Second Circuit to Resolve Disagreement Between Tax Court and Federal Circuit Over Availability of Interest Netting for Pre-1998 Tax Periods
April 24, 2012
[Note: Miller & Chevalier represents the taxpayer Exxon Mobil Corp. in this case.]
The government has appealed the Tax Court’s decision on interest netting in Exxon Mobil Corp. v. Commissioner, 136 T.C. No. 5 (Feb. 3, 2011). The briefing is now complete, and the Second Circuit (Judges Cabranes, Walker, and Winter) is scheduled to hear oral argument on April 25.
Congress expressly required global interest netting by enacting Code section 6621(d) in 1998. Before then, the IRS had sometimes taken advantage of the differential interest rates established in 1986 to collect net interest when no net tax was due. Specifically, for periods of overlapping indebtedness where the taxpayer owed money to the government for one tax year but the government owed money to the taxpayer for a different tax year, the government would not net the overlapping amounts, but rather would collect interest at the higher underpayment rate while paying interest to the taxpayer at the lower overpayment rate on the same amount. Section 6621(d) ended this practice by providing that “the net rate of interest . . . on such amounts” of overlapping indebtedness “shall be zero.” That net rate of zero can be implemented either by increasing the overpayment rate or decreasing the underpayment rate from what they would otherwise be in the absence of interest netting. Section 6621(d) applies prospectively without restriction, but Congress also enacted a “special rule” that makes the interest netting rule also apply to pre-1998 periods of overlapping indebtedness in certain circumstances. The Exxon Mobil case involves the construction of that special rule.
The special rule provides that section 6621(d) applies to past periods “[s]ubject to any applicable statute of limitation not having expired with respect to either a tax underpayment or a tax overpayment” on the July 22, 1998, effective date of section 6621(d). Taxpayers interpret this provision to mean that interest netting is available for past periods so long as the statute of limitations for either of the overlapping periods of indebtedness was open on the effective date. The IRS, however, has interpreted the special rule to mean that interest netting is available for past periods only if the statute of limitations for both of the overlapping periods of indebtedness was open on the effective date.
The issue was first litigated by Fannie Mae in the Court of Federal Claims, which held that the special rule was best read in accordance with the taxpayer’s position. The Federal Circuit, however, reversed. Federal National Mortgage Ass’n v. United States, 379 F.3d 1303 (Fed. Cir. 2004), rev’g, 56 Fed. Cl. 228 (2003). The Federal Circuit did not take issue with the statutory analysis of the Court of Federal Claims, but instead relied on an argument that was not presented below. The Federal Circuit reasoned that the special rule was a waiver of sovereign immunity that must be strictly construed in favor of the government; since the statutory language was capable of being read to support either position, the court concluded, that strict construction principle required a ruling in the government’s favor.
The Exxon Mobil case raises the same issue, but it arises in a different procedural posture and hence in a different jurisdiction. Exxon petitioned the Tax Court for a redetermination of deficiencies for its 1979-1982 tax years, and the Tax Court’s determination eventually resulted in an overpayment. The IRS paid interest on that overpayment at the regular overpayment rate, unadjusted for the fact that interest netting principles would have called for a higher rate because there was a period of overlapping indebtedness with underpayments that Exxon owed for the 1975-78 tax years. Exxon Mobil then filed a motion for redetermination of interest under Code section 7481 seeking additional overpayment interest through application of the interest netting rule of section 6621(d).
The IRS argued that interest netting did not apply because the statutes of limitations for the 1975-78 tax years were no longer open on July 22, 1998, although it acknowledged that the statutes of limitations for the overpayment leg of the overlap period were open on that date. The Tax Court declined to follow the Federal Circuit’s view and rejected the IRS’s argument. The Tax Court stated that “the special rule is not a waiver of sovereign immunity but an interest rate provision.” It added that, even if it were a waiver of sovereign immunity, that would not require the court to adopt a reading that contravened Congress’s intent to achieve the remedial purpose of relieving taxpayers from paying interest when no net tax was due.
The government’s brief on appeal relies heavily on the Federal Circuit’s decision in Fannie Mae. It argues that the Tax Court erred in rejecting the basic principle of that decision, asserting that “the special rule is a waiver of sovereign immunity because it authorizes recovery of certain retroactive refund claims for overpaid interest and thus ‘discriminates between those claims for overpaid interest Congress has authorized and those it has not’” (quoting Fannie Mae, 379 F.3d at 1310). Exxon Mobil’s brief first addresses the special rule independently, asserting that the natural reading of the text and the evident purpose of the special rule both indicate that Congress intended to allow interest netting for pre-1998 periods so long as one leg of the overpayment period was open. The brief then addresses the government’s sovereign immunity argument, arguing that the special rule is not a waiver of sovereign immunity and, even if it were, the canon of construction would not justify adopting an interpretation that would thwart the evident intent of Congress.
April 3, 2012
As we previously reported, Day 1 of last week’s oral argument in the Supreme Court on the challenges to the health care legislation focused on whether the Anti-Injunction Act bars the lawsuits. The excitement about the argument on that issue was largely gone as soon as it was over, because it was fairly apparent that the Court will not find the Act to be an obstacle to reaching the merits of the health care dispute. Indeed, Robert Long, the lawyer who argued as amicus for that position, has predicted that he will not get a single vote. Certainly the argument was almost completely forgotten by the next day when the Court’s questioning on the constitutionality of the individual mandate led many observers to conclude that the mandate will be invalidated. (For the record, my opinion is that the health care legislation will survive, but that topic is beyond the scope of this blog.) Still, the Court’s decision to reject the applicability of the Anti-Injunction Act could have precedential significance, depending on the rationale that the Justices use. Therefore, we briefly recount the argument here with that issue in mind.
There were two basic arguments made for holding that the health care lawsuits could proceed despite the Anti-Injunction Act. The primary argument was that the Act by its terms did not apply – that is, that for a variety of reasons the “penalty” for failing to obtain health insurance is not a “tax” within the meaning of the Anti-Injunction Act. Both the challengers to the health care legislation and the United States took this position. Preliminary to this statutory interpretation question, however, was the argument that the Anti-Injunction Act should not apply in this case — even if the health care penalty would ordinarily come within the ambit of the Anti-Injunction Act — because the government had waived the defense and urged that the lawsuits should proceed. The validity of that waiver argument turns on whether the Anti-Injunction Act is “jurisdictional,” meaning that it addresses a court’s jurisdiction or power to hear a case, as opposed to being a “claim processing rule.”
A court does not have the authority to create its own jurisdiction, even if both parties want it to hear the case. Thus, if a statute is “jurisdictional,” a court is obliged to examine jurisdiction on its own and to dismiss the case if it finds that the statutory conditions are not met. Conversely, if a statutory condition is not jurisdictional, then a party can waive satisfaction of that condition and the court can proceed to hear the case. (For example, many exhaustion requirements or statutes of limitations are not jurisdictional, see Reed Elsevier, Inc. v. Muchnick, 130 S. Ct. 1237 (2010); Day v. McDonough, 547 U.S. 198, 205-06 (2006)). In the health care litigation, although the United States wanted the lawsuits to proceed, it took the position that the Anti-Injunction Act is “jurisdictional” and therefore (in contrast to the challengers to the law) argued that the Court could proceed to hear the case only if it concluded that the Anti-Injunction Act by its terms did not apply to the “penalty” for failing to obtain insurance.
If the Court were to resolve the case on waiver grounds, concluding that the Anti-Injunction Act is not jurisdictional, that could create opportunities for taxpayers in future cases if a government attorney overlooks an Anti-Injunction Act defense. It also would give the government flexibility to assert the defense when it wants, but to allow cases like the health care challenge to go forward if the government determines that it wants a prompt answer. The government, however, is concerned about a holding that the Anti-Injunction Act is not jurisdictional, because courts are freer to adopt equitable exceptions to non-jurisdictional statutes.
At the oral argument, the Justices explored both possible grounds for resolving the issue. Chief Justice Roberts and Justice Alito seemed the most interested in concluding that the Act is not jurisdictional and thus giving effect to the government’s waiver. Roberts described as the “biggest hurdle” to the Anti-Injunction Act argument a 1938 case in which the Court had gone ahead and decided an issue apparently barred by the Act after the government had waived its defense. When counsel responded that the case was no longer good law and pointed out that the Court had since repeatedly referred to the Act as “jurisdictional,” Alito forced him to concede that the Court had never actually held that the statute was “jurisdictional” in a case where that characterization would make a difference. Justices Ginsburg, Kagan, and Sotomayor all joined in that line of questioning, pointing to similarly worded statutes or precedents that in Justice Sotomayor’s words indicated that “Congress has accepted that in the extraordinary case we will hear the case.”
As the argument progressed, however, it appeared less likely that a majority would coalesce around this position. Justice Breyer volunteered that he was inclined to agree that the Anti-Injunction Act is jurisdictional, but that he doubted it applied to the health care legislation. Sotomayor indicated that she thought this position, which is what was being espoused by the government, was the least problematic. Justice Ginsburg suggested that she sided with the position that the Act did not apply, observing that this conclusion would make it unnecessary to resolve the thornier “jurisdictional” question.
Although it is always hazardous to predict outcomes based on questions asked at oral argument, the most likely outcome appears to be that the majority of Justices will address the merits of the Anti-Injunction Act issue, rather than relying on the government’s waiver of the defense. If so, the decision will not foreclose courts in the future from applying the Anti-Injunction Act when the government has failed to raise the defense or deliberately chosen not to raise it.
A decision is expected in the last week of June, perhaps June 28, and will surely be overshadowed by the Court’s contemporaneous decision on the constitutionality of the health care legislation.
April 2, 2012
We present here a guest post by our colleague Patricia Sweeney:
On March 29, 2012, the Second Circuit heard oral arguments in Union Carbide Corp. & Subsidiaries v. Commissioner, No. 11-2552 (Judges Straub, Pooler & (visiting district court Judge) Korman on the panel ). The case involves Union Carbide Corp’s (“UCC’s”) claim for research and experimentation credits with respect to 106 research projects. In order to resolve the issues expeditiously, the parties agreed that the Tax Court would limit trial to the five largest projects. All five projects involved process experimentation after products were placed in commercial operation. For two of the five projects, the Tax Court held that the projects involved qualified research, but it determined that the costs of materials incurred to produce the commercial product necessary for UCC to conduct the process research were not eligible for the credit. UCC has appealed this holding, as well as the Tax Court’s separate conclusion that one of the other projects did not involve a process of experimentation because UCC did not conduct additional post-test analysis after it determined that the test was successful. The Second Circuit’s decision with respect to the costs of materials issue could have widespread implications for taxpayers claiming the research credit.
Background Regarding the R&E Credit. To be eligible for the research credit under section 41(a)(1) of the Code, a taxpayer must show that it has performed “qualified research.” To be qualified research: (1) the research must be eligible as research under section 174 of the Code; (2) the research must be undertaken for the purpose of discovering technological information; (3) the taxpayer must intend for the discovered information to be useful in the development of a new or improved business component; and (4) substantially all of the research activities must constitute elements of a process of experimentation. These tests are applied separately to each “business component.” Section 41(d)(2)(C) provides that the development of an improved process is a separate business component from the product being produced. At the same time, the Code generally provides in section 41(d)(4)(A) that research conducted after the beginning of commercial production of the business component is not eligible for the research credit. Because the products in issue were all in commercial production, the “improved process” rule is critical to UCC’s position to avoid the expenses from being declared ineligible for the research credit under section 41(d)(4)(A).
The research credit is computed by reference to qualified research expenses (“QREs”), including “any amount paid or incurred for supplies used in the conduct of qualified research . . .” other than land, improvements to land, and depreciable property. I.R.C. § 41(b)(2)(A). In a post-Union Carbide decision, TG Missouri v. Commissioner, 133 T.C. 278 (2009), the IRS did not challenge, and the Tax Court allowed without discussion, the inclusion in the QREs of material costs for production molds sold to clients. In Trinity Industries v. United States, 691 F.Supp. 2d 688, 697 (2010), the U.S. District Court for the Northern District of Texas expressly allowed the material costs incurred to construct ships because it concluded that the costs “are properly considered research expenditures in that the business component – the ship – could not have been developed without them.”
The Parties’ Briefs with Respect to the Supplies Argument. As previously noted, although the Tax Court agreed that 2 of the 5 projects in issue were qualified research and that the experiments could not have occurred without the supplies claimed by UCC (a process cannot be tested unless it is applied to the production of the product), it concluded that the supplies were not “used in the conduct of” the process of experimentation with respect to process development. Instead, the court allocated the supply costs to the product business component and disallowed all the supplies claimed. In essence, the court established a primary purpose test, and concluded that the primary purpose of the expenditures was to produce the commercial product, not to conduct research. In this context, the Tax Court noted that the supply costs in issue were “at best, indirect research expenditures excluded from the definition of QREs.”
In its briefs, UCC argues that the Tax Court has created an inappropriate distinction between product and process experimentation. It maintains that the Tax Court erred when it created two new requirements, i.e., that in addition to being used in the conduct of research, the costs must have been “primarily” incurred as a result of the process experimentation, and that supplies incurred both to produce a product and to be used in process experimentation are not primarily incurred in process experimentation. There is nothing in the Code, regulations or legislative history to indicate that the qualification of supplies as QREs turns on whether they are incurred with respect to process or product experimentation. Moreover, there is nothing in the Code, regulations or legislative history that either creates a hierarchy that would allocate the costs solely to the production of the product or disallows such costs as “indirect costs.” UCC argues that the Second Circuit should apply the plain meaning of the statute, which provides that costs “used in the conduct of qualified research” qualify for the credit. There is no dispute that UCC could conduct its process experimentation only if it produced the product in issue. As a result, UCC maintains that the material costs were necessary for the process experimentation, were at risk in that experimentation, and therefore, must be included in the QREs.
As noted by UCC, the government in its brief did not attempt to reconcile the Tax Court’s decisions in UCC and TG Missouri, or the Trinity decision. It distinguished these other two decisions as involving product rather than process experimentation, stating “in each of these cases . . . the supply costs found to be eligible for the credit were the direct, incremental cost of performing the qualified research.” The focus of the government’s argument in its brief is that the costs of supplies to produce goods for sale are “indirect research expenses,” which are excluded from the definition of QREs under Treas. Reg. § 1.41-2(b)(2) because they are expenses that would have been incurred regardless of any research activities. The government argues that Congress intended that only the incremental costs arising from the research are sufficiently “direct” to qualify as QREs. For support, it cites other limitations restricting the availability of the credit – such as the limitations on qualified research to a discrete business component or to pre-commercial production. It further asserts that it is unreasonable to claim supply costs that otherwise would have been incurred in commercial production.
The Potential Impact. As noted in an amicus brief filed by the National Association of Manufacturers, the American Chemistry Council, and the Chamber of Commerce of the United States, the Second Circuit decision could have a wide impact on U.S. manufacturers. If the Tax Court’s decision is sustained, the incentive that Congress intended by allowing the research credit could be significantly curtailed. Moreover, as pointed out by both the amicus brief and UCC, such a decision would ignore the significant risks associated with process research and create an inappropriate distinction between product and process research. However, there could be an even worse result for taxpayers than an affirmance of the Tax Court decision. As noted by UCC in its briefs, the government’s argument that production costs are not QREs could be extended to product research. If the Second Circuit accepted that extension, it could conclude that only incremental production costs qualify as QREs, a decision that would have much broader ramifications and that would be at odds with the decisions in both TG Missouri and Trinity. Time will tell.
We will return soon with a report on the oral argument.