Wells Fargo & Company v. United States

957 F.3d 840

• Court: Court of Appeals for the Eighth Circuit
• Decided: April 24, 2020
• Docket: 17-3578
• Status: Published

Opinion

United States Court of Appeals For the Eighth Circuit ___________________________

No. 17-3578 ___________________________

Wells Fargo & Company, on behalf of itself and the members of its affiliated group filing a consolidated return

lllllllllllllllllllllPlaintiff - Appellant

v.

United States of America

lllllllllllllllllllllDefendant - Appellee ____________

Appeal from United States District Court for the District of Minnesota - Minneapolis ____________

Submitted: November 12, 2019 Filed: April 24, 2020 ____________

Before SHEPHERD, GRASZ, and KOBES, Circuit Judges. ____________

SHEPHERD, Circuit Judge.

Wells Fargo & Company (Wells Fargo) appeals from the district court’s1 determination that it was not entitled to a tax credit on its 2003 tax return arising from

1 The Honorable Patrick J. Schiltz, United States District Judge for the District of Minnesota. a transaction that Wells Fargo entered into with a British bank. It also appeals from the court’s determination that Wells Fargo was liable for a “negligence penalty” after claiming that credit. Having jurisdiction under 28 U.S.C. § 1291, we affirm the judgment in its entirety.

I.

In 2002, Wells Fargo, a United States corporation, entered into a structured trust advantaged repackaged securities transaction (STARS) with Barclays Bank PLC (Barclays), a corporate citizen of the United Kingdom. Wells Fargo asserts that the purpose of STARS was to borrow a significant amount of money from Barclays at a very low interest rate, to diversify its funding sources, to reduce its liquidity risk, and to provide a stable source of funding for five years. The government, however, argues that STARS was an elaborate and unlawful tax avoidance scheme, designed to exploit the differences between the tax laws of the U.S. and the U.K. and generate U.S. tax credits for a foreign tax that Wells Fargo did not, in substance, pay.

Wells Fargo claimed foreign-tax credits on its 2003 federal tax return arising from STARS. The Internal Revenue Service (IRS) disallowed those credits and notified Wells Fargo that it owed additional taxes. Wells Fargo paid the resulting tax deficiency and filed this lawsuit in order to challenge the IRS’s decision and to obtain a refund. The government defended the IRS’s position, and it sought to impose a “negligence penalty” on Wells Fargo as an offset defense because Wells Fargo underpaid its 2003 taxes after claiming this credit. Following a jury trial, the government prevailed in part below, and Wells Fargo appeals.

A.

Before discussing STARS and the facts giving rise to this case, it is useful to briefly analyze the particular tax credit at issue. The U.S. government taxes the income

-2- of its citizens, including corporations, even when that income is earned abroad or is otherwise subject to taxation by another country. See 26 U.S.C. § 61(a) (defining gross income as “all income from whatever source derived”). To avoid the problem of double taxation on income that is taxed by a foreign jurisdiction, the Internal Revenue Code permits a taxpayer to claim a dollar-for-dollar tax credit against its federal tax liability for taxes paid to another country. See 26 U.S.C. § 901; see also Burnet v. Chicago Portrait Co., 285 U.S. 1, 7 (1932) (“[T]he primary design of [the foreign-tax credit] was to mitigate the evil of double taxation.”). This credit, which is known as the “foreign-tax” credit, is subject to various rules and limitations, see 26 U.S.C. §§ 901-909, including that the underlying transaction giving rise to the foreign- tax credit is a “valid transaction,” not a “sham transaction.” This means that the transaction must have some economic substance outside of its tax consequences. Bank of New York Mellon Corp. v. Comm’r (BNY), 801 F.3d 104, 108 (2d Cir. 2015).

To illustrate how the foreign-tax credit works, suppose a taxpayer earned $100 abroad and was subject to $22 of U.K. tax and $35 of U.S. tax. Without foreign-tax credits, the taxpayer would have an overall tax liability of $57 and would be double taxed on that income—once by the U.K. and once by the U.S. However, with the foreign-tax credit, the taxpayer could claim credits of $22 on its federal tax return, reducing its U.S. tax liability to $13 and its overall tax liability to $35. Now the taxpayer would effectively be taxed once on that income. The foreign-tax credit is, in short, supposed to create an economic “wash” to the taxpayer: every $1 it pays in foreign taxes offsets $1 of U.S. tax liability. Practically, this means that a taxpayer has no financial incentive to engage in a transaction simply to generate foreign-tax credits.

-3- B.

Turning to the facts of this case, we note that STARS is a sophisticated financial transaction with a fairly complex structure. See Wells Fargo & Co. v. United States (Wells Fargo I), 143 F. Supp. 3d 827, 831 (D. Minn. 2015) (“The STARS transaction was extraordinarily complicated—so complicated, in fact, that it almost defies comprehension by anyone (including a federal judge) who is not an expert in structured finance.”); Santander Holdings USA, Inc. v. United States, 977 F. Supp. 2d 46, 48 (D. Mass. 2013) (noting that STARS was “surpassingly complex and unintuitive; the sort of thing that would have emerged if Rube Goldberg had been a tax accountant”). By now, STARS has been thoroughly examined and explained by several circuit courts. See, e.g., Santander Holdings USA, Inc. v. United States, 844 F.3d 15 (1st Cir. 2016); BNY, 801 F.3d at 104; Salem Fin., Inc. v. United States, 786 F.3d 932 (Fed. Cir. 2015).

In this iteration of STARS, Wells Fargo placed approximately $6.7 billion of income-producing assets into a Delaware trust that had, as a trustee, another Wells Fargo entity. The trustee was a U.K. resident for tax purposes, which subjected the income generated by the trust to U.K. taxes, which the trust paid. Barclays then loaned Wells Fargo $1.25 billion, for a term of five years, by purchasing an interest in the trust. Wells Fargo was obligated to repay Barclays by repurchasing Barclays’s interest in the trust at the end of the five-year period. By virtue of its ownership of part of the trust, Barclays was also subject to taxation in the U.K. on the income produced and distributed to Barclays by the trust. As a result of certain features of U.K. tax law, Barclays obtained certain U.K. tax benefits from its ownership interest—these tax benefits were central to STARS.

Each month, Wells Fargo paid Barclays interest on the loan, while Barclays paid Wells Fargo a fixed cash payment called “Bx.” The Bx payments totaled

-4- approximately $32 million per year for each of the five years that STARS was operational. Wells Fargo and Barclays negotiated Bx to be equal to approximately 47.5% of the U.K. tax benefits that Barclays was expected to receive by participating in STARS, regardless of whether Barclays was actually allowed such benefits by the U.K. tax agency, Her Majesty’s Revenue and Customs (HMRC). Either party could terminate STARS before the end of the five-year period by giving the other party 30- days’ notice.

Stated differently—and more specifically—STARS comprises both a loan component and a trust component. In the loan component, as discussed above, Barclays lent to Wells Fargo $1.25 billion by purchasing an ownership interest in the trust. Wells Fargo paid monthly interest payments to Barclays. At the end of five years, Wells Fargo would repay the principal of the loan by repurchasing Barclays’s ownership interest in the trust. In the trust component, also discussed above, Wells Fargo transferred income-earning assets into the trust, which generated a certain amount of U.K. taxes. The trust set aside an amount to pay those taxes. Simultaneously, the trust also distributed income to Barclays, which had an ownership interest in the trust by virtue of the loan component, by placing those distributions into a “blocked account” at Wells Fargo. Barclays could not actually access the funds held in this blocked account, which were quickly returned to the trust before being distributed to Wells Fargo. By nominally holding the funds, even though it could not access them, and by reinvesting the funds into the trust, Barclays could claim a U.K. tax loss and a deduction on its U.K. taxes. Barclays also received a U.K. tax credit for the U.K. taxes already paid by Wells Fargo on behalf of the trust, and it received a separate U.K. tax deduction for making Bx payments to Wells Fargo. Through its Bx payments, Barclays gave to Wells Fargo an amount equal to 47.5% of the U.K. tax benefits that it generated from participating in STARS.

-5- To understand how STARS generated the relevant tax benefits to both parties, consider the following hypothetical. Santander, 844 F.3d at 20-21 (“The benefits for both parties can be illustrated by a hypothetical also employed by the Second and Federal Circuits.”). Suppose Wells Fargo places its income-producing assets into a trust, which generate $22 of U.K. taxes for every $100 of income produced by the trust. Wells Fargo pays the $22 in taxes to HMRC and receives a credit of $22 on its U.S. taxes, pursuant to the foreign-tax credit. At this point, the tax effect to Wells Fargo is an economic wash—it paid the taxes that it owed to HMRC, but it simultaneously reduced its tax liability to the IRS by the same amount. However, at the same time, Barclays obtains $18.70 in tax benefits in the U.K. as a result of its “re- investment” of its share of the trust income through the blocked account and through other U.K. tax consequences of participating in STARS. Barclays, via the Bx payment, subsequently returns $11 of that $18.70 to Wells Fargo. The net result is that Wells Fargo gains $11, Barclays gains $7.70, and HMRC gains $3.30.2 But it appears that the IRS is down exactly $22—the sum total of the gains to Wells Fargo, Barclays, and HMRC—which Wells Fargo deducted from its U.S. tax return. Wells Fargo, however, presented the transaction as tax-neutral to the IRS—it asserted that it paid $22 in U.K. taxes and merely offset its U.S. taxes by $22 because of the foreign-tax credit.3

2 The reason that HMRC gains $3.30 is because Wells Fargo paid $22 in U.K. taxes to HMRC while Barclays receives deductions or credits from HMRC equal to $18.70. The difference of $3.30 between what Wells Fargo pays and what Barclays receives is the reason why STARS is tax additive for the U.K. 3 Wells Fargo was not the only U.S. financial institution to enter into a STARS transaction that has been the subject of some tax controversy—the form of STARS in several recent circuit cases are materially similar to the one in this case. See, e.g., Santander, 844 F.3d at 19-20; BNY, 801 F.3d at 104; Salem, 786 F.3d at 937-39. In 2005, British authorities alerted the IRS to the fact that STARS was possibly an abusive tax shelter, despite the fact that the scheme was tax additive to HMRC. See

-6- C.

In 2004, after Wells Fargo set up STARS with Barclays, Wells Fargo filed its 2003 income-tax return. It claimed foreign-tax credits for approximately $70 million in U.K. taxes that it paid which arose out of STARS. The IRS disallowed the credits and asserted that Wells Fargo owed additional taxes for 2003, presumably because Wells Fargo underpaid its taxes after claiming the foreign-tax credits. Wells Fargo paid the deficiency and initiated this lawsuit in order to obtain a refund. The central issues in the district court—and now on appeal—are whether STARS was a sham transaction under the “sham-transaction” doctrine, also known as the “economic- substance” doctrine, and whether Wells Fargo is liable for a negligence penalty as a setoff. As discussed below in greater detail, the sham-transaction doctrine is a common law doctrine that prevents a taxpayer from receiving tax benefits if the transaction or occurrence giving rise to those tax benefits is not a “valid” economic transaction, but a “sham” transaction. See WFC Holdings Corp. v. United States, 728 F.3d 736, 742 (8th Cir. 2013). In the district court, the government took the position that STARS was a sham transaction and argued Wells Fargo was ineligible to claim foreign-tax credits for its payment of U.K. taxes arising from this transaction. The government also asserted a negligence penalty, under 26 U.S.C. § 6662, as a potential setoff defense for Wells Fargo’s initial underpayment of tax.

Santander, 844 F.3d at 25. The IRS later proposed and finalized regulations that disregard STARS transactions for tax purposes, but they do not apply retroactively. Id. at 17; see also Determining the Amount of Taxes Paid for Purposes of Section 901, 72 Fed. Reg. 15081, 15081 (March 30, 2007). “The regulations reflect an understanding that STARS transactions and similar complex financial structures for which foreign tax credits are sought both pose a danger to the federal fisc and do not serve the purposes intended by Congress in enacting the foreign tax credit regime.” Santander, 844 F.3d at 17.

-7- After the district court appointed a special master to address various pretrial matters, Wells Fargo moved for partial summary judgment. In its motion, Wells Fargo asked the special master to decide, among other things, whether Barclays’s payment of Bx to Wells Fargo constituted pre-tax income—that is, whether Bx was economic income to Wells Fargo that is unrelated to tax benefits for the purposes of the economic-substance doctrine. It also asked the special master to find that Wells Fargo did not owe a negligence penalty for the alleged underpayment of U.S. taxes. The special master submitted a report and recommendations, finding in favor of Wells Fargo that Bx constituted pre-tax income and that it was not liable for a negligence penalty. Both parties filed objections, and the district court denied partial summary judgment on both issues, effectively reserving a ruling until after trial. See Wells Fargo I, 143 F. Supp. 3d at 842, 853.

The case was then tried to a jury. The government argued at trial that STARS was actually comprised of two separate transactions—a trust component and a loan component—and each had to be analyzed separately under the economic-substance doctrine. The jury found that STARS was really two separate transactions, that STARS’s trust component lacked a reasonable possibility of pre-tax profit and a non- tax business purpose, that STARS’s loan component had a reasonable possibility of pre-tax profit but that it lacked a valid business purpose, and that Bx was a tax benefit. See Wells Fargo & Co. v. United States (Wells Fargo II), 260 F. Supp. 3d 1140, 1142- 43 (D. Minn. 2017). The issue of whether to characterize the foreign taxes paid by Wells Fargo as a pre-tax expense or a post-tax expense was not submitted to the jury—the district court seemingly agreed with Wells Fargo that it was, as a matter of law, a post-tax expense.

Following the jury’s verdict, the district court determined that STARS’s trust component was a sham transaction; that the loan component was not a sham

-8- transaction; that if the nature of Bx was a legal question for the court, it would find as a matter of law that Bx was a tax benefit; and that Wells Fargo was liable for the negligence penalty. Id. at 1142-43 & n.1. Accordingly, the district court entered a judgment requiring the government to refund to Wells Fargo a net amount of approximately $13.65 million. This figure reflected that STARS’s trust component, but not the loan component, was a sham transaction, and that Wells Fargo was subject to the negligence penalty as an offset. The judgment primarily restored to Wells Fargo an interest deduction that it claimed for STARS’s loan component. Wells Fargo appeals, arguing that the district court erred as a matter of law in finding that STARS’s trust component was a sham transaction and that it was subject to the negligence penalty.

II.

We first consider whether the district court erred in determining that the trust component of STARS should be disregarded for tax purposes under the sham- transaction doctrine. “The characterization of a transaction for tax purposes is a question of law that is subject to de novo review, while the underlying facts are reviewable for clear error.” Salem, 786 F.3d at 940 (citing Frank Lyon Co. v. United States, 435 U.S. 561, 581 n.16 (1978)). In conducting our review, we are mindful that the relevant factual findings were made by a jury following a trial and that the jury’s verdict is “entitled to extreme deference.” Craig Outdoor Adver., Inc. v. Viacom Outdoor, Inc., 528 F.3d 1001, 1009 (8th Cir. 2008).

The sham-transaction doctrine, also known as the economic-substance doctrine, allows the IRS and courts “to distinguish between structuring a real transaction in a particular way to obtain a tax benefit, which is legitimate, and creating a transaction to generate a tax benefit, which is illegitimate.” Salem, 786 F.3d at 942 (emphasis

-9- omitted) (quoting Stobie Creek Invs. LLC v. United States, 608 F.3d 1366, 1375 (Fed. Cir. 2010)). Even if a “transaction complies with the literal terms of the relevant statutes and regulations that create the tax benefits,” Wells Fargo I, 143 F. Supp. 3d at 834, courts must “disregard a transaction that a taxpayer enters into without a valid business purpose in order to claim tax benefits not contemplated by a reasonable application of the language and the purpose of the [Internal Revenue] Code or the regulations thereunder . . . .” WFC Holdings, 728 F.3d at 742. Put simply, if a transaction is a “sham”—meaning that it lacks economic substance outside of its tax considerations—a taxpayer is not entitled to claim credits for any foreign taxes that it paid which arose from that transaction. Therefore, if STARS’s trust component lacks economic substance, it is a sham transaction as a matter of law and we must affirm the district court.

“In determining whether a particular transaction is a ‘sham,’ [this Court] has traditionally applied the two-part test set forth in Rice’s Toyota World, Inc. v. Commissioner, 752 F.2d 89, 91-92 (4th Cir. 1985).” Wells Fargo I, 143 F. Supp. 3d at 834. First, we consider whether the transaction lacks economic substance because no real potential for profits exists apart from its tax benefits, which is known as the objective test. WFC Holdings, 728 F.3d at 743. Second, we consider whether the taxpayer was motivated to enter into the transaction by any economic purpose outside of tax considerations—this is known as the subjective test. Id.

In applying the sham-transaction doctrine, we have noted that the “transaction[s] must be viewed as a whole, and each step, from the commencement of negotiations to the consummation of the sale, is relevant.” IES Indus., Inc. v. United States, 253 F.3d 350, 356 (8th Cir. 2001) (alteration in original) (quoting Comm’r v. Court Holding Co., 324 U.S. 331, 334 (1945)). Moreover, it is the taxpayer, not the government, who

-10- bears the burden of proving that the transaction is not a sham. Wells Fargo I, 143 F. Supp. 3d at 834.4

We hold that STARS’s trust component had no real potential for profit outside of its tax implications and that Wells Fargo had no valid purpose for entering into it outside of tax considerations. For this reason, we decline to address “whether a transaction that has [either economic substance or a valid business purpose] but not the other is a sham.” Id.; see also WFC Holdings, 728 F.3d at 744. Accordingly, we affirm the district court’s conclusion that STARS’s trust component was a sham transaction that must be disregarded for tax purposes.

As discussed above, the first part of the sham-transaction doctrine is to inquire whether the transaction at issue had “a reasonable possibility of profit . . . apart from tax benefits.” Shriver v. Comm’r, 899 F.2d 724, 726 (8th Cir. 1990) (quoting Rice’s Toyota World, 752 F.2d at 94). The parties suggest that this analysis centers on two

4 Importantly, this Court has never decided—and has previously declined to decide—whether courts must rely on both parts of the sham-transaction doctrine in characterizing a particular transaction. WFC Holdings, 728 F.3d at 743-44. Some circuits have adopted a disjunctive approach, in which a transaction is valid—and not a sham—if the transaction has either economic substance or a valid business purpose. Id. at 744 n.3. Others have employed a conjunctive approach, in which a transaction is valid only if it has economic substance apart from tax considerations and the taxpayer has a valid business purpose for entering into it. Id. Some circuits have been more rigid, looking only to the objective, economic-substance prong without concern for the taxpayer’s subjective motivation. Id. Others still have adopted a more flexible approach, relying on the objective and subjective prongs as factors to consider in determining how to characterize a particular transaction. Id.

-11- distinct questions. The first is whether Bx, which comprised Wells Fargo’s income from the trust, constituted “pre-tax income” or a “post-tax benefit.” See Salem, 786 F.3d at 940 (“The characterization of the Bx payment is important to the resolution of this [issue].”). If Bx is “pre-tax income,” then Wells Fargo earned significant income from the trust component, meaning that it will have an easier time showing that it had a reasonable possibility of profit apart from the transaction’s tax considerations. On the other hand, if Bx is simply a “post-tax benefit,” then Wells Fargo did not earn much, if at all, from STARS’s trust component that can be counted towards its profits. The second question is whether Wells Fargo’s payment of U.K. taxes was a “pre-tax expense” or a “post-tax expense.” If it was a pre-tax expense, then Wells Fargo’s expenses of operating the trust greatly increase, meaning that it is more difficult for it to show a reasonable possibility of profit from the trust. If, on the other hand, the payment of U.K. taxes are a “post-tax expense,” then they do not count in determining whether the transaction had a reasonable possibility of profit apart from tax benefits.

At the outset, we note that the answer to the first question—how to characterize the Bx payments—does not ultimately matter. See Santander, 844 F.3d at 23 (“We see no need to address the government’s characterization of the Bx payment as a [tax] rebate, not income, because we hold that whether the Bx payment is best characterized as a rebate or as income, [the taxpayer’s] argument still fails.”). Even assuming, without deciding, that Bx is pre-tax income to Wells Fargo, and not a post-tax benefit, STARS’s trust component is still profitless “because the ‘profit’ to [Wells Fargo] from the Bx payment comes at the expense of exposure to double the Bx payment’s value in U.K. taxes.” Id.; see also Stobie, 608 F.3d at 1378 (noting that to determine whether

-12- a transaction has potential for profit, its expected non-tax income must be compared with the expected operating “costs and fees”).5

In arriving at this conclusion, we adopt the reasoning of the First Circuit, which is fairly consistent with that of the Second and Federal Circuits. Santander, 844 F.3d at 15; BNY, 801 F.3d at 104; Salem, 786 F.3d at 932. As explained in Santander, even if Wells Fargo receives Bx payments of $11 for every $100 of income generated by the trust, there is no possibility of profit outside of tax considerations because each $11 Bx payment is earned at the expense of Wells Fargo’s payment of $22 in U.K. taxes. 844 F.3d at 23-24; see also Salem, 786 F.3d at 951. In other words, every $1 that Wells Fargo makes via the Bx payment comes at a cost of $2 in U.K. taxes, to which Wells Fargo intentionally subjected itself. Santander, 844 F.3d at 23. “When the primary transaction cost of the Bx payment, the U.K. taxes, are factored into the pre-tax profitability calculation, the Trust transaction is plainly profitless.” Id. at 23- 24. The only reason the scheme ends up making money for Wells Fargo is because Wells Fargo simultaneously obtains U.S. foreign-tax credits for $22, which correspond to its payment of that $22 in U.K. taxes, in addition to the Bx payment, which works out to $11, or half of its U.K. tax liability. See id. at 23. In sum, because Wells Fargo’s pre-tax expenses dwarf any income it receives from the trust in the form of Bx, STARS’s trust component simply has no reasonable possibility of profit outside of its tax features. See BNY, 801 F.3d at 122 (noting that regardless of how Bx is characterized, “the benefit of the [Bx payments] was more than offset by the additional transaction costs that [the taxpayer] incurred to obtain [Bx]”).

5 We note that the district court determined that Wells Fargo’s payment of U.K. taxes was a post-tax expense, and it instructed the jury to disregard Wells Fargo’s payment of U.K. taxes when determining whether the trust had a reasonable possibility of pre-tax profit.

-13- Contrary to Wells Fargo’s arguments, we think it is proper to count Wells Fargo’s U.K. tax expenses as a pre-tax expense, not as a post-tax expense, in this calculation. See Salem, 786 F.3d at 948-49. “Because exposure to U.K. taxation was the necessary and sufficient condition of the Bx payment, the U.K. taxes were an expense incurred by [Wells Fargo] for the ‘profit’ generated by the Trust transaction.” Santander, 844 F.3d at 24 n.11. Stated differently, if Bx is assumed to be pre-tax economic income to Wells Fargo, the payment of the U.K. taxes similarly must be treated as an expense incurred in earning that income because the two are directly and inextricably linked to one another. See id. Wells Fargo only received the Bx payment because it subjected its own income-generating assets in the trust to U.K. taxation, thus generating the relevant tax credits for itself and for Barclays in order to make the transaction worthwhile for both parties. For the reasons discussed above, “when the U.K. taxes are recognized as expenses, there is no pre-tax profit, and the Trust transaction lacks a cardinal feature of an economically substantial transaction: a reasonable prospect of pre-tax profit.” Id. “[A]s a result, it is not a transaction for which Congress intended to give the benefit of the foreign tax credit.” Id. at 23.

Wells Fargo argues that our decision in IES, which was followed by the Fifth Circuit in Compaq Computer Corp. v. Comm’r, 277 F.3d 778 (5th Cir. 2001), requires us to treat its U.K. tax payments as a post-tax expense and not as a pre-tax expense. Admittedly, there is a certain logic to characterizing any payment of tax as a post-tax expense—pre-tax expenses are ordinarily, and perhaps even definitionally, those that are incurred in a transaction apart from tax liability. Nevertheless, we do not read IES or Compaq to require us to treat the U.K. tax payments as post-tax expenses. Indeed, the transactions in “[t]hose cases did not analyze STARS transactions and so are distinguishable factually.” Santander, 844 F.3d at 24 n.11; see also BNY, 801 F.3d at 116 (noting that Compaq and IES arose out of “factually different contexts”). We emphasized in IES that our decision was based on the particular “facts and

-14- circumstances of [that] case,” including that the transaction had no “straw entities” or artificial devices, but that it involved actual economic risk and real dividends paid by a foreign corporation which was undertaking actual business activity in a foreign jurisdiction. IES, 253 F.3d at 355-56. Unlike this case, in which Wells Fargo’s incurrence and payment of foreign taxes on U.S.-source income was necessary for the generation of Barclays’s tax benefits and Bx payments, the taxpayers in IES and Compaq received dividend income paid in the ordinary course of business, where a foreign tax was routinely imposed on that income. That tax was a necessary consequence of obtaining the dividend income, which was incurred without a taxpayer intentionally subjecting U.S.-source income to foreign taxation. Under those facts, we did not treat the foreign taxes paid by the taxpayer as a pre-tax expense, but our decision did not create a per se rule that foreign taxes must always be treated as a post- tax expense. Here, we think it is proper to treat the U.K. tax expenses as a pre-tax expense because Wells Fargo artificially generated this tax by engaging in an economically meaningless activity which was specifically designed to create foreign- tax liability. This foreign-tax liability, in turn, would be recovered by Wells Fargo by obtaining U.S. foreign-tax credits. At the same time, Barclays recovered part of that foreign-tax liability and shared it with Wells Fargo via the Bx payments. Accordingly, we hold that STARS’s trust component lacked economic substance because there was no reasonable possibility of profit apart from the transaction’s tax consequences.

It is worth noting that the foregoing analysis does not mean that any transaction which does not have an immediate possibility of profit, apart from its tax considerations, is a sham. As the First and Federal Circuits have noted, “some transactions that are not immediately profitable without tax benefits, such as investments in ‘nascent technologies,’ may have economic substance.” Santander, 844 F.3d at 25. However, a STARS trust “transaction is not comparable to such transactions because it does not ‘meaningfully alter the taxpayer’s economic position

-15- (other than with regard to the tax consequences).’” Id. (quoting Salem, 786 F.3d at 950). Indeed, unlike many longer-term investments which may not initially be profitable, but which nonetheless have economic substance, STARS’s trust component involved little to no economic risk outside of tax risk—that the IRS or HMRC might disallow some of the credits that were necessary for the transaction to be profitable to both parties. Id. And although Bx was not necessarily conditioned on Barclays actually receiving its particular tax benefits from HMRC, it was also “not truly independent of the expected U.K. tax effects.” Id. Indeed, Wells Fargo’s “ability to benefit economically from the Bx payments depended on Barclays’[s] receipt of its expected tax benefits, which in turn depended on the Trust’s U.K. tax payments.” Salem, 786 F.3d at 944.

Our conclusion is bolstered by our observation that, outside of its tax benefits, STARS’s trust component was essentially comprised of economically meaningless and circular cash flows. See Wells Fargo & Co. v. United States, 641 F.3d 1319, 1330 (Fed. Cir. 2011) (noting that “purely circular transactions that elevate form over substance” are often characteristics of “abusive tax shelters”). The assets in the trust appeared to be effectively under the control of Wells Fargo at all times, see also BNY, 801 F.3d at 118-19, and their placement in this trust created no value for Wells Fargo outside of generating the foregoing transaction costs and expected tax benefits. See Santander, 844 F.3d at 25. Again, this is further supported by the fact that there was seemingly no economic risk, apart from tax risk, in this operation. See Salem, 786 F.3d at 951 (“Rather than being a genuine business transaction involving economic risk, the STARS Trust transaction was simply a money machine. . . . The artificiality of the transaction is shown by its unlimited capacity to generate gains, without any additional exposure or commitment of resources.”).

-16- B.

Next, we consider the second part of the sham-transaction test—whether the taxpayer had any subjective business purpose in entering into the transaction outside of tax considerations. See WFC Holdings, 728 F.3d at 743. We find no error in the jury’s determination that Wells Fargo lacked a valid business purpose, apart from tax reasons, to enter into the transaction.

Wells Fargo argues that the Bx payments gave it an economic benefit of $32 million per year—therefore, it contends that it had at least one valid business reason for entering into the transaction. See IES, 253 F.3d at 353 (noting that a taxpayer satisfies the business-purpose test if it was “motivated by any economic purpose outside of tax considerations” (emphasis added)). Alternatively, it argues that this Court should adopt a flexible approach to the sham-transaction doctrine, under which both economic substance and business purpose are simply two considerations in determining whether a particular transaction is a sham. Despite suggesting a flexible approach, it also seems to advocate for a disjunctive approach, as it urges us to hold that STARS’s trust component is not sham so long as it had a reasonable possibility of pre-tax profit.

Wells Fargo’s arguments on this issue are unconvincing. For the reasons given above, STARS’s trust component had no reasonable possibility of pre-tax profit, even assuming that Bx was pre-tax income to Wells Fargo. For this reason, we reject Wells Fargo’s argument that Barclays’s payment of Bx was, as a matter of law, sufficient to show that Wells Fargo had a non-tax economic purpose for entering into the transaction. Moreover, the jury was presented with ample evidence that Wells Fargo was motivated to engage in the transaction solely for tax purposes, including Wells Fargo’s and Barclays’s assessment of STARS as a tax-driven transaction— indeed,

-17- there was significant evidence presented at trial showing that Wells Fargo viewed STARS as a “tax strategy” or “tax trade.” This included evidence that tax advisers such as KPMG were promoting STARS to Wells Fargo as a prepackaged tax strategy to reduce Wells Fargo’s tax liability, that Wells Fargo knew that KPGM’s proposed fee was a percentage of the tax benefits Wells Fargo obtained through STARS, and that Wells Fargo and Barclays internally characterized the benefits of STARS as tax driven. Wells Fargo has not offered any persuasive argument for why the jury’s assessment is incorrect in light of the record evidence.6

III.

Next, we turn to Wells Fargo’s appeal from the district court’s application of the negligence penalty. We review de novo Wells Fargo’s arguments that the district court erred as a matter of law in imposing the negligence penalty. See Chemtech Royalty Assocs., L.P. v. United States, 823 F.3d 282, 287 (5th Cir. 2016); see also Scherbart v. Comm’r, 453 F.3d 987, 989 (8th Cir. 2006).

A taxpayer is liable for a “negligence penalty” of twenty percent of an underpayment of its taxes attributable to its “negligence.” 26 U.S.C. § 6662(b)(1). In this context, “negligence” is defined both by statute, see 26 U.S.C. § 6662(c), and by regulation, see 26 C.F.R. § 1.6662-3(b)(1). The applicable regulation, however, creates a defense to the negligence penalty if the taxpayer’s “return position” was

6 Because Wells Fargo has not demonstrated that the transaction had either economic substance or a valid business purpose, we need not address whether this Court uses a disjunctive, conjunctive, flexible, or rigid approach in applying the sham- transaction doctrine. See WFC Holdings, 728 F.3d at 743-44.

-18- “reasonably based on one or more of the authorities set forth in § 1.6662-4(d)(3)(iii) (taking into account the relevance and persuasiveness of the authorities, and subsequent developments).” 26 C.F.R. §§ 1.6662-3(b)(1), (3). This is known as the “reasonable-basis” defense. A “return position” is a particular position taken by the taxpayer on its tax return. 26 C.F.R. § 301.6114-1(a)(2)(i) (“A taxpayer is considered to adopt a ‘return position’ when the taxpayer determines its tax liability with respect to a particular item of income, deduction or credit.”).

The parties dispute whether the reasonable-basis defense requires evidence that a taxpayer actually relied on relevant legal authority which supports its return position. Wells Fargo argues that its return position was objectively reasonable under the relevant legal authorities. Accordingly, it contends that it is irrelevant whether it actually relied upon those authorities in forming its return position. The government, however, asserts that a taxpayer cannot “base” its return position on the relevant authorities without showing that it actually relied on those authorities. Because Wells Fargo did not submit any evidence that it subjectively based its return position on legal authority, the government submits that the district court correctly applied the negligence penalty. Alternatively, the government argues that Wells Fargo lacked an objectively reasonable basis for its return position.

We agree with the government that the reasonable-basis defense requires evidence of actual reliance on the relevant authority on the part of the taxpayer. We start with the plain language of the regulation, see Solis v. Summit Contractors, Inc., 558 F.3d 815, 823 (8th Cir. 2009), which provides a defense to the negligence penalty only when the taxpayer’s “return position is reasonably based on one or more [relevant] authorities.” 26 C.F.R. § 1.6662-3(b)(3) (emphasis added). The plain or common usage of the word “base” suggests that one is relying on particular information in order to form an opinion or a position about something. See Base,

-19- Black’s Law Dictionary (10th ed. 2014) (defining “base,” in part, as “[t]o use (something) as the thing from which something else is developed”). Thus, in order to “base” a return position on particular legal authority, a taxpayer must show that it actually relied upon those authorities in forming its position. As the district court noted, “[i]t is difficult to know how a taxpayer could ‘base’ a return position on a set of authorities without actually consulting those authorities, just as it is difficult to know how someone could ‘base’ an opinion about the best restaurant in town on Zagat ratings without actually consulting any Zagat ratings.” Wells Fargo II, 260 F. Supp. 3d at 1148. Indeed, the regulation does not require the taxpayer’s position to be simply “consistent with” or “supported by” the relevant legal authority. If it did, then it might be sufficient that the relevant authorities supported the taxpayer’s position, regardless of whether the taxpayer relied upon them. But in order for a taxpayer to “base” its position on relevant authority, it must have actually known about those authorities and actually relied upon them when forming its return position. See Candyce Martin 1999 Irrevocable Trust v. United States, 822 F. Supp. 2d 968, 1013 (N.D. Cal. 2011) (“Reasonable basis requires reliance on legal authorities.”) aff’d in part, rev’d in part on other grounds, 739 F.3d 1204 (9th Cir. 2014); accord Blue Mountain Energy, Inc. v. United States, No. 2:14-cv-418-DN, 2016 WL 4179366, at *11 (D. Utah Aug. 5, 2016) (denying summary judgment on reasonable-basis defense because there was “an issue of fact whether [taxpayer] was aware and relied on [relevant authorities]”). But see TIFD III-E Inc. v. United States, 8 F. Supp. 3d 142, 151 (D. Conn. 2014) (rejecting the government’s position that evidence of taxpayer’s subjective or actual reliance was necessary), rev’d on other grounds, 604 F. App’x. 69 (2d Cir. 2015).

Moreover, we think that such a reading of the regulation is sensible in light of the broader context of the statute and accompanying regulatory definitions. Again, the government is seeking to impose a “negligence penalty,” which suggests that the focus of the inquiry must be, at least in part, on the taxpayer’s actual conduct—whether it

-20- met the requisite standard of care in preparing its tax return and considering its return position—rather than simply determining whether its legal position finds support in the relevant legal authority. See 26 U.S.C. § 6662(c) (defining “negligence” as “any failure to make a reasonable attempt to comply with the provisions of this title”). Indeed, in discussing the negligence penalty, we have explicitly held that “the burden is on the taxpayer to prove that he did not fail to exercise due care or do what a reasonable and prudent person would do under similar circumstances.” Chakales v. Comm’r, 79 F.3d 726, 729 (8th Cir. 1996). Additionally, requiring evidence of actual reliance is supported by the fact that a taxpayer adopts a particular “return position” only when it actually “determines its tax liability with respect to a particular item of income, deduction or credit.” 26 C.F.R. § 301.6114-1(a)(2)(i). Accordingly, reading the phrase “reasonably based” to require evidence of actual reliance is more consistent with the broader statutory and regulatory framework.

We find unpersuasive Wells Fargo’s arguments on this issue. First, it argues that the plain text of the regulation does not obviously demand evidence of actual reliance—it asserts that Congress and the drafters of the regulations presumably knew how to explicitly require evidence of reliance in such regulatory provisions. In particular, Wells Fargo points to different statutory provisions and other regulations—including 26 U.S.C. § 6404(f)(2)(A) (requiring a taxpayer relying on the advice of an IRS employee to prove “it reasonably relied upon” that advice) and 26 C.F.R. § 1.6662-4(g)(1)(i) (requiring proof that a taxpayer actually “analyzed the pertinent facts and authorities . . . and in reliance upon that analysis” formed its conclusion)—in support of its argument. It contends that, in the absence of language explicitly requiring evidence of actual reliance, 26 C.F.R. § 1.6662-3(b)(1) requires only that a taxpayer show that its position was objectively reasonable under the relevant legal authorities. For the reasons given above, we think that this argument is contrary to a plain reading of the term “base” and is inconsistent with the broader statutory

-21- context, which centers on the taxpayer’s actual conduct. That other statutory provisions or regulations use different language in creating an actual reliance requirement does not mean that the provision at issue in this case requires only that the taxpayer’s position be objectively reasonable with respect to the relevant legal authorities.

Second, Wells Fargo argues that a subjective or actual reliance standard would likely require a taxpayer to waive attorney-client privilege in order to prove that it actually relied on the relevant legal authority. While this argument has some appeal, this adverse consequence may also be true of other provisions of the Internal Revenue Code and Treasury regulations that create defenses to penalties so long as the taxpayer can demonstrate actual reliance on IRS or independent legal advice or legal authorities. See, e.g., 26 C.F.R. § 1.6662-4(g)(1)(i). Therefore, this argument, standing alone, is not a convincing reason to adopt Wells Fargo’s reading of the regulation.

Third, Wells Fargo argues that as a policy matter, it should not matter whether a taxpayer can demonstrate actual reliance. Instead, Wells Fargo asserts that if it “gets to a reasonable position” it is irrelevant how it actually arrived there—whether by “much deliberation” or simply “by sheer luck.” We disagree. Again, the purpose of the negligence penalty is to compel taxpayers to make “a reasonable attempt to comply with the provisions” of the Internal Revenue Code. 26 U.S.C. § 6662(c); see also 26 C.F.R. § 1.6662-3(b)(1). It penalizes those who “fail to exercise due care or do what a reasonable and prudent person would do under similar circumstances.” Chakales, 79 F.3d at 729. Accordingly, there is a sound policy reason underlying a subjective or actual reliance requirement—it incentivizes taxpayers to actually conform to the requisite standard of care rather than simply taking the chance that there may be a reasonable basis for their underpayment of tax. It also reflects the understanding that a taxpayer who carefully studies the relevant legal authorities but arrives at an incorrect

-22- conclusion of law, albeit with a reasonable basis for its position, is perhaps less blameworthy or culpable than a taxpayer which simply ignored the existing authorities in forming its tax position and attempts to generate a reasonable basis as a post-hoc justification for its underpayment.

For these reasons, we hold that the reasonable-basis defense under 26 C.F.R. § 1.6662-3 requires evidence that a taxpayer actually relied on the relevant legal authorities that form the reasonable basis for its position. Because it did not provide evidence of actual reliance, we need not address whether Wells Fargo’s return position actually had a reasonable basis under the existing legal authorities.7

B.

Finally, Wells Fargo claims that the district court erred in applying the negligence penalty because the government failed to satisfy the requirements of 26 U.S.C. § 6751. The statute states that “[n]o penalty under this title shall be assessed unless the initial determination of such assessment is personally approved (in writing) by the immediate supervisor of the individual making such determination or such higher level official as the Secretary may designate.” Id. § 6751(b)(1). A sister circuit has held that this requirement is “a mandatory, statutory element of a penalty claim.”

7 We note that the district court found that the regulation was ambiguous, requiring it to give Auer deference to the IRS’s interpretation that the regulation requires evidence of actual reliance. Wells Fargo II, 260 F. Supp. 3d at 1148-49; see also Definition of Reasonable Basis, 63 Fed. Reg. 66433, 66433 (Dec. 2, 1998). As we do not find the regulation to be ambiguous, we decline to reach this issue or consider whether it is still appropriate to defer to the government’s interpretation of the regulation in light of Kisor v. Wilkie, 139 S. Ct. 2400 (2019) (imposing restrictions on the application of Auer deference).

-23- Chai v. Comm’r, 851 F.3d 190, 222 n.26 (2d Cir. 2017). Because the government failed to present any evidence at trial that the negligence penalty was “personally approved (in writing)” by an appropriate official under the statute, Wells Fargo asserts that the government failed to satisfy one of its statutory elements. Accordingly, Wells Fargo contends that the imposition of the negligence penalty must be reversed as a matter of law.

We do not think that the prior-approval requirement in § 6751 applies in this case. By its terms, the statute requires prior written approval to be obtained when the government “assesses” a penalty against a taxpayer. But the negligence penalty in this case is an offset defense in a refund action, not an actual “assessment” by a revenue agent during an audit or by the IRS in a deficiency lawsuit. And because the negligence penalty in this case is an offset in litigation, it will, therefore, never be “assessed” against the taxpayer or even collected from the taxpayer by the IRS—rather, the amount is merely withheld from the amount that the government is ordered to refund to the taxpayer. The existing body of caselaw draws a distinction between the government “assessing” a penalty and asserting a penalty as a litigation offset, noting that the government’s procedural requirements in the latter are significantly relaxed. For example, the statute of limitations generally does not apply when the government asserts the penalty as an offset defense in litigation, even when the statute of limitations would otherwise bar the government from seeking the penalty as an assessment against the taxpayer. See Lewis v. Reynolds, 284 U.S. 281, 283 (1932) (“Although the statute of limitations may have barred the assessment and collection of any additional sum, it does not obliterate the right of the United States to retain payments already received when they do not exceed the amount which might have been properly assessed and demanded.”); see also Allen v. United States, 51 F.3d 1012, 1014-15 (11th Cir. 1995). In short, because the negligence penalty in this case is an offset defense in a refund

-24- action—not an “assessment”—the prior-approval requirement in § 6751(b)(1) does not apply.8

IV.

For these reasons, we affirm the judgment of the district court.

GRASZ, Circuit Judge, dissenting in part.

I join all but Section III of the court’s opinion. I respectfully dissent from the court’s decision to affirm the district court’s imposition of the negligence penalty on Wells Fargo. The district court wrongly required Wells Fargo to show actual reliance on certain authorities to qualify for the reasonable-basis defense permitted under 26 C.F.R. § 1.6662-3(b)(3).

The district court reached this decision, in part, because it believed the regulation was ambiguous and thus it needed to defer to the IRS’s interpretation of its own regulation. See Auer v. Robbins, 519 U.S. 452, 461 (1997). The court today avoids giving Auer deference, but affirms the district court’s decision because it interprets 26 C.F.R. § 1.6662-3(b)(3) to require actual reliance. Ante at 19. I disagree with both conclusions.

8 The district court did not reach this issue, finding that Wells Fargo waived this argument, an issue which the parties continue to dispute on appeal. We need not address the waiver issue, however, because even assuming, without deciding, that Wells Fargo properly preserved the issue, its argument fails as a matter of law. See Lane v. Peterson, 899 F.2d 737, 742 (8th Cir. 1990) (noting that “[w]e may affirm a judgment on any ground supported by the record even if not relied upon by the district court”).

-25- I.

First, I believe the district court wrongly gave so-called Auer deference to the IRS’s interpretation of its own regulation. As the Supreme Court recently explained, in order for the Auer-deference doctrine to apply, certain factors must be present — (1) the regulation must be genuinely ambiguous; (2) the agency’s interpretation of the regulation must be reasonable; (3) the interpretation must be the agency’s authoritative or official position; (4) the interpretation must in some way implicate the agency’s substantive expertise; and (5) the interpretation must reflect fair and considered judgment — in other words, we should not defer to an agency’s convenient litigating position or post hoc rationalization. See Kisor v. Wilkie, 139 S. Ct. 2400, 2415–18 (2019).

Here, I believe the regulation is ambiguous and the IRS’s position is reasonable. But I doubt any of the three remaining factors are present, particularly the requirement that the IRS’s interpretation must implicate its substantive expertise. As the Supreme Court explained, courts should presume Congress intended to invest interpretative power in whichever actor was best positioned to develop expertise about the issue. See id. at 2417. When interpreting a technical rule, an agency is likely best positioned to interpret the ambiguity. Id. But “[s]ome interpretive issues may fall more naturally into a judge’s bailiwick.” Id. The question here is whether the reasonable-basis exception requires proof of actual reliance. In effect, a legal standard is at issue. I believe the judiciary is better positioned than the agency to decide this question of the proper legal standard. No expertise possessed by the IRS warrants us taking a backseat on this issue. Thus, in light of Kisor’s directive, I see no reason why the judicial branch should defer to the agency here.

-26- II.

The court today avoids deferring to the IRS, while reaching the same conclusion as the district court based on its de novo interpretation of 26 C.F.R. § 1.6662-3(b)(3). Ante at 19. This court focuses on the use of the word “based” and holds “in order to ‘base’ a return position on particular legal authority, a taxpayer must show that it actually relied upon those authorities in forming its position.” Id. I am unconvinced.

I begin with the text of the regulation. It states, “[a] return position that has a reasonable basis as defined in paragraph (b)(3) of this section is not attributable to negligence.” 26 C.F.R. § 1.6662-3(b)(1). Paragraph (b)(3) explains a “[r]easonable basis is a fairly high standard of tax reporting, that is, significantly higher than not frivolous or not patently proper” and “is not satisfied by a return position that is merely arguable or that is merely a colorable claim.” 26 C.F.R. § 1.6662-3(b)(3). It further provides that “[i]f a return position is reasonably based on one or more of the authorities set forth in § 1.6662-4(d)(3)(iii) . . . , the return position will generally satisfy the reasonable basis standard even though it may not satisfy the substantial authority standard as defined in § 1.6662-4(d)(2).” Id. (emphasis added). And § 1.6662-4(d)(3)(iii) provides the “authorit[ies] . . . for determining whether there is [a reasonable basis]9 for the tax treatment of an item,” and such authorities include the “[a]pplicable provisions of the Internal Revenue Code and . . . court cases.” 26 C.F.R. § 1.6662-4(d)(3)(iii) (footnote added).

9 The regulation as written applies to the substantial-authority exception. But when setting forth the reasonable-basis exception, § 1.6662-3(b)(3) cross references § 1.6662-4(d)(3)(iii) for the list of permissible authorities on which to base the decision.

-27- Nowhere in this language, which is cast in objective terms, does the term “reliance” appear. This is notable because § 1.6662-4, which is cross-referenced in § 1.6662-3(b), includes a reliance element when describing the “reasonable belief” defense. See 26 C.F.R. § 1.6662-4(g)(4) (explaining “a taxpayer is considered reasonably to believe that the tax treatment of an item is more likely than not the proper tax treatment if . . . (A) [t]he taxpayer analyzes the pertinent authorities in the manner described in paragraph (d)(3)(ii) of this section, and in reliance upon that analysis . . .; or (B) [t]he taxpayer reasonably relies in good faith on the opinion of a professional tax advisor . . . .”) (emphasis added).

The Supreme Court has explained that when “Congress includes particular language in one section of a statute but omits it in another section of the same Act, it is generally presumed that Congress acts intentionally and purposely in the disparate inclusion or exclusion.” Russello v. United States, 464 U.S. 16, 23 (1983) (quotation omitted). I see no reason why the same canon of statutory construction would not apply when interpreting the regulation here. See Black & Decker Corp. v. C.I.R., 986 F.2d 60, 65 (4th Cir. 1993) (“Regulations, like statutes, are interpreted according to canons of construction.”). If the IRS wanted to require actual reliance on the specified authorities to satisfy the reasonable-basis defense, it could have expressly said so, as it did in setting forth eligibility for the reasonable-belief defense. Its failure to do so indicates actual reliance is not required.

In support of its contrary conclusion, the court borrows an incisive and colorful analogy offered by the district court — “[i]t is difficult to know how a taxpayer could ‘base’ a return position on a set of authorities without actually consulting those authorities, just as it is difficult to know how someone could ‘base’ an opinion about the restaurant in town on Zagat ratings without actually consulting any Zagat ratings.” Ante at 19–20 (quoting Wells Fargo & Co. v. United States, 260 F. Supp. 3d 1140,

-28- 1148 (D. Minn. 2017)). The analogy raises a good point. But I believe it is based on a faulty premise.

It assumes the taxpayer must base its position on the specified authorities before the return is filed. The regulation makes no such demand. Instead, 26 C.F.R. § 1.6662-3(b)(3) simply provides that a return position will generally be considered — presumably by the agency or the courts — to have a reasonable basis if it is based on one of the authorities designated in 26 C.F.R. § 1.6662-4(d)(3)(iii). And § 1.6662- 4(d)(3)(iii) further indicates that the agency or courts should consider only such designated authority to make its determination. Reading these regulations together, I believe the agency and/or the courts — not the taxpayer — are to make the determination whether there was a reasonable basis for a return position based on the specified authorities.

To illustrate this distinction, let us alter the district court’s restaurant analogy. Suppose three friends try to decide where to go for dinner. Two of the friends, Friend A and Friend B, offer differing suggestions, each claiming his suggestion is the best restaurant in town. Tasked with resolving the dispute, Friend C consults Zagat to see which of the two recommended restaurants is indeed “the best,” and, after doing so, sides with Friend A. Friend C’s decision was indeed based on the Zagat ratings. But Friend A did not rely on the Zagat ratings when taking his position. In other words, Friend C’s determination was based on Zagat, regardless of whether Friend A ever relied on the service.

In my view, the court is more like Friend C, in that we are tasked with resolving the debate between the United States and Wells Fargo as to whether Wells Fargo’s position had a reasonable basis. To decide, the court may find a reasonable basis if the

-29- position is supported by authorities designated in the regulation. This is true whether or not Wells Fargo actually relied on these authorities.

In sum, I do not believe the reasonable-basis defense required Wells Fargo to show it actually relied on the relevant authorities in forming its return position. Admittedly, that does not end the analysis. It still must be decided whether there was indeed an objectively reasonable basis for Wells Fargo’s position. Because this analysis was not reached by the district court, I would remand the case to the district court to decide this issue in the first instance. ______________________________

-30-