Altera Deconstructed: A Nuanced Alteration in Tax Law
Tax Court’s Unanimous Opinion Deals With Unusual Regulatory Situation

On July 27, the Tax Court issued a unanimous opinion¹ in Altera Corporation v. Commissioner, 145 T.C. No. 3 (2015), in which the court held that Treas. Reg. § 1.482-7(d)(2) is invalid. It is relatively rare for a court to invalidate Treasury regulations, and, for that reason alone, the decision is worthy of review. However, Altera arose in an unusual regulatory context, and interpreting that case as heralding a new era of judicial activism by the Tax Court is likely to lead to disappointment. The lessons from Altera are there, but they are more nuanced than a seismic shift in tax jurisprudence.

Altera Corporation is a Silicon Valley company that develops, manufactures, and markets programmable logic devices (e.g., integrated circuits). As is common in high-tech industries, Altera licensed to its foreign subsidiary, Altera International, the right to use and exploit, everywhere except within the United States and Canada, the pre-existing intellectual property for Altera’s products. In exchange, Altera International agreed to pay a royalty.

In addition, Altera and Altera International entered into a cost-sharing agreement with respect to the development of future intellectual property, through which they agreed to share the cost of continuing research and development, including research using the licensed intangible property already in existence. Altera and Altera International computed all of the costs of research and development activities they performed after the date of the license and shared those costs in accordance with the reasonably anticipated benefits each would receive from exploiting the newly developed technology. Each could then exploit the new technology as it saw fit.

The taxpayer might not have prevailed on its motion for summary judgement in Altera without the numerous comments submitted to Treasury that specifically addressed whether businesses include stock-based costs in arm’s-length cost-sharing agreements.

Capturing R&D Costs

One issue in evaluating cost-sharing agreements is whether the parties are properly capturing all of the costs attributable to the research and development activity in their cost pool. If, as is often the case, most or all of the research is conducted in the United States and the costs are understated, the payment from the foreign subsidiary to the U.S. parent will be less, resulting in lower taxable income in the United States. Altera’s cost-sharing agreement included the salary cost of the personnel in the United States performing the research but not the value of stock-based compensation they received. Altera’s position was that, in a cost-sharing arrangement between unrelated parties, stock-based compensation costs would not be part of the shared cost pool, so they should not be included in the computations for Altera International.

This issue had previously been decided in taxpayers’ favor pursuant to prior regulations under Section 482.² However, for the years at issue in Altera, the Treasury Department had promulgated new Treas. Reg. § 1.482-7(d)(2), which requires participants in qualified cost-sharing arrangements to share stock-based compensation costs to achieve arm’s-length results. In Altera, the parties cross-moved for summary judgment on the narrow legal issue of whether Treas. Reg. § 1.482-7(d)(2) is arbitrary and capricious and therefore invalid. In a relatively rare outcome, the Tax Court held that Treas. Reg. § 1.482-7(d)(2) is invalid.

The question on the minds of tax professionals is whether this decision has broader implications for judicial review of tax regulations. To answer that question, one should first consider the standards under which the validity of regulations is reviewed. After that, one should carefully analyze the unique considerations that led to the Tax Court’s conclusion and determine how those considerations might be applied more broadly.

Interpreting Mayo

When the Supreme Court issued its decision in Mayo,³ it settled an arguably open question as to the standard of review to which tax regulations are subject. Prior to Mayo, many tax practitioners believed that tax regulations were subject to a less-deferential standard of review than other agency actions, pursuant to the Supreme Court’s decision in National Muffler.⁴ However, Mayo held that formal Treasury rulemaking activity is subject to review under the deferential standard announced in Chevron.⁵ Under Chevron, a court must first analyze whether a statute is ambiguous or has gaps that have been left for the agency to fill—whether Congress has directly addressed the question at issue. Where the statute is unambiguous, a regulation will be invalid unless it is consistent with the precise dictates of the unambiguous and comprehensive statute. If the statute has left a gap, the agency action will be upheld unless it is “arbitrary or capricious in substance, or manifestly contrary to the statute.”⁶

What is often lost in discussions of Mayo is the heavy emphasis the Court placed on the occurrence of bona fide notice-and-comment rulemaking. The Mayo Court emphasized the important role that notice-and-comment rulemaking plays in validating agency action, relying in three separate places on the occurrence of such procedures as a key justification for Chevron-level deference. As a result, tax professionals both inside and outside the government should have been reminded that the administrative law rules, including the Administrative Procedures Act (APA), apply equally to tax regulations and that they matter, if a regulation is to be accorded deference and respected as an appropriate exercise of administrative power.

These conclusions, and particularly this renewed emphasis on the APA, came simultaneously with a growing sense within the tax community that the IRS has concluded that many existing anti-abuse rules do not have sufficient claws to curb putatively aggressive tax planning and that Congress cannot be relied upon to legislate the necessary. As a result, Treasury and the IRS seem to be trying to take more of a legislative role, promulgating regulations that may change and expand existing statutes rather than merely interpreting them and filling in the gaps.⁷ Those regulations are likely to come under increased scrutiny in the coming years, and to the extent Altera is the logical result of Mayo, as explained below, such regulations may stand or fall on whether the dictates of the APA were followed.

According to the APA, in promulgating regulations, the agency must (1) publish notice in the Federal Register, (2) provide interested parties the opportunity to participate in the development of the regulations through written comments, and (3) after consideration of the comments presented, provide a general statement of the basis and purpose of the regulations.⁸ In reviewing an agency’s action, courts will not substitute their own judgment for that of the agency; instead, under Motor Vehicles Association v. State Farm Mutual Automobile Insurance Company, 463 U.S. 29, 43 (1983), courts must ensure that the agency engaged in reasoned decision-making by examining the relevant facts and comments and articulating a satisfactory explanation for its actions, including demonstrating a rational connection between the facts found and the choice made. A court must hold unlawful and set aside agency actions, findings, and conclusions that are arbitrary, capricious, an abuse of discretion, or otherwise not in accordance with the law.⁹

Typically, disputes over the validity of a regulation turn on whether the regulation represents a fair interpretation of the statutory law.¹⁰ The parties argue over the meaning of the particular words of the statute. The applicable Treasury regulations at issue in Altera presented a somewhat unique situation. Section 482 authorizes the Commissioner to allocate income and expenses among related entities to ensure that taxpayers clearly reflect income relating to transactions between related parties. This is a very broad grant of authority, essentially inviting the promulgation of any and all rules that impose on related parties prices comparable to what unrelated parties would have charged: “The standard to be applied in every case is that of a taxpayer dealing at arm’s length with an uncontrolled taxpayer.”¹¹ In 2003, Treasury promulgated final regulations that required related entities that have entered into a cost-sharing agreement to share stock-based compensation costs (e.g., the value of an option on the grant date). What made this unusual was that determining whether unrelated parties in fact share the cost of stock-based compensation is an objective, observable fact, so that a rule such as the one promulgated should only be valid if there is at least a reasonable predicate for Treasury’s conclusion that including such compensation is, in fact, what arm’s-length parties would have done.

Notice of Proposed Rulemaking

Treasury had issued the notice of proposed rulemaking in connection with the regulations that ultimately became Treas. Reg. § 1.482-7(d)(2) in July 2002. The proposed regulations were undergirded by Treasury’s belief that unrelated parties entering into cost-sharing agreements would generally share stock-based compensation costs. However, Treasury did not engage in any fact-finding efforts to confirm its belief, nor did Treasury’s file include empirical evidence supporting that belief. Treasury received twelve comments from industry groups, law firms, accounting firms, taxpayers, and a professor. Several of the comments, many with empirical support, expressed the view that stock-based compensation costs are not shared in cost-sharing relationships between unrelated parties, including the particularly damning evidence that federal acquisition regulations prohibit the practice. Nevertheless, Treasury finalized Treas. Reg. § 1.482-7(d)(2) without changing the requirement that stock-based compensation costs be taken into account in determining the arm’s-length rate. Moreover, the preamble to the final regulations did not refute the information provided by the commenting parties and did not evidence reasoned consideration of their input.

In this unusual situation—where the objective data was contrary to Treasury’s assumption of market practice—Treasury failed to respond to this data, and Treasury failed to offer any affirmative evidence of its own. The Tax Court held in Altera that Treas. Reg. § 1.482-7(d)(2) was not based on reasoned decision-making and was arbitrary and capricious and, thus, invalid. A rule that treats a particular allocation as required to mirror what unrelated parties would do cannot survive when there is no evidence that, in fact, it is what unrelated parties do. Because Treasury seemed to believe that Section 482 gave it authority to do almost anything, and it had been trying to use the regulatory process to overturn a decision with which it disagreed, its failure here was particularly noteworthy.

However, Treasury regulations that are premised upon observable facts and empirical evidence are the exception. As noted above, it is much more common for Treasury regulations to interpret a statute. In that context, the comments Treasury receives will usually focus on legal arguments about the meaning and purpose of words of a statute or practical arguments about why a rule is harmful to the taxpayers that the statute intends to benefit. Treasury presumably should be able to respond with its own reasoned arguments about why its interpretation is reasonable. Thus, the Altera decision may not have as far-reaching implications as one might first believe. Altera is most relevant where Treasury is basing a regulation on a factual premise that is objectively unreasonable and is shown to be so by comments sent to Treasury and where Treasury fails to respond adequately to the comments.

Altera also is important because it shows once again that Treasury action may be invalidated, even in the post-Mayo world, where the dictates of the APA are not followed.

Lessons Learned

Nonetheless, Altera offers several lessons for taxpayers beyond the narrow confines of the decision and is directly applicable to the more common disputes over the validity of regulations that are contested in the current environment.

1. The APA’s Notice-and-Comment Rulemaking Process Matters. The taxpayer might not have prevailed on its motion for summary judgment in Altera without the numerous comments submitted to Treasury that specifically addressed whether businesses include stock-based costs in arm’s-length cost-sharing agreements. To withstand judicial scrutiny, Treasury had to reconcile the facts set out in the comments with the ultimate choice made in the final regulations. Because those objective facts ran counter to Treasury’s presumption of market practice, Treasury was unable or unwilling to do so. This made it easy for the Tax Court to find Treasury’s reasoned decision-making lacking.
   Frequently, few if any comments are submitted in response to proposed regulations, and hearings are often canceled because nobody is interested enough to offer verbal testimony. More attention needs to be placed on submitting thoughtful, specific comments to Treasury on issues that are of interest to the corporate taxpayer community. Comments from a variety of sources, including TEI, are a vital part of influencing the outcome and ensuring that Treasury is promulgating reasoned rules. Even when the arguments offered are legal or practical, as opposed to factual, if Treasury fails to consider and address them in a cogent fashion, a court may decide that the requirements of notice-and-comment rulemaking have not been met.
2. Settlement Short of Litigation Based on Invalid Regulations May Be Impossible. In the tax controversy context, Exam and Appeals will rarely, if ever, settle based on the taxpayer’s arguments that regulations are invalid. As a result, if the taxpayer’s position depends upon arguing that Treasury regulations are invalid, the taxpayer should assume that the issue will likely go to litigation, at least through summary judgment, and, if successful at the trial court, the taxpayer will likely face an appeal by the government. Nonetheless, you should present your arguments regarding a regulation’s invalidity at the administrative level, along with other arguments to support your position. Experience indicates, for example, that where you convince an Appeals officer that a Treasury regulation is invalid, the Appeals officer may resolve the case more favorably for you based on your other arguments, with the invalidity argument placing an invisible thumb on the scale. Given this, you should carefully examine the comments submitted and the way the final regulations address them so you can lead the Appeals officer to think not only about the merits of your challenge to the substance of the regulations but also about whether the regulations might be infirm for Treasury’s failing to consider and address responsible comments.
   In all events, if a key element of your case depends on the invalidity of a Treasury regulation, you should begin early to preserve and organize your own working files and the evidence necessary for trial. In Altera, it was crucial to show the Tax Court the extensive nature of the comments that were submitted to Treasury in response to the proposed regulations. As soon as you decide that you might take a position contrary to regulations, in addition to collecting publicly available materials, submit your Freedom of Information Act request for a copy of the administrative file for the regulatory project at issue. Despite the statutory timeline for FOIA responses, the IRS can be very slow to produce these files, and an analysis of the files will be crucial to assess your likelihood of prevailing in the assertion that Treasury has not engaged in the requisite reasoned decision-making. This may help you at Appeals and will be crucial in court.
   In dealing with Exam and Appeals, keep in mind that the dispute will likely go to trial. Involve outside counsel early in the process to preserve privileges and manage the production of documents and other evidence.
3. The Regulatory Context Matters. To some extent, particularly with respect to rulemaking that is done through temporary regulations or even Revenue Rulings, Treasury may try to cure any deficiency in its reasoned decision process through subsequent actions. Home Concrete¹² is an example of a situation where Treasury promulgated final regulations after losing in the trial court. At the appellate court, the government argued that deference should be given to the final regulations, even though those regulations were obviously promulgated in reaction to the trial court loss.¹³ An attack on whether agency action satisfies the APA may only produce a pyrrhic victory outside the context of final regulations. Make sure to preserve your alternative arguments for such an event.
4. Challenging the Regulation in Court Is Always Possible. Altera does not stand for the proposition that in the absence of comments, one cannot challenge the factual predicate for a regulation. If a regulation, such as the one at issue, turns on what unrelated parties would do, it would seem that, even in the absence of comments during the rulemaking process, a litigant could seek to demonstrate that the factual predicate for a regulation that depends on such a foundation is flawed. Assume that in Altera no comments had been filed. The taxpayer could still have tried to show (through an expensive trial) that unrelated parties would not consider stock-based compensation and that the regulation was thus invalid. But that is more difficult where the APA was followed.
5. Be Prepared for Harmless Error Arguments. If an older regulation promulgated before Mayo is at issue, Treasury is more likely to have taken various shortcuts through the APA. Harmless errors do not produce invalid regulations.¹⁴ Consider whether Treasury shortcuts are mere harmless error or whether the error would affect the outcome. In the context of objective facts on market practice, Altera was an example of where the error in reasoned decision-making, if corrected, would have produced a different outcome, and hence that shortcoming led the Tax Court to invalidate the regulations. In many other, less objective regulatory contexts, it will be easier for the government to argue that any failure to comply with the APA was harmless.
   In summary, Altera is an important decision in the context of cost-sharing agreements and because it makes it very clear that Treasury must consider and critically evaluate comments in order for a tax regulation to be entitled to substantial deference. It also is important because it shows once again that Treasury action may be invalidated, even in the post-Mayo world, where the dictates of the APA are not followed. But it does not augur a new era of heightened scrutiny of the reasonableness of agency interpretations of legal provisions and is largely a very fact-driven evaluation of a particular agency action.

Matthew Lerner and Kevin Pryor are partners at Sidley Austin LLP.

Endnotes

1. Judges Morrison and Pugh did not participate in the decision.
2. See Xilinx Incorporated v. Commissioner, 567 F.3d 482 (9th Cir. 2009), rev’g 125 T.C. 37 (2005), withdrawn, 592 F.3d 1017 (2010).
3. Mayo Found. for Med. Educ. & Research v. United States, 562 U.S. 44, 53 (2011).
4. National Muffler Dealers Assn. Inc. v. United States, 440 U.S. 472 (1979).
5. Chevron U.S. Inc. v. NRDC Inc., 437 U.S. 837 (1984).
6. Household Credit Services Incorporated v. Pfennig, 541 U.S. 232, 242 (2004) (quoting United States v. Mead Corporation, 533 U.S. 218, 227 (2001)).
7. See, for example, Treas. Reg. § 1.7874-4T, which purports to determine that the rule that stock of the foreign acquiring corporation purchased for cash in a public offering can be ignored in computing Section 7874’s ownership fraction applies to stock purchased in a private placement, notwithstanding the different words in the statute and the fact that the provision in the temporary regulation was proposed and rejected in the Senate on three separate occasions.
8. See 5 U.S.C. § 553.
9. 5 U.S.C. § 706(2)(A).
10. For example, in Mayo the question was what the appropriate definition of a full-time employee is.
11. Treas. Reg. § 1.482-1(b)(1).
12. United States v. Home Concrete and Supply LLC, 132 S.Ct. 1836 (2012).
13. Ultimately, this effort failed based on the Supreme Court’s interpretation of the import of prior judicial determinations in evaluating the validity of regulations, so that the level of deference applied to after-enacted regulations was not decided.
14. 5 U.S.C. § 706.