Prof. William Byrnes (Texas A&M Law) is the author of the treatises Practical Guide to U.S. Transfer Pricing and Taxation of Intellectual Property & Technology. 

“an arm’s length result is not simply any result that maximizes one’s tax obligations”

In a double take two-to-one decision because of a withdrawn decision due to the death of a judge, a Ninth Circuit panel in Altera reversed a unanimous en banc decision of the Tax Court that the qualified cost sharing arrangements (QCSA) regulations[1] were invalid under the Administrative Procedure Act.[2]  The renown Professor Calvin Johnson (Texas) and I shared comments on this case.  Professor Johnson’s pragmatism is worth noting (see his latest Altera article here) in the context of Altera: “$100 million of stock options is a $100 million cost, as a matter of law.” Because it is a cost for public accounting, Calvin states it is incredulous hat Altera would enter into a arm’s length negotiation in which the counterparty invests $200 cash, and Altera invests $200 cash plus $100 million stock options, but then Altera agrees to ignore its additional $100 million cost and agrees to split equally.  Altera wants to deduct its $100 million of stock cost domestically but pass on the associated income to the foreign-controlled group member.  This is bad policy.

I agree with Professor Johnson that it is bad policy.  But I think that Treasury is taking shortcuts to generate the result that it wants instead of going through the steps necessary to effect a change in policy. Most of my academic colleagues support the majority’s opinion of the proposition that Congress bestowed such latitude to Treasury in IRC § 482.  I agree that the latitude is within the Code Section, but that Treasury to date has regulated a policy dependent on the arm’s length and comparables, as the dissent enunciates and the Ninth Circuit panel majority supported in Xilinx II.  Treasury may change its policy approach, but that requires a formal procedural process laid out by the APA, I argue in favor of the dissent’s approach.  Even with the new language added to IRC § 482 by the TCJA of 2017, Treasury, I propose, must still formally open a public process that it is changing tact from arm’s length and comparables to something else like apportionment of profits and loss by formulae.

The last word has not been heard in Altera.  I expect that Altera will request an en banc hearing.  However, Altera II may be the case that the two newest members, in particular, Justices Kavanaugh and Gorsuch, of the Supreme Court have been waiting for to weigh in on Chevron and State Farm.  Expect Altera III.

Altera I and Altera II (withdrawn)

The Ninth Circuit’s issuance, withdrawal, and re-issuance of a CSA decision is also a double take of Xilinx.^[3] However unlike Altera, after the withdrawal of its initial Xilinx decision favoring the IRS position, the Ninth Circuit rejected the IRS’ position that the (pre-2003) QCSA Regulations required treating deductions for stock-based compensation as costs that must be shared by the foreign related party in cost-sharing arrangements. The former QCSA regulations, and current ones still, require that related entities share the cost of employee stock compensation in order for their cost-sharing arrangements to be classified as qualified cost-sharing arrangements. Treasury has consistently stated that the previous and current versions of the QCSA regulations are consistent with the arm’s length standard whereas the Tax Court has consistently disagreed with the IRS position.

At the Tax Court level for Altera, the Court held that the current QCSA regulations are a legislative rule because the regulations have the force of law, as opposed to an interpretive rule, and thus the State Farm standard applied.[4] The Tax Court concluded that Treasury did not undertake “reasoned decision making” required by State Farm in issuing the cost-sharing regulations because Treasury failed to support with any evidence in the administrative record its opinion that unrelated parties acting at arm’s length would share stock-based compensation (SBC) costs.[5] The Tax Court held that Treasury’s decision-making process relied on speculation rather than on hard data and expert opinions and that Treasury ignored public comments evidencing that unrelated party cost-sharing arrangements did not share stock compensation costs.

The Ninth Circuit’s first panel’s opinion, now withdrawn, held that Treasury did not exceed its authority delegated by Congress under IRC § 482.[6] That panel explained that IRC § 482 does not speak directly to whether Treasury may require parties to a QCSA to share employee stock compensation costs in order to receive the tax benefits associated with entering into a QCSA. The first panel held that the Treasury reasonably interpreted IRC § 482 as an authorization to require internal allocation methods in the QCSA context, provided that the costs and income allocated are proportionate to the economic activity of the related parties and concluded that the regulations are a reasonable method for achieving the results required by the statute. Thus, the first panel granted Chevron deference to the QCSA regulations.

The primary issue of Altera I and II, and the cases that precede it that have found in favor of the taxpayers is whether the arm’s length standard requires the comparability standard be met through a method seeking evidence of empirical data or known transactions?  Alternatively, is Treasury afforded deference to disregard a comparability method to instead seek an arm’s length result of tax parity that relies on an internal method of allocation to allocate the costs of the U.S. employee stock options between the U.S. and foreign related parties in proportion to the income enjoyed by each, determined post facto (after the fact) of the cost-sharing agreement?[7]

Altera II’s majority, relying on Frank,[8] states that the arm’s length standard need not be based solely on comparable transactions for reallocating costs and income, though recognizing that Frank is limited[9] to situations wherein it is difficult to hypothesize an arm’s length transaction.  The dissenting Judge provided a descriptive history that Treasury has repeatedly asserted that a comparability analysis is the only way to determine the arm’s length standard. Regarding Frank, the dissent stated, “The majority’s attempt to breathe life back into Frank is, simply, unpersuasive.” The Judge emphasized that the Ninth Circuit had declared Frank an outlier because (a) the parties in Frank had stipulated to applying a standard other than the arm’s length, (b) “there was no evidence that arms-length bargaining upon the specific commodities sold had produced a higher return,” and (c) that the complexity of the circumstances surrounding the services rendered by the subsidiary made it “difficult for the court to hypothesize an arm’s length transaction.”[10]

Pre Altera

The regulatory rules for cost-sharing arrangements (“CSAs”) at issue in Altera I and II, issued in temporary form January 5, 2009[11] and in subsequent final form effective December 16, 2011,[12] are different from the previously issued CSAs. The rules for earlier CSAs are subject to grandfather provisions.  For periods before January 5, 2009, the status of an arrangement as a CSA and the operative rules for complying arrangements, including rules for buy-in transactions, were determined under the qualified cost sharing arrangement regulations issued in 1995 and substantively amended in 1996 and 2003 (the “2003 QCSA Regulations”).[13]

The Ninth Circuit, in Xilinx,^[14] rejected the position of the Service that the pre-2003 QCSA Regulations in effect in 1997–99 required treating deductions for stock-based compensation as costs that must be shared in cost-sharing arrangements.

The purpose of the regulations is parity between taxpayers in uncontrolled transactions and taxpayers in controlled transactions. The regulations are not to be construed to stultify that purpose. If the standard of arm’s length is trumped by 7(d)(1), the purpose of the statute is frustrated. If Xilinx cannot deduct all its stock option costs, Xilinx does not have tax parity with an independent taxpayer. Xilinx, Inc. v. Comm’r, 598 F.3d 1191, 1196, 2010 U.S. App. LEXIS 5795, *14 (9th Cir 2010)

The Xilinx concurring opinion summarizes the positions at odds between Xilinx and the IRS:

The parties provide dueling interpretations of the “arm’s length standard” as applied to the ESO costs that Xilinx and XI did not share. Xilinx contends that the undisputed fact that there are no comparable transactions in which unrelated parties share ESO costs is dispositive because, under the arm’s length standard, controlled parties need share only those costs uncontrolled parties share. By implication, Xilinx argues, costs that uncontrolled parties would not share need not be shared.

On the other hand, the Commissioner argues that the comparable transactions analysis is not always dispositive. The Commissioner reads the arm’s length standard as focused on what unrelated parties would do under the same circumstances, and contends that analyzing comparable transactions is unhelpful in situations where related and unrelated parties always occupy materially different circumstances. As applied to sharing ESO costs, the Commissioner argues (consistent with the tax court’s findings) that the reason unrelated parties do not, and would not, share ESO costs is that they are unwilling to expose themselves to an obligation that will vary with an unrelated company’s stock price.  Related companies are less prone to this concern precisely because they are related — i.e., because XI is wholly owned by Xilinx, it is already exposed to variations in Xilinx’s overall stock price, at least in some respects. In situations like these, the Commissioner reasons, the arm’s length result must be determined by some method other than analyzing what unrelated companies do in their joint development transactions. Xilinx, Inc. v. Comm’r, 598 F.3d 1191, 1197, 2010 U.S. App. LEXIS 5795, *16-17 (9th Cir 2010)

The concurring Judge concludes: “These regulations are hopelessly ambiguous and the ambiguity should be resolved in favor of what appears to have been the commonly held understanding of the meaning and purpose of the arm’s length standard prior to this litigation.”

The Treasury amended the QCSA in 2003 to explicitly provide that the intangible development costs that must be shared include the costs related to stock-based compensation. From January 5, 2009, the 2009/2011 QCSA Regulations apply (the “2009 QCSA Regulations”). For periods starting with January 5, 2009, a pre-January 5, 2009 arrangement that qualified as a CSA under the 2003 QCSA Regulations is subject in part to the 2003 QCSA Regulations and in part to the 2009 QCSA Regulations. Arrangements that qualified as CSAs under the 2003 QCSA Regulations, whether or not materially expanded in scope on or after January 5, 2009, are known as “grandfathered CSAs.” The IRS contends that grandfathered CSAs are subject, with significant exceptions, to the 2009 QCSA regulations provisions for cost sharing transactions (“CSTs”) and platform contribution transactions (PCTs).  The significant exceptions for the grandfathered CSAs include that, unless the CSA is later expanded by the related parties, the original pre-2009 CSA is not subject to the 2009 QCSA regulations ‘Divisional Interest’ and Periodic Adjustment rules.

However, the IRS attempted to adjust the application of the 2003 QCSA Regulations by issuing a Coordinated Issue Paper on Section 482 CSA Buy-In Adjustments on September 27, 2007 (the “2007 CSA-CIP”).[15] The CSA-CIP was de-coordinated effective June 26, 2012, after the rejection of its concepts in the 2009 Tax Court decision in the VERITAS case. [16] The CSA-CIP provided that the Income Method and the Acquisition Price Method, similar to the specified transfer pricing methods set forth in the 2009 QCSA Regulations, are to be considered ‘best methods’ under the 2003 QCSA Regulations even though they only could be applied as ‘unspecified methods’. The Tax Court in VERITAS, addressing assessments for the tax years 2000 and 2001, neither cited nor followed the IRS methods of its 2007 CSA-CIP. Note that VERITAS survives Altera II because the 2009 QCSA Regulations years were not yet promulgated for the years of concern.  From the IRS’ perspective, though it does not acquiesce in the decision, it cured VERITAS by including the Income Method and the Acquisition Price Method as specified methods for determining “buy-in” payments for the 2009 QCSA regulations buy-ins.  Thus, the IRS continues to aggressively litigate in favor of these methods, exemplified by the appeal from Altera[17] and Amazon[18] in 2017.

Post Altera

Although the IRS withdrew the CSA-CIP in 2012, it continues to pursue cases under the pre-2009 Treasury Regulations as is the CSA-CIP remained in place. Amazon filed a Tax Court petition in December of 2012 challenging a $2 billion transfer pricing adjustment related to a qualified cost sharing arrangement between Amazon.com Inc. and its European subsidiary pre-2009.  Amazon claimed that the IRS erred in relying on a discounted cash flow method which the tax court clearly rejected in VERITAS. In the 207-page Amazon opinion, the Tax Court ruled that the IRS’s adjustment with respect to a buy-in payment for the intragroup CSA was arbitrary, capricious, and unreasonable.

Moreover, the IRS has an ongoing CSA controversy against Microsoft for the 2004-06 tax years for which President George Bush’s former Treasury Secretary John Snow promised at a February 7, 2006 hearing to then Chairman of the Committee Senator Charles E. Grassley that the IRS would bring a substantial CSA adjustment.[19] Microsoft reported an effective tax rate for fiscal years 2016, 2017, and 2018 of 15 percent, eight percent, and 19 percent respectively.[20] Microsoft reported that this unresolved transfer pricing issue is the primary cause for it to increase its tax contingency from $11.8 billion to $13.5 billion to $15.4 billion.[21] The IRS has not issued a deficiency because the controversy remains in the IDR stage of the audit currently due to litigation over the issues of legal privilege and the issue of the IRS’ contract with a third party law firm to assist in the audit.[22]

The IRS announced in 2016 and 2018 a CSA adjustment against Facebook for the tax years 2010 and subsequent of at least $5 billion, and of 2011 – 2013 of approximately $680 million.[23] Facebook reported an effective tax rate of 13 percent for the second quarter of 2017 and 2018.[24] The controversy remains in the procedural phase on the docket of the Tax Court. The Microsoft and Facebook controversies appear to be further second take of Amazon and Altera.

Based on Treasury’s litigation stances and the 2015 temporary CSA regulations proposals, Treasury updated several International Practice Service Transaction Units’ audit guidelines relevant for CSAs, including (1) Pricing of Platform Contribution Transaction (PCT) in Cost Sharing Arrangements (CSA)—Initial Transaction, (2) Change in Participation in a Cost Sharing Arrangement (CSA)—Controlled Transfer of Interests and Capability Variation, (3) Pricing of Platform Contribution Transaction (PCT) in Cost Sharing Arrangements (CSA) Acquisition of Subsequent IP, (4) Comparison of the Arm’s Length Standard with Other Valuation Approaches—Inbound, and (5) IRC 367(d) Transactions in Conjunction with Cost Sharing Arrangements (CSA).

Altera’s Double Take Analysis Of Majority and Dissenting Opinions (Read the Altera II Decision here)

The Ninth Circuit Court majority evaluated the validity of Treasury’s regulations under both Chevron and State Farm, which the Court stated: “provide for related but distinct standards for reviewing rules promulgated by administrative agencies.”[25] The majority distinguished State Farm from Chevron in that State Farm “is used to evaluate whether a rule is procedurally defective as a result of flaws in the agency’s decision-making process,” whereas Chevron “is generally used to evaluate whether the conclusion reached as a result of that process—an agency’s interpretation of a statutory provision it administers—is reasonable.”  The majority first turned to the Chevron analysis that:[26]

“When Congress has ‘explicitly left a gap for an agency to fill, there is an express delegation of authority to the agency to elucidate a specific provision of the statute by regulation,’ and any ensuing regulation is binding in the courts unless procedurally defective, arbitrary or capricious in substance, or manifestly contrary to the statute.”

The Ninth Circuit Court panel’s majority resolved that IRC § 482 is ambiguous because it does not address share employee stock compensation costs.[27] The majority stated that it is not persuaded by Altera’s argument that stock-based compensation is not “transferred” between parties because only intangibles in existence can be transferred. Altera argues that QCSAs to “develop” intangibles does not constitute a “transfer” of intangibles. The majority instead concludes that the transfer of intangibles may include the transfer of future distribution rights to intangibles which stock-based compensation are albeit yet to be developed. The majority relies upon the expansive meaning of the statutory word “any” for IRC § 482 (“any” transfer . . . of intangible property).[28] But the dissent counters that “any” does not modify “intangible property.” Rather, “any” precedes and thus, applies only to “transfer.”[29]

The majority accepts Treasury’s new explanation that the taxpayer’s agreement to “divide beneficial ownership of any Developed Technology” constitutes a transfer of intangibles.[30]  The dissenting Judge points out that Treasury never made, much less supported, a finding that QCSAs constitute transfers of intangible property.[31] The dissent states that:[32]

“No rights are transferred when parties enter into an agreement to develop intangibles; this is because the rights to later-developed intangible property would spring ab initio to the parties who shared the development costs without any need to transfer the property. And, there is no guarantee when the cost-sharing arrangements are entered into that any intangible will, in fact, be developed.”

The majority next turned to the reasonableness of Treasury ignoring the comparables presented by the Taxpayer and during the regulatory comment period.  The majority quotes from an aspect of the legislative history:[33]  

 “There are extreme difficulties in determining whether the arm’s length transfers between unrelated parties are comparable. . . . [I]t is appropriate to require that the payment made on a transfer of intangibles to a related foreign corporation be commensurate with the income attributable to the intangible.”

The majority concludes that Congress granted Treasury the authority to develop methods that did not rely on the analysis of ‘problematic’ comparable transactions and that Treasury promulgated the QCSA based on this authority because Treasury stated, “The uncontrolled transactions cited by commentators do not share enough characteristics of QCSAs involving the development of high-profit intangibles…”.[34]

The dissenting Judge pointed out that Treasury merely cited to the general legislative history IRC § 482 1986 amendment but that Treasury “did not explain what portions of the legislative history it found pertinent or how any of that history factored into its thinking.”[35] The dissenting Judge holds out that the majority accepts the “ever-evolving post-hoc rationalizations” of Treasury and then “goes even further to justify what Treasury did here”.[36] Commentators of the 2009 QCSA regulations submitted comparable transactions demonstrating that unrelated companies do not share the cost of stock-based compensation. Treasury distinguished these uncontrolled transactions as not sharing enough characteristics of QCSAs involving the development of high-profit intangibles. The dissent agreed with the Tax Court which held that Treasury’s explanation for its regulation was insufficient under State Farm because Treasury “failed to provide a reasoned basis” for its “belief that unrelated parties entering into QCSAs would generally share stock-based compensation costs.”[37]

The dissenting Judge explained that the legislative history and plain reading of the second sentence of IRC § 482 did not offer Treasury the flexibility to depart from a comparability analysis required by the first sentence but for a limited context of “any transfer (or license) of intangible property”.  The Judge then pointed out that Treasury’s 1988 White Paper also stated: “intangible income must be allocated on the basis of comparable transactions if comparables exist.”[38]  Thus, the Tax Court’s found for Xilinx because the IRS had not provided evidence that unrelated parties transacting at arm’s length share expenses related to stock-based compensation.[39]  The Ninth Circuit majority upheld the finding in favor of Xilinx because the arm’s length standard required that stock-based compensation expenses would not be shared in the absence of evidence that unrelated parties would share these costs.[40]

The majority next concludes that Treasury complied with the procedural requirements of the Administrative Procedures Act (“APA”) so that the 2009 QCSA survives a State Farm analysis.[41] The State Farm analysis second step requires that the Treasury “must consider and respond to significant comments received during the period for public comment.”[42] The majority summarizes Altera’s four arguments that Treasury did not meet this requirement: (1) Treasury improperly rejected comments submitted in opposition to the proposed rule, (2) Treasury’s current litigation position is inconsistent with statements made during the rulemaking process, (3) Treasury did not adequately support its position that employee stock compensation is a cost, and (4) a more searching review is required under Fox,[43] because the agency altered its position.  Boiled down, Altera argues that Treasury stated its intent to coordinate the new regulations with the arm’s length standard and then dismissed submissions addressing arm’s length comparables.

The majority was not persuaded by Altera’s argument that an arm’s length analysis requires actual transactional analysis. Altera submitted that “unrelated parties do not share stock compensation costs because it is difficult to value stock-based compensation, and there can be a great deal of expense and risk involved.”[44] Treasury responded in the 2009 QCSA that “the uncontrolled transactions cited by commentators do not share enough characteristics of QCSAs involving the development of high-profit intangibles to establish that parties at arm’s length would not take stock options into account in the context of an arrangement similar to a QCSA.”[45] The majority sided with Treasury’s justification that the lack of similar transactions led it to “employ a methodology that did not depend on non-existent comparables to satisfy the commensurate with income test and achieve tax parity.”[46]  The majority also concluded that Treasury’s use of an internal method of reallocation is consistent with the arm’s length standard because the internal method attempts to bring parity to the tax treatment of controlled and uncontrolled taxpayers as does a comparison of comparable transactions when they exist.[47]

Finally, the majority distinguished the previous, contrary, 2010 holding of the majority in Xilinx that stock-based compensation is not required to be included for a CSA. This majority stated that administrative authority was not at issue in Xilinx and that the previous panel was not called upon to consider the “commensurate with income.  The Xilinx panel had to reconcile a conflict between two rules: the specific methods of the 1994 arm’s length rule and the pre-2003 QCSA Regulations.[48]

The dissenting panel member instead concluded that the two-member majority justified Treasury’s about-face by (a) providing “a reasoned basis for the agency’s action that the agency itself has not given”,[49] (b) encouraging “executive agencies’ penchant for changing their views about the law’s meaning almost as often as they change administrations”,[50] and (c) endorsing a practice of requiring interested parties to engage in a scavenger hunt to understand an agency’s rulemaking proposals.[51]  The dissenting Judge was troubled that Treasury stated “for the first time and with no explanation that it may, instead, employ the “commensurate with income” standard to reach the required arm’s length result.”[52]

Based on the Tax Court decision in Xilinx and in Altera that the taxpayer had presented sufficient evidence of comparable transactions, the dissent agreed with the Tax Court’s finding that Treasury was required at least to attempt to gather empirical evidence before declaring that no such evidence was available, in the face of such evidence being available.  In light of this evidence, Treasury concedes the comparables issue in its appellate brief and instead pivots its justification for the 2009 QCSA that Treasury is not required to undertake an analysis of what unrelated entities do under comparable circumstances. Treasury’s argument is that it was statutorily authorized to dispense with a comparability analysis in this narrow context and thus Treasury does not need to investigate whether the uncontrolled transactions were comparable.[53] The dissenting Judge would hold that the APA requires Treasury to state that it was taking this new position in a stark departure from its previous regulations.[54]

In my opinion, Treasury had to concede the comparables point.  The issues remain the same as explained by the Xilinx concurring Judge above.  Treasury’s argument, regarding CSAs, is that related parties should be treated differently because as a group the parties have more information and more control over the other party as regards the share options.  Given the group relationship, the U.S. and the foreign party will split the costs of the U.S. employees’ share options.  But the application of the arm’s length standard has been understood to treat the related parties and unrelated.  If unrelated, then the assumption of information is unfounded.  Moreover, why would the foreign party bear the costs of the share options of the U.S. employees without negotiating on behalf of its employees to also receive such options?  What is the quid pro quo for the foreign subsidiary?  Yet, I also consider that potentially such lopsidedness in favor of the U.S. party can be brought to bear by the economic dominance of the U.S. party. which can potentially occur in an outsourcing relationship.  However, Altera and amicus industry groups provided agreements evidencing the contrary and the IRS chose not to seek rebuttal evidence (or it could not locate any). 

The issues of comparables and comparability, at least in my perspective, are distinguishable.  The first step is to identify transaction comparables, which Altera clearly has, and the second is to then to adjust for the commonly accepted (market, economic) variances between the comparables.  By example, size of parties in relation to each other, size of market and competition within, term, etc the factors of the Treasury Regs and other arm’s length differences that would require adjustments.  I disagree with the underlying premise of the “three percent”. Stated another way, 97 percent of transactions are therefore incomparable.  That’s a lot of “unicorns”.  But business is not like our fingerprints and rarely generates unicorns.  More often, competitors develop distinguishing approaches that can be adjusted for.  Said another way, I disagree with the lack of comparables, and base my disagreement on the managerial sciences like supply and value chain.  The economy does produce unicorns and we call these unicorns first movers.  Sometimes we grant patent protection to maintain unicorn status for a period of time.  And sometimes first movers develop a new formation of the supply and full value chain that we call a business method.  But for the issue of a monopoly or concentrated oligopoly, such first movers eventually experience competitors and comparables begin to emerge.  Thus, the argument for a lack of comparable transactions within an industry or industry segment necessarily requires believing that unicorns are common.

Also, the “three percent” must be viewed in historical context.  Firstly, that report was written at a time when there was a lack of available information via the Internet and electronic (pay for) databases that captured such information, cleaned it, and tagged it.  Secondly, the domestic economy itself was less mature and robust, with much less competition and thus much less transactions to be compared. Thirdly, the world was not a globalized competitive economy as it is today.  The OECD and Treasury still state a lack of comparable transactions today with regard to “hard to value intangibles”.  My academic sense thinks that it is just hard, laborious work to find them.  (And arguably for simplicity maybe as a policy we should move away from the arm’s length). 

The dissenting Judge finds that in 1986 Congress could not have legislated against the backdrop of stock-based compensation and cost-sharing arrangement because these activities did not develop until the 1990s. Thus, the dissenting Judge concludes that while “Congress may choose to address this practice now, it cannot be deemed to have done so then.”[55] In his conclusion, the Judge states “… an arm’s length result is not simply any result that maximizes one’s tax obligations.”[56] In my opinion, the ball is in Treasury’s court, not Congress’.

Lexis’ Practical Guide to U.S. Transfer Pricing is 36 chapters, 3,000 pages, updated annually to help multinationals cope with the U.S. transfer pricing rules and procedures, taking into account the international norms established by the OECD and UN. It is also designed for use by tax administrators, both those belonging to the U.S. Internal Revenue Service and those belonging to the tax administrations of other countries, and tax professionals in and out of government, corporate executives, and their non-tax advisors, both American and foreign. Fifty co-authors contribute subject matter expertise on technical issues faced by tax and risk management counsel.  Free download of chapter 2 here

END NOTES

[1]Treas. Reg. § 1.482-7A(d)(2).

[2] Altera Corp. v Commr, __ F.3d. __ (9^th Cir., June 7, 2019) (case no. 16-70496) [hereafter “Altera II”] reversing Altera Corp. v. Commr, 145 TC No 3 (July 27, 2015) [hereafter “Altera I”].

[3] Xilinx, Inc v Commr, 125 TC 37 (2005), affd, 598 F 3d 1191 (9th Cir 2010). It is noted that in 2009 the Ninth Circuit issued an opinion accepting the position of the Service, but withdrew that opinion on Jan. 13, 2010.

[4] See Am. Mining Cong. v. Mine Safety & Health Admin., 995 F.2d 1106, 1109 (D.C. Cir. 1993). Interpretive rules are excluded from the general notice requirement for proposed rulemaking by 5 U.S.C. sec. 553(b)(3)(A). See Chevron, U.S.A., Inc. v. Natural Res. Def. Council, Inc., 467 U.S. 837 (1984) that the Tax Court held incorporates the State Farm standard.

[5] Motor Vehicle Mfrs. Ass’n of the U.S. v. State Farm Mut. Auto Ins. Co., 463 U.S. 29 (1983).

[6] The Ninth Circuit’s majority stated that the summary of the first panel’s withdrawn opinion constitutes no part of the opinion of the second panel.

[7] Altera II at 6, citing Comm’r v. First Sec. Bank of Utah, 405 U.S. 394, 400 (1972) (quoting 26 C.F.R. §1.482-1(b)(1) (1971)).

[8] Frank v. Int’l Canadian Corp., 308 F.2d 520, 528–29 (9^th Cir. 1962).

[9] Oil Base, Inc. v. Comm’r, 362 F.2d 212, 214 n.5 (9th Cir. 1966).

[10] Altera II dissent at 54.

[11] 74 Fed Reg 340 (Jan 5, 2009) (the “Temporary Regulations”).

[12] 76 Fed Reg 80,082 (Dec 22, 2011) (the “Final Regulations”).

[13] Treas. Reg. § 1.482-7A. The “A” was added to the QSCA Regulations effective on January 5, 2009, when the Temporary Regulations were published.

[14] Xilinx, Inc v Commr, 125 TC 37 (2005), affd, 598 F 3d 1191 (9th Cir 2010).

[15] Coordinated Issue Paper on Section 482 CSA Buy-In Adjustments, LMSB-04-0907-62 [hereinafter CSA-CIP].

[16] VERITAS Software Corp v Commr, 133 TC 297 (2009), nonacq, 2010-49 IRB (Dec 6, 2010) (detailed explanation of the IRS’ reasoning available at http://www.irs.gov/pub/irs-aod/aod201005.pdf, assessed June 7, 2019).

[17] Altera I.

[18] Amazon.Com, Inc. v Commr, 148 TC No 8 (March 23, 2017).

[19] Unofficial Transcript of Finance Hearing on Fiscal 2007 Budget is Available, 2006 TNT 31-15 (Feb 15, 2006).

[20] Fiscal year end of June 30 for 2016 and 2017, last three months ending December 31, 2018.  Microsoft 10-K (2017) at 38; Microsoft 10-K (2018); Microsoft 10-K (2Q 2019) at Note 11-Income Taxes.

[21] Microsoft 10-K (2017) at 39; Microsoft 10-K (2Q 2019) at Note 11-Income Taxes.

[22] United States v Microsoft Corp, No 2:15-cv-00102 (WD Wash May 5, 2017).

[23] See U.S. v Facebook Inc ND Cal, No 3:16-cv-03777 (pet filed July 6, 2016).

[24] Facebook 10-Q (2Q 2017) at 20; Facebook 10-K (2018) at 35, 48.

[25] Catskill Mountains Chapter of Trout Unlimited, Inc. v. EPA, 846 F.3d 492, 521 (2d Cir. 2017).

[26] Chevron, 467 U.S. at 843–44.

[27] Altera II at 25.

[28] The Court cites United States v. Gonzales, 520 U.S. 1, 5 (1997) (“Read naturally, the word ‘any’ has an expansive meaning . . . .”) and Republic of Iraq v. Beaty, 556 U.S. 848, 856 (2009) (“Of course the word ‘any’ (in the phrase ‘any other provision of law’) has an ‘expansive meaning, giving us no warrant to limit the class of provisions of law [encompassed by the statutory provision].”

[29] Altera II dissent at 79.

[30] Altera II dissent at 67.

[31] Altera II dissent at 73.

[32] Altera II dissent at 73.

[33] See H.R. Rep. No. 99-426, at 425.

[34] Citing Compensatory Stock Options Under Section 482, 68 Fed. Reg. 51,171-02, 51,173 (Aug. 26, 2003).

[35] Altera II dissent at 63.

[36] Altera II dissent at 67.

[37] Altera II dissent at 65.

[38] Study of Intercompany Pricing under Section 482 of the Code (“White Paper”), I.R.S. Notice 88-123, 1988-1 C.B. 458, 474;

[39] Xilinx v. Commissioner (“Xilinx I”), 125 T.C. 37, 53 (2005).

[40] Altera II dissent at 58.

[41] Altera II at 33.

[42] 5 U.S.C. § 553(c); Perez v. Mortg. Bankers Ass’n, 135 S. Ct. 1199, 1203 (2015).

[43] FCC v. Fox Television Stations, Inc., 556 U.S. 502 (2009).

[44] Altera II at 36.

[45] Compensatory Stock Options under Section 482 (Preamble to Final Rule), 68 Fed. Reg. 51,171-02, 51,172–73 Aug. 26, 2003).

[46] Altera II at 39.

[47] Altera II at 41.

[48] Treas. Reg. § 1.482-1(b)(1) (1994).

[49] Motor Vehicle Mfrs. Ass’n of U.S., Inc. v. State Farm Mut. Auto. Ins. Co., 463 U.S. 29, 43 (1983) (citing SEC v. Chenery Corp. (“Chenery II”), 332 U.S. 194, 196 (1947))

[50] BNSF Ry. Co. v. Loos, 586 U.S. ___, No. 17-1042, slip op. at 9 (2019) (Gorsuch, J., dissenting)

[51] Altera II dissent at 51.

[52] in its preamble to § 1.482-7A(d)(2),

[53] Altera II dissent at 66 citing Appellant’s Br. 64.

[54] Altera II dissent at 68 citing FCC v. Fox Television Stations, Inc., 556 U.S. 502, 515 (2009) (“[T]he requirement that an agency provide reasoned explanation for its action would ordinarily demand that it display awareness that it is changing position.”).

[55] Altera II dissent at 80.

[56] Altera II dissent at 81.