The Black & Decker Corporation v. United States

Affirmed in part, reversed in part, and remanded by published opinion. Judge MICHAEL wrote the opinion, in which Judge LUTTIG and Judge WILLIAMS joined. OPINION MICHAEL, Circuit Judge. A corporate taxpayer paid $561 million to a controlled subsidiary in exchange for 10,000 shares of the subsidiary’s stock and the subsidiary’s assumption of a $560 million contingent liability of the taxpayer. The taxpayer then sold the shares for $1 million, claimed a $560 million capital loss on its federal income tax return, and sought a refund based on that loss. The Internal Revenue Service declined to pay because it concluded that the capital loss stemmed from an illegal tax shelter. After the taxpayer sued, the district court denied the IRS’s summary judgment motion and granted the taxpayer’s summary judgment motion. The IRS appeals. We conclude that neither the IRS nor the taxpayer is entitled to summary judgment under the controlling tax statutes. Under the sham transaction doctrine, however, the validity of the claimed loss turns on unresolved issues of material fact. Accordingly, we affirm the denial of the IRS’s motion, reverse the grant of the taxpayer’s motion, and remand for further proceedings. I. A. The Black & Decker Corporation (BDC), its wholly-owned subsidiary Black & Decker Inc. (BDI), and their direct and indirect subsidiaries constitute a major manufacturer of power tools and home im *433 provement products. (The subsidiaries involved are Emhart Industries, Inc., Price Pfíster, Inc., and Kwikset Corp., all of which are domestic corporations controlled by BDC. BDC is also owner of Black & Decker Canada, Inc. (the “Canadian subsidiary”), which does not file a tax return in the United States. We will refer to BDC and its domestic direct and indirect subsidiaries as, collectively, “Taxpayer.”). Taxpayer provides medical and dental insurance benefits to its current and retired employees, who number in the thousands. The aggregated future health benefits claims constitute a contingent liability because their precise cost is not known in the present. Thus, Taxpayer can estimate but cannot predict with certainty how many of its employees or retirees will be diagnosed with particular illnesses in future years. In 1998 Taxpayer realized nearly $303 million in capital gains income from the sale of three businesses. Taxpayer sought to offset that income against a large loss to prevent the imposition of a substantial federal income tax obligation. To this end Taxpayer executed a transaction that gave rise to what it intended to be a significant capital loss. The Deloitte & Touche accounting firm had designed the transaction and advised some 30 corporate clients, including Taxpayer, on its implementation as a tax strategy. The transaction involved a subsidiary called Black & Decker Healthcare Management Inc. (BDHMI). Taxpayer owned all of BDHMI’s common stock. BDHMI’s preferred shareholders included an affiliate of William M. Mercer, Inc. (Taxpayer’s benefits consultant and a subsidiary of Marsh & McLennan Companies, Inc.) and Taxpayer’s Canadian subsidiary. The participation of outside investors permitted BDHMI to file a federal income tax return separate from Taxpayer’s. The transaction consisted of two phases. Phase One was an exchange on November 25, 1998, between Taxpayer and its Canadian subsidiary on the one hand and BDHMI on the other. Taxpayer and the Canadian subsidiary paid BDHMI approximately $561 million in cash with funds Taxpayer had borrowed from its banks for 30 days. In return BDHMI (1) gave Taxpayer and the Canadian subsidiary 10,000 shares of BDHMI’s series C preferred stock and (2) assumed liability for the future health benefits claims against Taxpayer and the Canadian subsidiary from 1999 to 2007, which had an estimated net present value of $560 million. According to the exchange agreement the companies executed, BDHMI’s assumption of liability “[did] not constitute either a legal defea-sance of the Benefits Liabilities by [Taxpayer] or a novation and consequently, [Taxpayer] ... continue[d] to be primarily liable for the payment and performance of the Benefits Liabilities.” J.A. 217. Thus, Taxpayer remained liable on the underlying obligations transferred to BDHMI. The companies executed Phase Two on December 29, 1998. Taxpayer and the Canadian subsidiary sold the 10,000 BDHMI shares at a price of $1 million to an unrelated third-party trust benefitting a former BDC executive. Also that day, BDHMI promised to lend BDI approximately $564 million, most of which was to be repaid in monthly installments according to the terms of three lending agreements. BDI’s installment payments on the loans were “designed to provide [BDHMI] with sufficient funds to pay” the benefits liabilities as they came due. J.A. 127. Although BDHMI continued to hold the benefits liabilities, Taxpayer reported all of the income from the businesses and employees that gave rise to those liabilities. *434 As one of BDHMI’s outside investors put it, “The rationale behind the establishment of the subsidiary [BDHMI] is that a loss equal to the reserve for the liabilities can be recognized upfront for tax purposes and the [special purpose vehicle, BDHMI] may be able to deduct the amount of the claims a second time as they are actually incurred.” J.A. 3235. Formally, the investors in BDHMI stood to earn a positive return on their investment in the event that Taxpayer’s actual health care liabilities fell short of the expected cost of those liabilities, so that Taxpayer’s repayment on the loan would exceed BDHMI’s payments on the medical claims, creating net income for BDHMI. But as a practical matter Taxpayer expected BDHMI to generate net operating losses because the health expenses were likely to consistently exceed BDHMI’s interest income from the loan to Taxpayer. Or, in the words of Taxpayer’s in-house accountants, “BDHMI will generate net operating losses since the interest income on the note receivable is not likely to exceed annual claims paid which are recognized for tax purposes when paid.” J.A. 148. B. In its 1998 tax return Taxpayer characterized Phase One as the purchase of the BDHMI shares for $561 million and Phase Two as the sale of those shares for $1 million, generating a $560 million capital loss. Taxpayer claimed that its basis in the BDHMI stock was equal to the cash payment without reduction by the benefits liabilities BDHMI assumed in Phase One. The large capital loss offset the capital gains from Taxpayer’s divestitures earlier in the year. Further, the large loss had both retrospective and prospective tax-reducing effects under the Internal Revenue Code of 1986, as amended, 26 U.S.C. (“IRC”), allowing Taxpayer to file for refunds on its returns for the 1995 through 2000 tax years. The refunds sought totaled approximately $57 million. Because the IRS did not pay the refunds for more than six months, Taxpayer commenced a civil action in the U.S. District Court for the District of Maryland. 26 U.S.C. §§ 7422; 6532(a)(1). The IRS filed a counterclaim for tax, interest, and penalties of approximately $215 million. The IRS construed the tax laws as requiring Taxpayer to state its basis in the shares sold as $1 million, not $560 million. Taxpayer took the opposite view. Moving for summary judgment, the IRS argued that under IRC § 357(c)(3)(A) Taxpayer, the transferor, had to reduce its stock basis by the amount of the contingent liability assumed by BDHMI, the transferee. The district court denied the motion. Taxpayer then moved for summary judgment, arguing that the transfer of the benefits liabilities was not a sham transaction that could be disregarded for tax purposes. The district court granted the motion, Black & Decker Corp. v. United States, 340 F.Supp.2d 621 (D.Md.2004). The district court stated that because BDHMI was constituted for a valid business purpose, BDHMI and all of its transactions were “objectively reasonable” and thus could not be disregarded for tax purposes. Id. at 623-24. This appeal by the IRS followed. II. The IRS first contends that it was entitled to summary judgment because, on the undisputed facts, the statutes governing the transaction required the basis reduction. Preliminarily, we note that if Taxpayer were to engage in the contingent liability transfer today, it would be required to reduce its basis by the amount of the transferred liabilities under IRC § 358(h), which Congress enacted as part *435 of the Community Renewal Tax Relief Act of 2000. P.L. 106-554, § 1(a)(7), 114 Stat. 2763 (2000). That statute does not control here because it does not apply retroactively. In addition, in 2001 the IRS issued Notice 2001-17 on contingent liability tax shelters, which put corporations on notice that the IRS would challenge transactions of the type at issue here. These legal and enforcement policy developments took place after Taxpayer engaged in the contingent liability transaction. Consequently, the question before us is whether the law at the time Taxpayer executed this transaction required Taxpayer to reduce its basis in the stock by the amount of the benefits liabilities transferred. In answering this question, we first briefly survey the pertinent statutes. Then we turn to the parties’ specific statutory arguments on whether summary judgment should have been granted in the IRS’s favor. A. On appeal the parties agree that the transaction is to be analyzed under IRC § 351. Section 351(a) provides that no gain or loss shall be recognized — that is, the transaction is tax-free — if property (here, the cash from Taxpayer) is transferred to a corporation (BDHMI) by “one or more persons” (here, members of a consolidated group, Taxpayer) solely in exchange for voting stock in BDHMI, and immediately after the exchange the trans-feror is in control of the transferee. See 26 U.S.C. § 351(a). Taxpayer did not, however, transfer property to BDHMI “solely” in exchange for BDHMI stock because BDHMI also assumed the contingent liability. So the transfer does not fit within the exact terms of § 351(a). Taxpayer would nevertheless continue to ’ enjoy the tax-free benefit of § 351(a) if Taxpayer could successfully invoke § 357(a). Under § 357(a) transactions are treated as § 351(a) tax-free transactions if they would satisfy § 351(a) were it not for the fact that the transferee, in consideration for the transferor’s property, not only gave its stock but also assumed the transferor’s liability. In such cases, § 357(a) prevents the transferee’s assumption of liability from being treated as taxable money or property received by the transferor. This tax-free benefit is not available, however, when the exceptions of §§ 357(b) and (c) apply. See 26 U.S.C. § 357(a). (Only the § 357(b) exception is in dispute here. Infra part III.A.) Separate from the concept of gain or loss recognition is the concept of basis computation. The income tax consequences of selling property hinge on the taxpayer’s basis in that property. Except as otherwise provided, “the basis of property shall be the cost of such property.” 26 U.S.C. § 1012. (Usually, basis is “the original cost of property used in computing capital gains or losses for income tax purposes.” Lessinger v. Comm’r, 872 F.2d 519, 525 n. 3 (2d Cir.1989) (quoting Webster’s Third New International Dictionary 182 (1963)). In general, when a taxpayer sells an asset for more than its basis, he records a capital gain; when he sells the asset for less than its basis, he records a capital loss.) In the § 351 transaction at issue here, Taxpayer’s basis in the BDHMI stock is determined under § 358. See 26 U.S.C. § 351(h)(2). Under § 358(a)(1) a § 351 transferor’s basis in the stock received from the transferee is the same as the basis of the property the transferor surrendered, reduced by the amount of any “money received.” 26 U.S.C. § 358(a)(1). Section 358(d)(1) further provides that, in § 358 analysis, an assumption of liability by the transferee shall be treated as “money received” by the transferor. A transferor reduces its basis in the stock received from the trans *436 feree by the amount of any liability the transferee assumed in exchange. But the statute also provides that § 358(d)(1) does not apply if the liability assumed is one that would be excluded under § 357(c)(3). 26 U.S.C. § 358(d)(2). Section 357(c)(3), in turn, excludes “liability the payment of which ... would give rise to a deduction.” 26 U.S.C. § 357(c)(3)(A). B. To prevail on summary judgment the IRS must demonstrate that as a matter of law Taxpayer was not entitled to the § 357(c)(3) exception. This demonstration turns on statutory meaning. “It is well established that when the statute’s language is plain, the sole function of the courts — at least where the disposition required by the text is not absurd — is to enforce it according to its terms.” Lamie v. United States Trustee, 540 U.S. 526, 534, 124 S.Ct. 1023, 157 L.Ed.2d 1024 (2004) (punctuation omitted); see Coleman v. Cmty. Trust Bank (In re Coleman), 426 F.3d 719, 724-25 (4th Cir.2005). When the language of the statute is unclear, we “may consult its legislative history as a guide to congressional intent.” Yi v. Fed. Bureau of Prisons, 412 F.3d 526, 533 (4th Cir.2005). The fact that a tax statute is at issue here does not render inapplicable these general principles of statutory interpretation. The benefits liabilities Taxpayer transferred to BDHMI fall within the plain terms of the § 357(c)(3) exception if, in the event that the claims are paid, a person would be able to deduct the amounts paid from that person’s taxable income. The statute does not specify, however, whether the person taking the deduction would be the liability’s transferee (BDHMI) or the transferor (Taxpayer). Because the statute does not clearly identify the person who would take the hypothetical deduction the statute envisions, it is appropriate to examine the legislative history so far as is necessary to identify that person. Section 357(c) was first enacted in 1978 and rewritten into its present form two years later by the next Congress. The Senate Report that presaged the 1980 revision explained: In general, liabilities the payment of which would give rise to a deduction include trade accounts payable and other liabilities (e.g., interest and taxes) which relate to the transferred trade or business. However, such liabilities may be excluded under this provision only to the extent payment thereof by the trans-feror would have given rise to a deduction. S.Rep. No. 96-498, at 62 (1979), reprinted in 1980 U.S.C.C.A.N. 316, 372. The second sentence quoted resolves the uncertainty in § 357(c)(3)’s “would give rise to a deduction” phrase by clarifying that Congress had in mind the deductibility by the transferor. A liability falls within the § 357(c)(3) exception so long as, if the transfer had not taken place and the trans-feror had paid the liability, the transferor could have taken a deduction. To see why this condition is satisfied here, let us suppose that the health care claims had remained with Taxpayer. In that case, when Taxpayer incurred medical expenses as its employees and retirees received health care, Taxpayer could have deducted those expenses from income. The IRS presses two arguments for why Taxpayer cannot claim the § 357(c)(3) exception. The first argument relies on legislative history. The IRS focuses on sentences such as the first of the two from the Senate Report quoted. It contends that Congress crafted the exception to protect a parent corporation from a tax double whammy when transferring both assets and associated liabilities to a subsidiary in *437 exchange for stock. From this perspective, Congress wanted to prevent such parent corporations from being twice penalized by (1) deprivation of the right to deduct the transferred liabilities as they accrued and (2) mandatory reduction of the stock basis by the amount of the liabilities transferred. Since Taxpayer only transferred the health claims but not the assets generating those claims, the IRS argues that Congress did not intend for Taxpayer to benefit from § 357(c)(3). Further, the IRS reads the quoted phrase “would have given rise to a deduction,” S.Rep. No. 96-498, at 62, to mean a deduction unavailable to the transferor once the liability has been transferred. The legislative history argument does not persuade us. The prototypical transaction Congress had in mind in drafting § 357(c)(3) may well have been one in which a corporation exchanged liabilities as part of a transfer of an entire trade or business to a controlled subsidiary, but nothing in the section’s plain language embraces such a limitation. As a result we find no ambiguity in the statute that requires us to parse the congressional record and discern what type of business transactions Congress originally envisioned in enacting the section. The Senate Report’s use of the phrase “would have given rise” also does not go as far as the IRS would have us take it. On the contrary, we agree with one commentator’s observation that this language “does not imply ... that Congress silently contemplated a case in which liabilities are transferred but the deduction is retained by the transferor and [then] concluded that § 357(c)(3) should not apply.” Ethan Yale, Reexamining Black & Decker’s Contingent Liability Tax Shelter, 108 Tax Notes 223, 234 (July 11, 2005). The IRS’s second argument is based on sound administration of the tax laws, because the Taxpayer should not be allowed to take the “functional equivalent of a double deduction.” Appellant’s Br. at 59. Although Taxpayer has not claimed the employee health expenses as a deduction (BDHMI, not a party to this suit, claims them instead), the IRS argues that Taxpayer has the legal right to seek these deductions as health care costs accrue. In the IRS’s view, the $560 million loss that Taxpayer reported effectively accelerates deductions for uncertain future health care costs through the year 2007. Such acceleration would contravene the prohibition against claiming a deduction in a given tax year for an estimate of liabilities that have not become fixed by the end of that year. Here, receipt of medical care and filing of proper claims forms would fix the annual health care liability. United States v. Gen. Dynamics Corp., 481 U.S. 239, 242-45, 107 S.Ct. 1732, 95 L.Ed.2d 226 (1987). Again, we are not convinced that the language of § 357(c)(3) is so unclear as to permit us to rely on this policy argument and adopt the IRS’s reading. In addition, because BDHMI files a tax return separate from Taxpayer’s and has been taking the deductions for the health care expenses as the expenses are incurred, the “double deduction” argument would only work if we were to treat BDHMI and Taxpayer as a single entity. We see no justification on the present record for disregarding the distinct corporate taxpayer identities of BDHMI and Taxpayer. Rather, we agree with Taxpayer: “BDHMI pays the claims; BDHMI takes the deductions — not Taxpayer.” Appel-lee’s Br. at 27. We conclude that the contingent liability Taxpayer transferred to BDHMI falls within the § 357(c)(3) exception for “liability the payment of which ... would give rise to a deduction.” Therefore, under § 358(d)(2)’s exception to the general rule *438 of § 358(d)(1), the liability need not be treated as “money received” by Taxpayer for basis reduction purposes. For this reason the district court’s denial of the IRS’s summary judgment motion was correct. III. The IRS next advances two arguments for why the district court erred in granting summary judgment in Taxpayer’s favor. The first argument is based on the IRC and the second on the judge-made sham transaction doctrine. A. Section 357(b) of the IRC is an anti-abuse rule. It provides that if a taxpayer’s principal purpose in transferring liability was to avoid federal income tax on the § 351 exchange, or was otherwise not a bona fide business purpose, then the assumption of liability shall, for § 351 analysis, be considered “money received” by the taxpayer on the exchange. 26 U.S.C. § 357(b)(1). In this refund suit the burden is on the taxpayer to prove that the assumption of liability “is not to be treated as money received by the taxpayer.” 26 U.S.C. § 357(b)(2). Thus, if § 357(b)(1) applied, Taxpayer would be required to prove “by the clear preponderance of the evidence” that it had a bona fide business purpose for transferring the benefits liabilities to BDHMI. Id. The IRS argues that Taxpayer’s business purpose was a disputed fact requiring trial. Next the IRS contends that this issue is material because if Taxpayer could not meet this burden at trial, then Taxpayer would be obligated to treat the benefits liabilities assumed by BDHMI as money Taxpayer received for § 351 purposes. The purposes of § 351, in the IRS’s view, include the basis computation rules of § 358. In particular, the IRS endeavors to connect the anti-abuse rule of § 357(b) to the basis-computation rule of § 358. The IRS would have us travel along one of two routes between these sections. The first route is linguistic. The IRS concentrates on the phrase “money received,” which appears in both § 357(b)(1) and in § 358(a)(1). From the IRS’s perspective, if Taxpayer lacked a bona fide business purpose in the exchange, then BDHMI’s assumption of liability would be “money received” by Taxpayer under § 357(b)(1) and also under § 358(a)(l)(A)(ii), so that Taxpayer’s basis in the BDHMI shares would be “decreased” by the amount of the liability BDHMI assumed. See § 358(a)(1) (“The basis of the property permitted to be received ... without the recognition of gain or loss shall be the same as that of the property exchanged decreased by ... the amount of any money received by the taxpayer”). This argument is plausible but ultimately unavailing. It cannot overcome the absence of any reference to § 358 on the face of § 357(b). Rather, § 357(b)(1) refers (as pertinent here) only to the “purposes” of § 351, which are the purposes of gain or loss recognition, as indicated by the opening words of § 351(a) (“No gain or loss shall be recognized ... ”). Those purposes are distinct from the purposes of § 358, which concerns the computation of basis. Section 351(h) certainly refers to § 358, but this reference has no substantive content; rather, it simply guides the reader of § 351 to the basis computation rule for exchanges that qualify for § 351’s tax-free treatment. The Treasury regulation interpreting § 357 also undercuts the IRS’s argument. The regulation explains that when a taxpayer lacks a bona fide business purpose in transferring liabilities through a § 351 *439 exchange, “the total amount of liabilities assumed ... shall, for the purpose of determining the amount of gain to be recognized upon the exchange in which the liabilities are assumed or acquired, be treated as money received by the taxpayer upon the exchange.” Treas. Reg. § 1.357-l(e) (emphasis added). The emphasized language clarifies that the focus of § 357(b) analysis is gain recognition only. The regulation’s complete silence on basis assessment indicates that § 357(b) is not concerned with such assessment. By declining to endorse the IRS’s view that “money received” means the same thing in the two different sections, we adhere to the teaching that “[a] word is not a crystal, transparent and unchanged, it is the skin of a living thought and may vary greatly in color and content according to the circumstances and the time in which it is used.” Towne v. Eisner, 245 U.S. 418, 425, 38 S.Ct. 158, 62 L.Ed. 372 (1918) (Holmes, J.). Only by masking the conceptual difference between gain or loss recognition and basis computation could we treat the phrase “money received” as having a single meaning, irrespective of “the circumstances ... in which it is used.” The IRS has not provided compelling reasons to do that. The second route the IRS identifies between § 357(b) and § 358 is structural. Here, the IRS argues that if Taxpayer could not meet § 357(b)’s business purpose test, Taxpayer could not use § 358(d)(2) to invoke § 357(c)(3)’s exclusion of deduction-generating liabilities. See supra Part II.B. That is because in the IRS’s view, to be deductible under § 357(c)(3) a liability must fall within § 357(c)(1), but § 357(c)(1) does not apply to any liability that falls within § 357(b)(1). See 26 U.S.C. § 357(c)(2) (§ 357(c)(1) “shall not apply to any exchange to which subsection (b)(1) of this section applies.”). The IRS’s argument falters in taking this last step. We read the statute to mean that as a general matter liability need not exceed basis — thereby falling within § 357(c)(1) — in order to qualify for the § 358(d)(2) exception. Again, § 358(d)(2) provides that the ordinary basis reduction rule of § 358(d)(1), under which assumed liabilities are treated as money received by the transferor, does “not apply to the amount of any liability excluded under section 357(c)(3).” 26 U.S.C. § 358(d)(2). Section 357(c)(3)’s exclusion, as relevant here, covers “a liability the payment of which ... would give rise to a deduction.” Supra part II.B. In referring to § 357(c)(3), § 358(d)(2) does not by implication refer to § 357(c)(1) or (2). If Congress intended such a broader reference, it could have built one into § 358(d)(2) by, for example, providing that § 358(d)(1) does “not apply to the amount of any liability excluded under § 357(c).” But that is not what the statute says. The central difficulty with the IRS’s reasoning is that it would limit liabilities eligible for the § 358(d)(2) exclusion to those that exceed basis under § 357(c)(1), even in transactions that have a perfectly legitimate business purpose and thus do not transgress the § 357(b) anti-abuse provision. Under § 357(c)(1) “if the amount of the liability transferred [in a § 351 exchange] is greater than the basis of the property transferred, then, in general, the transferor-taxpayer recognizes a capital gain on the amount by which the liability exceeds the basis of the transferred property.” Estate of Kanter v. Comm’r, 337 F.3d 833, 865 (7th Cir.2003), rev’d on other grounds sub nom Ballard v. Comm’r, 544 U.S. 40, 125 S.Ct. 1270, 161 L.Ed.2d 227 (2005). While § 357(c)(1) is concerned *440 with assumed liabilities that exceed the basis in the property transferred, § 358(d)(2) has no such concern; its only concern is the liabilities that “would give rise to a deduction” as described in § 357(c)(3). We agree with the IRS that under § 357(c)(2), when a transaction lacks a bona fide business purpose under § 357(b)(1), § 357(c)(1) is not applied. But accepting this much of the argument does not compel the conclusion that, when § 357(b)(1) applies, it also knocks out the § 357(c)(3) exception for basis computation purposes under § 358(d)(2). Or, as the IRS put it in a concededly non-prece-dential advisory memorandum to its field agents, even if a taxpayer has “abused the rules” in violation of § 357(b), § 358, which “does not inquire into taxpayer intent, is not the means of correcting that.” Field Service Advice 199905008 (Feb. 5, 1999), at J.A. 644. Taxpayer points to a published revenue ruling that supports this analysis. The ruling views § 358(d)(2) as providing “that section 358(d)(1) does not apply to the amount of any liabilities defined in section 357(c)(3).” Rev. Rui. 80-198, 1980-2 C.B. 113 (emphasis added). There is no language in the ruling that ties § 358(d)(2) either to § 357(c)(l)’s command regarding liabilities that exceed basis or to § 357(c)(2)’s displacement of § 357(c)(1) altogether when ' § 357(b)(1) applies. Of course, the revenue ruling does not control our analysis for at least two reasons. First, while such rulings constitute “precedents to be used in the disposition of other cases,” Aeroquip-Vickers, Inc. v. Comm’r, 347 F.3d 173, 181 (6th Cir.2003) (quoting Rev. Proc. 89-14, 1989-1 C.B. 815), the precise degree of deference owed to these rulings — particularly in the face of the IRS’s decision to interpret the statute differently in litigation — is unclear. See id. at 181 (concluding that “some deference” was owed to a revenue ruling). Cf. Dominion Res., Inc. v. United States, 219 F.3d 359, 366 (4th Cir.2000) (“[Revenue rulings are entitled to considerably less deference than an agency’s properly promulgated regulations.”). Second, the hypothetical § 351 exchange analyzed in the ruling is a transfer of assets and liabilities undertaken for a bona fide business purpose, which is factually distinguishable from the unusual transfer of liabilities alone in Taxpayer’s case. With these caveats in mind we nevertheless take the revenue ruling to be fully consistent with our reading of the statute’s plain language. That plain language is our most important guide to resolve the competing claims. It is of no importance that the district court never evaluated the IRS’s argument under § 357(b). For the reasons identified we do not find the argument convincing. Accordingly, we cannot reverse the summary judgment in Taxpayer’s favor on this basis. B. The sham transaction doctrine permits the IRS to disregard a transaction that literally complies with the terms of the IRC but that is devoid of any legitimate business purpose. In Frank Lyon Co. v. United States, 435 U.S. 561, 98 S.Ct. 1291, 55 L.Ed.2d 550 (1978), the Supreme Court described when, under this doctrine, a transaction should be. recognized as valid for tax purposes: [W]here ... there is a genuine multiple-party transaction with economic substance which is compelled or encouraged by business or regulatory realities, is imbued with tax-independent considerations, and is not shaped solely by tax-avoidance features that have meaningless labels attached, the Government should honor the allocation of rights and duties effectuated by the parties. *441 Id. at 583-84, 98 S.Ct. 1291. Our court distilled from this guidance a two-part test: “To treat a transaction as a sham, the court must find [ (1)] that the taxpayer was motivated by no business purposes other than obtaining tax benefits in entering the transaction, and [ (2)] that the transaction has no economic substance because no reasonable possibility of a profit exists.” Rice’s Toyota World, Inc. v. Comm’r, 752 F.2d 89, 91 (4th Cir.1985). “Whether under this test a particular transaction is a sham is an issue of fact.” Id. at 92. We applied this test in several subsequent cases. Hunt v. Comm’r, 938 F.2d 466, 472 (4th Cir.1991); Hines v. United States, 912 F.2d 736, 739 (4th Cir.1990); Friedman v. Comm’r, 869 F.2d 785, 792 (4th Cir.1989). As we clarified in Hines, the first prong of the test is subjective, while the second is objective. Nevertheless, “[wjhile it is important to examine both the subjective motivations of the taxpayer and the objective reasonableness of the investment, in both instances our inquiry is directed to the same question: whether the transaction contained economic substance aside from the tax consequences.” Hines, 912 F.2d at 739. In replying to the IRS’s arguments opposing Taxpayer’s motion for summary judgment, Taxpayer conceded for purposes of deciding the motion that “tax avoidance was the sole motivation underlying Black & Decker’s decision to outsource its healthcare management function to BDHMI.” J.A. 3363. Taxpayer thus effectively conceded that the test’s subjective prong was satisfied. To defeat Taxpayer’s summary judgment motion, all that remained was for the IRS to show that there were genuine issues of material fact concerning the test’s objective prong. The district court’s approach to the objective prong strayed from our precedents. Although the district court quoted the pertinent language from Rice’s Toyota, see 340 F.Supp.2d at 623, it went on to assert: “A corporation and its transactions are objectively reasonable, despite any tax-avoidance motive, so long as the corporation engages in bona fide economically-based business transactions.” Id. at 623-24. In so reasoning, the district court mischaracterized the Rice’s Toyota test, which focuses not on the general business activities of a corporation, but on the specific transaction whose tax consequences are in dispute. “The second prong of the sham inquiry, the economic substance inquiry, requires an objective determination of whether a reasonable possibility of profit from the transaction existed apart from tax benefits.” Rice’s Toyota, 752 F.2d at 94 (emphasis added). Thus, many of the undisputed facts upon which the district court relied in concluding that Taxpayer was entitled to summary judgment — including the facts that BDHMI “maintained salaried employees” and paid health claims as they came due with BDHMI assets, 340 F.Supp.2d at 624 — were simply not germane to the proper inquiry under the second prong of our circuit’s sham transaction test. We do not agree with Taxpayer’s contention that the Supreme Court’s decision in Moline Properties v. Commissioner, 319 U.S. 436, 63 S.Ct. 1132, 87 L.Ed. 1499 (1943), supports the district court’s analysis of the objective prong under Rice’s Toyota. The Court in Moline held that a corporate taxpayer, the petitioner, “had a tax identity distinct from its stock-holder,” an individual, such that gain realized on sales in two tax years were to be treated as income taxable to the corporation, not to the individual. 319 U.S. at 440, 63 S.Ct. 1132. To reach this conclusion, the Court examined the purposes for the individual’s establishment of the corporation. Id. at 439-40, 63 S.Ct. 1132. The Court recog *442 nized that in some tax eases “the corporate form may be disregarded where it is a sham or unreal.” Id. at 439, 63 S.Ct. 1132. Moline is not implicated, however, by the IRS’s allegation that under the objective prong of the sham transaction test there was no reasonable profit opportunity in the two-phase transaction Taxpayer executed with BDHMI in November and December 1998. The IRS is not arguing at this point that BDHMI’s corporate identity separate from Taxpayer must be disregarded for tax purposes, such that income earned by one is to be attributed to the other. Shams under Rice’s Toyota are distinct from shams under Moline. In particular, a shareholder’s transaction with a controlled corporation may be a sham under Rice’s Toyota even if the corporation is entitled to regard its income as distinct from its shareholder’s because the corporation is not itself a sham under Moline. Hines illustrates the proper analysis under the objective prong of the Rice’s Toyota. Hines involved an IRS challenge to investment interest and depreciation deductions stemming from a taxpayer’s purchase and lease back to the seller of used computer equipment. We first noted that the payments on the transaction would leave the taxpayer “with a loss of $127,324 over the eight years of the lease” to the seller. 912 F.2d at 739. We next identified all of the possible sources of revenue on the transaction and weighed them against this loss. See id. at 739-40. Viewing the evidence in the light most favorable to the taxpayer, which had won at trial, we nevertheless concluded that the transaction “fail[ed] to yield any reasonable expectation of a profit.” Id. at 739. Hines clarifies that under this circuit’s firmly established Rice’s Toyota standard, the objective prong of the sham transaction test focuses on reasonable expected profits from a transaction. There is no basis here for abandoning our standard by scrutinizing the transaction for its “real economic effects,” despite Taxpayer’s argument that we should do so based on the law of another circuit. See United Parcel Serv. of Am., Inc. v. Comm’r, 254 F.3d 1014, 1019 (11th Cir.2001). In evaluating Taxpayer’s motion for summary judgment, the evidence must be viewed in the light most favorable to the IRS, the non-moving party. Consistent with Hines, the essential question posed in Taxpayer’s motion is whether the IRS adduced sufficient facts to go to trial on its argument that Taxpayer lacked “any reasonable expectation of a profit” from the transaction that generated the claimed $560 million capital loss reported on the 1998 return. We conclude that the IRS offered ample evidence to permit a reasonable trier of fact to find in the IRS’s favor. Four expert witnesses retained by the IRS explained in their expert reports why the only economically substantial value to Taxpayer in transferring its contingent liability to BDHMI was in tax savings. These experts were (1) University of Kansas economist Mark Hirschey, (2) Harvard economist Oliver D. Hart, (3) Ohio State University health economist John A. Rizzo, and (4) benefits consultant John G. Kontner. Taxpayer in turn pointed to a countervailing expert of its own, Harvard business professor Michael C. Jensen. The district court did not cite any of this evidence, let alone evaluate it. Weighing all of this expert testimony should have been left for trial because witness credibility cannot be assessed on summary judgment. The IRS’s evidence, in sum, was sufficient to create a triable issue on the reasonable profit expectations attaching to Taxpayer’s transaction. The trial to resolve whether Taxpayer’s transaction was a sham must determine whether both prongs of the Rice’s Toyota *443 test are satisfied, since the Taxpayer’s concession that the subjective prong was met applied only for purposes of deciding the summary judgment motion. Regarding the subjective prong, it bears repeating that a taxpayer’s “mere assertion” of subjective belief in the profit opportunity from a transaction “particularly in the face of strong objective evidence that the taxpayer would incur a loss, cannot by itself establish that the transaction was not a sham.” Hines, 912 F.2d at 740. The “ultimate determination of whether an activity is engaged in for profit is to be made ... by reference to objective standards.” Id. For these reasons, proper analysis of our well-established test under Rice’s Toyota hinged on genuine issues of material fact that remained very much in dispute when the district court granted summary judgment in Taxpayer’s favor. Accordingly, we reverse and remand for a trial to resolve the sham transaction question. IV. We conclude that IRC §§ 357 and 358 do not require Taxpayer to reduce its basis in the BDHMI stock by the amount of the liabilities transferred to BDHMI; we therefore affirm the district court’s denial of the IRS’s motion for summary judgment. Because the district court erred in granting summary judgment in favor of Taxpayer under the sham transaction doctrine, we reverse that judgment and remand for further proceedings. AFFIRMED IN PART, REVERSED IN PART, AND REMANDED