims dismissed the portion of the FPAA at issue before it considered the FPAA’s merits. At this stage, we cannot conclude with certainty that the trial court would accept the IRS’s proposal if this matter were to proceed to trial. 6

Tellingly, while the taxpayers argue that the allocation would not change under the IRS’s proposal, they remain silent as to whether they agree with the IRS’s proposal. They claim that “the government has not pointed to anything that suggests ... [RRMC] Group’s partnership items would be allocated to anyone other than the Sands family member who made the transfer to the CRUT,” but they do not reveal whether they would stipulate to such an allocation if the trial court were to consider the merits of it. Thus, the record leaves open a live issue concerning allocation, which the Court of Federal Claims must address on the merits. The IRS’s proposal, therefore, does not control the jurisdictional issue.

The taxpayers also caution that the conclusion we reach here is tantamount to placing a continuing burden on partnerships to ascertain the identity of their partners. The taxpayers point out that many partnerships have hundreds or even thousands of partners whose partnership interests may be actively traded, and that Congress never intended to burden partnerships with the responsibility of ascertaining their “true” partners.

The taxpayers are correct — at least in the abstract. It is true, for example, that, after partners exchange their partnership interests, a partnership is not required to note such an exchange on its return until it is notified of the exchange. I.R.C. § 6050K(c). The taxpayers also observe, and the government does not dispute, that the IRS does not require partnerships to do anything more than rely on their known partners of record to satisfy their obligation to file a return identifying the individuals who would be entitled to the partnership’s distributive shares under I.R.C. § 6031(a). Thus, the taxpayers correctly observe that partnerships are entitled to take their partners of record at face value.

The taxpayers are incorrect, however, in arguing that our holding, which is limited to the particular facts of this case, places a new and additional burden on all partnerships to ascertain the identity of their true partners. Congress created TEFRA proceedings such as this one to provide a vehicle for determining partner identity when partner identity is challenged by the IRS; sometimes that is appropriate at the partnership level and sometimes it is appropriate at the partner level, depending upon the circumstances. Partnerships do not bear the burden of identifying their true partners in all filings; rather, the IRS bears the burden of establishing that the partners are other than those identified as known partners in a partnership’s tax filings. Where, as here, that challenge, if successful, would directly impact the number of distributive shares attributable to other members of the partnership, resolution of that question at the partnership level is appropriate.

For the foregoing reasons, the identity of the partners in RRMC Group is appropriately determined at the partnership level. 7

Ill

With respect to the forty-percent gross valuation misstatement penalty, the Court of Federal Claims erred when it granted the taxpayers summary judgment on that penalty. The Court of Federal Claims was wrong to conclude that it was not obligated to determine whether the taxpayers’ underpayments are attributable to a valuation misstatement merely because the taxpayers conceded the gain and loss adjustments in the FPAAs. The cases on which the trial court relied for this conclusion misapply the valuation misstatement penalty, and, thus, do not properly characterize the law in this circuit. We review the Court of Federal Claims’s ruling de novo. Merino v. Comm’r, 196 F.3d 147, 155 n. 12 (3d Cir.1999) (“[T]he question of whether the valuation overstatement statute applies to a particular taxpayer’s situation is a question of law that we subject to plenary review.” (citing Gainer v. Comm’r, 893 F.2d 225, 226 (9th Cir.1990))). Upon doing so, we vacate that ruling and remand for further penalty proceedings.

The tax code provides that a penalty “shall be added” “to any portion of an underpayment of tax required to be shown on a return ... which is attributable to ... [a] ny substantial valuation misstatement.” I.R.C. § 6662(a) & (b)(3). A penalty may apply for both a substantial valuation misstatement and a gross valuation misstatement. The substantial valuation misstatement penalty applies when the value of any property or the adjusted basis of any property is 200% or more of the amount the IRS determines to be the correct amount; the gross valuation misstatement penalty applies when the claimed value or basis is 400% or more of the correct amount. I.R.C. § 6662(e)(1)(B)(i), (h)(2)(A)(ii)(I). Here, the IRS sought to impose a forty-percent gross-valuation misstatement penalty based on its determination that the partnerships inflated the cost basis of the Constellation stock by more than 900%, the cost basis of the Yahoo stock by more than 3,700%, and the cost basis of the Corning stock by more than 3,900%.

The statute requires that any underpayment of tax on which a valuation misstatement penalty is based be “attributable to” the valuation misstatement. I.R.C. § 6662(b)(3). The taxpayers conceded the IRS’s capital gain and loss adjustments in the FPAAs solely on the ground of I.R.C. § 465. They then argued that the underpayment of tax they conceded was not “attributable to” a valuation misstatement because I.R.C. § 465 does not address a valuation misstatement. That I.R.C. § 465 does not address a valuation misstatement, however, does not absolve the partnerships from liability for the penalty.

I.R.C. § 465 does not address the valuation of any claimed gain or loss. It provides that an individual’s deduction of certain losses must be limited to the amount considered “at risk.” I.R.C. § 465(a)(1). The provision does not disallow the existence of a loss; it limits the deduction of that loss. See id. Because the deduction of a loss is typically at issue only with respect to an individual partner’s income tax return, the Tax Court has held that any question concerning whether an taxpayer meets the “at-risk” requirement of I.R.C. § 465 must be addressed in a partner-level, rather than partnership-level, proceeding. Hambrose Leasing 1984-5 LP v. Comm’r, 99 T.C. 298, 310 (1992) (“[T]he application of section 465 is not a determination ‘required to be taken into account for the partnership’s taxable year.’ ” (quoting I.R.C. § 6231(a)(3))).

Thus, the partnerships’ alleged failure to report the alleged gains in this action is conceptually distinct from the issue presented in the context of I.R.C. § 465. First, I.R.C. § 465 addresses the deduction of losses. The IRS, however, did not take issue in the FPAAs with the amount of loss deduction that the partnerships claimed. Rather, the IRS took issue with the partnership’s failures to report their gains. I.R.C. § 465 does not apply to gains. See I.R.C. § 465(a)(1). Second, the “at-risk” provision of I.R.C. § 465, is generally a partner-level item. Hambrose Leasing, 99 T.C. at 310. The IRS, therefore, could not rely on I.R.C. § 465, a provision that limits the deduction of a loss in a partner-level proceeding, to penalize the partnerships for undervaluing gains in partnership-level proceeding. For the same reason, the partnerships could not rely on I.R.C. § 465 to concede the adjustments underlying the penalty that the IRS sought. 8

The Court of Federal Claims appeared to acknowledge that I.R.C. § 465 does not provide a valid basis for the partnerships’ concession. Rather than confront this reality, however, the trial court concluded it was not required to address at all whether the partnerships relied on a valid basis for the concession or to look behind that concession to determine the real cause of the tax underpayment. The trial court stated that it was not required to “endorse the validity of the ground on which plaintiffs made their concession,” as long as the “[pjlaintiffs did not concede the adjustments on grounds relating to valuation that could cause the penalties to be applied.” Alpha I, L.P. v. United States, 84 Fed.Cl. 622, 632 n. 6 (2008) (“Penalty Op.”). The trial court missed the point, however. The taxpayers could not revise the grounds upon which their underpayment of taxes actually is attributable by choosing to concede that underpayment on some other, invalid theory.

The Court of Federal Claims cited several cases to support its decision to defer to the terms of the partnerships’ concession without further scrutiny, two of which it found particularly persuasive: Todd v. Commissioner, 862 F.2d 540 (5th Cir.1988), and Gainer v. Commissioner, 893 F.2d 225 (9th Cir.1990). Not only do we find those cases factually distinguishable from this one, we disagree with the legal analysis employed in Todd and Gainer, finding it flawed in material respects.

In Todd, the Fifth Circuit affirmed the Tax Court’s ruling that penalties sought for valuation overstatements did not apply because the taxpayers’ underpayments were not attributable to the valuation overstatements. 862 F.2d at 543. There, the IRS disallowed certain depreciation deductions and credits because the property in question had not been placed in service during the tax years in issue. Id. The IRS separately imposed the penalty for valuation overstatement. Id. The tax liability after adjusting for the failure to place the property in service did not differ from the tax liability after adjusting for the valuation overstatements. Id. Any alleged valuation overstatement, moreover, was irrelevant, because the property that was the subject of the alleged overvaluation misstatement was never placed in service during the relevant tax year. Id. The Fifth Circuit, therefore, held that the tax underpayment could not be attributed to the valuation overstatement. Id.

Gainer involved the same fact pattern as Todd. 893 F.2d at 226. Like the Fifth Circuit, the Ninth Circuit declined to impose valuation overstatement penalties where the IRS separately disallowed depreciation deductions and credits because the property at issue was not placed in service during the relevant tax year. Id. at 228. Because a valid, independent basis for disallowing the deductions and credits existed, the court concluded that “[plaintiffs] overvaluation bee[ame] irrelevant to the determination of any tax due.” Id. 9

The Court of Federal Claims concluded that this case is analogous to those in the Todd and Gainer category because, in its view, the “adjustments were made on grounds unrelated to valuation....” Penalty Op. at 630. The trial court disregarded that a valid, independent basis for the adjustments existed in those cases. The only grounds on which the partnerships conceded the adjustments in this case was the inapplicable I.R.C. § 465 adjustment. The trial court, accordingly, analogized this case to cases in which a valid, independent basis existed for the adjustment, even though no such basis existed here. The analogy is inapt. Thus, assuming we were persuaded by the legal theory employed in Todd and Gainer, we would find that theory inapplicable to the facts before us.

Even if factually identical to this case, moreover, we are not persuaded that Todd, Gainer, and their progeny accurately apply the valuation misstatement penalty. Indeed, the flaws in the analysis employed in Todd and Gainer are so apparent that subsequent panels of the circuit courts deciding those eases have questioned their holdings.

In Todd, the Fifth Circuit looked to guidance from the “Blue Book,” a post-enactment summary of the legislation prepared by the staff of the Joint Committee on Taxation, to determine whether the tax underpayment at issue was attributable to a valuation misstatement. 862 F.2d at 542-43 (interpreting Staff of the Joint Committee on Taxation, General Explanation of the Economic Recovery Tax Act of 1981, at 333 (Comm. Print 1981) (“Blue Book”)). The Blue Book proposed applying the following analytical rule: “The portion of a tax underpayment that is attributable to a valuation overstatement will be determined after taking into account any other proper adjustments to tax liability-” Id. (quoting Blue Book at 333). The Blue Book then proposed a formula for applying that rule: The tax underpayment attributable to the valuation overstatement equals the difference between (i) “ ‘actual tax liability (i.e., the tax liability that results from a proper valuation and which takes into account any other proper adjustments)’ and (ii) “ ‘actual tax liability as reduced by taking into account the valuation overstatement.’ ” Id. (quoting Blue Book at 333). The Blue Book then provided an example of the application of that rule: If an improper $20,000 deduction “was claimed by the taxpayer as a result of a valuation misstatement,” and “another deduction of $20,000 is disallowed totally for reasons apart from the valuation overstatement,” then the overvaluation penalty should apply only to the former valuation-related deduction, not to the latter unrelated deduction. Id. at 543 (quoting Blue Book at 333 n. 2). Thus, if the taxpayer’s filing reflected a taxable income of $40,000, but the actual taxable income after adjusting for each improper deduction was $80,000, then the tax underpayment attributable to the valuation overstatement is $80,000(r) minus $60,000(r), where r is the tax rate. See id.

The Blue Book, in sum, offers the unremarkable proposition that, when the IRS disallows two different deductions, but only one disallowance is based on a valuation misstatement, the valuation misstatement penalty should apply only to the deduction taken on the valuation misstatement, not the other deduction, which is unrelated to valuation misstatement.

The court in Todd mistakenly applied that simple rule to a situation in which the same deduction is disallowed based on both valuation misstatement- and non-valuation-misstatement theories. There, the IRS disallowed the claimed deduction because the property was not placed in service during the relevant tax year and was overvalued. Id. at 543. Because placing the property in service was a prerequisite for taking the deduction, the court believed that any alleged overvaluation of the same property was irrelevant where the property never met the prerequisite in the first instance. Id.

The Blue Book does not describe or apply to the scenario presented in Todd, however. The Fifth Circuit recognized this flaw in a recent case in a concurring opinion joined by the entire panel. See Bemont Invs., LLC v. United States, 679 F.3d 339, 351-52 (5th Cir.2012) (Prado, J., concurring, joined by Reavley and Davis, JJ.) While the panel in Bemont Investments was obligated to affirm a district court decision denying a valuation misstatement penalty on the basis of Todd, it wrote separately to express its disagree ment with Todd’s reasoning. As the concurrence correctly observed:

The Blue Book only covers the case of two unrelated deductions, one of which is caused by overvaluation. Accordingly, the Blue Book does not suggest that the overvaluation penalty should not apply if overvaluation is one of two possible grounds for denying the same deduction and the ground explicitly chosen is not overvaluation.

Id. at 352. The court in Bemont Investments further observed that every circuit court to have addressed the issue, except the Ninth Circuit in Gainer, has rejected Todd’s reasoning. Id. at 353-55 (citing Fid. Int’l Currency Advisor A Fund, LLC v. United States, 661 F.3d 667, 673-74 (1st Cir.2011); Merino, 196 F.3d at 158; Zfass v. Comm’r, 118 F.3d 184, 191 (4th Cir.1997); Ill es v. Comm’r, 982 F.2d 163, 167 (6th Cir.1992); Gilman v. Comm’r, 933 F.2d 143, 151 (2d Cir.1991); Massengill v. Comm’r, 876 F.2d 616, 619-20 (8th Cir.1989)). Even the Ninth Circuit has recognized that Gainer might have been incorrectly decided. Keller v. Comm’r, 556 F.3d 1056, 1060-61 (9th Cir.2009) (holding that the Ninth Circuit is “constrained by” Gainer, which “rested in large part” on Todd, but recognizing the “sensible method” of “many other circuits”). We, too, part with the Todd and Gainer panels.

We agree that an underpayment of tax may be attributable to a valuation misstatement for purposes of the statute even when the IRS asserts both a valuation misstatement ground and a non-valuation-misstatement ground for the same adjustment. When considering whether such a “dual-cause” scenario falls within a statute’s reach, courts consider the context and policy underlying the statute. Fid. Int’l, 661 F.3d at 673 (citing W. Page Keeton et al., Prosser and Keeton on Torts §§ 41-42 (5th ed.1984)). There is little doubt that Congress intended to prevent abuse of the tax code when it enacted the valuation misstatement penalty. See, e.g., id. at 673 (“One might think that it would be perverse to allow the taxpayer to avoid a penalty otherwise applicable to his conduct on the ground that the taxpayer had also engaged in additional violations that would support disallowance of the claimed losses.”); Gilman, 933 F.2d at 152 (holding that a transaction disregarded for lack of economic substance — a non-valuation-related ground — nevertheless may be subject to a valuation misstatement penalty because “[a] transaction that lacks economic substance generally reflects an arrangement in which the basis of the property was misvalued in the context of the transaction” and that Congress “intended to penalize” such transactions); Merino, 196 F.3d at 158-59 (agreeing with Gilman); and Clearmeadow Invs. LLC v. United States, 87 Fed.Cl. 509, 534 (2009) (“[I]t is particularly dubious that Congress intended to confer ... largesse upon participants in tax shelters, whose intricate plans for tax avoidance often run afoul of the economic substance doctrine.”). An interpretation of the statute that allows imposition of a valuation misstatement penalty even when other grounds are asserted furthers the congressional policy of deterring abusive tax avoidance practices.

The Court of Federal Claims erred both by adopting the legal analysis applied in Todd and Gainer and by expanding that flawed analysis even beyond the facts presented there. On remand, the trial court must determine whether the taxpayers’ underpayments are attributable to a valuation misstatement. That determination is a fact-driven one, focusing on the role that any valuation misstatements played in attaining any improper tax benefits. The trial court, for example, may examine the features of the transactions at issue and consider whether any valuation misstate ments were “the vehicle for generating” inappropriate basis calculations or losses, Fid. Int’l, 661 F.3d at 673; whether a transaction “reflects” an improper valuation, Gilman, 933 F.2d at 152; and whether any improper valuation of the property in question “is an essential component of the tax avoidance scheme,” Merino, 196 F.3d at 158. This assessment must be made despite, and independent of, the “concession” of capital gain and loss adjustments that the partnerships offered. We leave it to the trial court to make its determination in the first instance.

IV

Finally, the taxpayers appeal the Court of Federal Claims’s grant of summary judgment to the government on the twenty-percent penalty for negligence, substantial understatement, and failure to act reasonably and in good faith. The taxpayers’ appeal is premature. If the Court of Federal Claims concludes on remand that the forty-percent gross valuation misstatement penalty applies, the taxpayers’ cross appeal of the twenty-percent penalty potentially will be moot. Even if the trial court were to find that both penalties apply, it would be permitted to impose only the highest of those because the gross valuation misstatement penalty and accuracy-related penalty may not be stacked. See 26 C.F.R. § 1.6662-2(c).

If the Court of Federal Claims were to impose the forty-percent penalty, and this court were to affirm that penalty, arguments regarding the propriety of the twenty-percent penalty would be moot. This court would not be required to address the twenty-percent penalty to affirm the trial court’s judgment in those circumstances. It is possible that this court will still be required to address the twenty-percent penalty — if, for example, this court were to reverse the trial court’s imposition of the forty-percent penalty, or the trial court were to decline to impose the forty-percent penalty in the first instance and the twenty-percent penalty were the only issue presented on appeal. Unless and until such an appeal is filed, however, we find it premature to address a penalty that may not be determinative in resolving this case.

V

We reverse the Court of Federal Claims’s holding that it lacked jurisdiction to determine the identity of RRMC Group’s partners. We also vacate its holding that the gross valuation penalty is inapplicable. Finally, we dismiss the taxpayers’ appeal of the negligence penalty as premature, without prejudice to reassertion of the arguments relating to that appeal if and when appropriate. This action is remanded to the trial court for further proceedings consistent with this opinion.

REVERSED IN PART, DISMISSED IN PART, and REMANDED

1 . A short sale is a sale of securities that are not owned by the seller. Securities are borrowed, usually from a brokerage house, and then sold on the open market. The seller holds the proceeds from the sale but is required to replace the borrowed property in kind — referred to as "closing out” the short sale — at some future date. See Zlotnick v. TIE Commc’ns, 836 F.2d 818, 820 (3d Cir.1988).

2 . The taxpayers dispute the trial court’s finding that they engaged in a tax shelter. Cross Appellants' Br. 59-61. Because they dispute that finding in the cross-appeal that we dismiss as premature, we express no opinion on the tax shelter finding. For that reason, we refer to the taxpayers’ activities as “transactions” rather than "tax shelters.”

3 . In Keener, the parties debated whether the reference to “Subtitle A” in Section 6231(a)(3) modifies "any item required to be taken into account” — which would restrict the meaning of "partnership item" to only those items appearing in Subtitle A — or modifies "the partnership’s taxable year” — -thereby encompassing anything that affects, or is “required to be taken into account for,” the partnership’s taxable year, including items outside Subtitle A. 551 F.3d at 1363. Because courts had been inconsistent in their construction of the phrase, we concluded that the phrase was ambiguous, requiring deference to agency regulations defining the term "partnership item,” when those regulations are reasonable. Id. While we did not undertake an independent grammatical analysis of the statutory text in Keener, the conclusion that we must give Chevron deference to regulations interpreting the interplay between the term "partnership item” and the reference to Subtitle A of the tax code in Section 6231(a)(3) is unmistakable and, thus, controlling.

4 . In Keener, the court did not specifically address whether Treasury Regulation § 301.6231(a)(3)-1(a) is entitled to Chevron deference. The court held only that Treasury Regulation § 301.6231(a)(3)-1(b) was reasonable, and, thus, was entitled to deference. The parties here do not dispute whether Treasury Regulation § 301.6231(a)(3)-1(a) is entitled to Chevron deference, however. All parties, in fact, rely on Keener and urge deference to governing regulations. For purposes of this appeal, therefore, we assume, but do not decide, that Treasury Regulation § 301.6231(a)(3)-1(a) is entitled to Chevron deference.

5 . The taxpayers argue that the Sands ceased being partners of record, and that the taxable year closed with respect to them as a matter of law, when they transferred their partnership interests to the CRUTs. See I.R.C. § 706(c)(2)(A) (providing that the taxable year of a partnership closes with respect to a partner whose entire interest in the partnership terminates). That argument is circular. The trial court has yet to determine whether the Sands’ partnership interests terminated because it has yet to determine whether their transfers to the CRUTs were valid.

6 . We disagree with the taxpayers that the government waived this argument. The Court of Federal Claims ruled on the issue of whether partner identity affects allocation when it stated that "the question of whether the partnership interests in [RRMC] Group were validly transferred does not affect the allocation of income, gain, or losses among other partners.” On appeal, the government is merely developing its argument on that same issue, namely, whether a live issue concerning allocation exists.

7 . While the government also attempts to argue that partner identity falls within the scope of Treasury Regulation § 301.6231(a)(3)-1(b), we find that argument unpersuasive. Treasury Regulation § 301.6231(a)(3) — 1(b) provides, in relevant part, that a " 'partnership item’ includes ... the legal and factual determinations that underlie the determination of the amount, timing, and characterization of items of income, credit, gain, loss, deduction, etc." 26 C.F.R. § 301.6231 (a)(3)-1(b). The government proffers arguments as to why partnership identity, in the abstract, underlies the determination of items such as the amount, timing, and characterization of items of income, credit, gain, loss, deduction. Appellant's Br. 43-45. The government, however, fails to explain why partner identity would be necessary to determine those items on the facts of this particular case. As explained above, whether a particular item qualifies as a partnership item turns on the specific facts of the case. The government's abstract argument, therefore, is insufficient.

8 . The IRS did cite I.R.C. § 465 in the FPAAs and assert that the transactions did not increase the partners’ amounts at risk. While the Court of Federal Claims placed substantive significance on the citation to I.R.C. § 465, remarking that "the § 465 adjustment was listed right along with the defendant’s other theories for adjusting plaintiffs’ basis and gain," the IRS’s decision to cite I.R.C. § 465 does not convert the section into one that applies to loss deductions in the face of its plain language to the contrary.

9 . The other cases the trial court cited to support its conclusion also involved fact patterns where an alternative basis for the IRS’s adjustments existed. See Derby v. Comm’r, 95 T.C.M. (CCH) 1177, 2008 WL 540271, *25, 2008 Tax Ct. Memo LEXIS 46, at *90-91 (Tax Ct. Feb. 28, 2008) (“[Bjecause there is a separate, independent ground for disallowing [the] deductions, the overvaluation penalty may not be imposed against the petitioners.”); McCrary v. Comm’r, 92 T.C. 827, 851-55 (1989) (holding that valuation overstatement penalties did not apply where the taxpayers “conceded that they were not entitled to [an] investment tax credit because the agreement was a license and not a lease,” which were grounds unrelated to valuation); Rogers v. Comm’r, 60 T.C.M. (CCH) 1386, 1990 WL 195716, at *---, 1990 Tax Ct. Memo LEXIS 695, at *47-49 (Tax Ct. Dec. 10, 1990) (holding that overvaluation penalty did not apply where the taxpayers conceded other grounds of adjustment in the notice of deficiency, including that they lacked a profit objective, that the property at issue was not qualifying property, and that the property was not placed in service); and Weiner v. United States, 389 F.3d 152, 153 & 162-63 (5th Cir.2004) (declining to impose a valuation penalty where the taxpayers conceded an FPAA listing several independent grounds for the adjustments, including the sham and economic substance doctrines).