DOMINION RESOURCES, INC. v. U.S., Cite as 109 AFTR 2d 2012-XXXX, 05/31/2012
DOMINION RESOURCES, INC., Plaintiff-Appellant, v. UNITED STATES, Defendant-Appellee.
Court Name: United States Court of Appeals for the Federal Circuit,
Docket No.: 2011-5087,
Date Decided: 05/31/2012.
Eric R. Fox, Ivins, Phillips & Barker, Chartered, of Washington, DC, argued for plaintiff-appellant. With him on the brief were Leslie J. Schneider and Patrick J. Smith.
Karen G. Gregory, Attorney, Tax Division, United States Department of Justice, of Washington, DC, argued for defendant-appellee. With her on the brief were Tamara W. Ashford, Deputy Assistant Attorney General, and Jonathan S. Cohen, Attorney.
Paul L. Gale, Troutman Sanders LLP, of Washington, DC, for amicus curiae.
United States Court of Appeals for the Federal Circuit,
Appeal from the United States Court of Federal Claims in Case No. 08-CV-195, Judge Charles F. Lettow
Before Rader, Chief Judge, Clevenger, and Reyna, Circuit Judges.
Opinion for the court filed by Chief Judge Rader. Opinion concurring in part and concurring in the result filed by Circuit Judge Clevenger.
Judge: Rader, Chief Judge.
The United States Court of Federal Claims granted the United States' motion for summary judgment against Dominion Resources, Inc. See Dominion Res., Inc. v. United States, 97 Fed. Cl. 239 [107 AFTR 2d 2011-1033] (2011). This case presents an issue of first impression for any appellate court. The CFC held that Treasury Regulation § 1.263A-11(e)(1)(ii)(B) is a permissible construction of the statute I.R.C. § 263A. Because the associated property rule in Treasury Regulation § 1.263A-11(e)(1)(ii)(B) as applied to property temporarily withdrawn from service is not a reasonable interpretation of I.R.C. § 263A and because the Treasury acted contrary to 5 U.S.C. § 706(2) in failing to satisfy the State Farm requirement to provide a reasoned explanation when it promulgated that regulation, this court reverses.
Dominion provides electric power and natural gas to individuals and businesses. In 1996, it replaced coal burners in two of its plants. When making those improvements, it temporarily removed the units from service — one unit for two months, the other for three months. During that time, Dominion incurred interest on debt unrelated to the improvements.
On its corporate tax returns, Dominion deducted some of that interest from its taxable income. The IRS disagreed with Dominion's computation under Treasury Regulation § 1.263A-11(e)(1)(ii)(B), the regulation at issue here. The IRS applied the regulation to capitalize $3.3 million of that interest, instead of deduct. A deduction occurs immediately in that tax year, while capitalization occurs over later years. Under a settlement, the IRS allowed Dominion to deduct 50% and capitalize 50% of the disputed amount.
Still asserting that the entire disputed amount is deductible, Dominion filed this suit seeking a refund of $297,699 in corporate income tax. Dominion thus sought to invalidate Treasury Regulation § 1.263A-11(e)(1)(ii)(B). The CFC denied Dominion's claim and granted summary judgment to the United States. The CFC held that the regulation was a permissible construction of I.R.C. § 263A and that Treasury promulgated that regulation with a reasoned explanation that satisfied 5 U.S.C. § 706(2) and Motor Vehicles Mfrs. Ass'n of the United States, Inc. v. State Farm Mut. Auto. Ins. Co., 463 U.S. 29 (1983).
The Tax Reform Act of 1986 enacted I.R.C. § 263A (“Capitalization and Inclusion in Inventory Costs of Certain Expenses”). Generally, the statute requires capitalization of certain costs incurred in improving real property, instead of deduction. In broad terms, interest appears as a cost covered by the capitalization requirement.
The relevant statutory provisions in I.R.C. § 263A comprise five subsections. Each subsection refers to the next. A careful reading of the five subsections shows that each rule or definition refers to another rule or definition in a circular progression that brings the law back to the place it began with little elucidation of legal standards and definitions. In simple words, the statute is circular.
First, I.R.C. § 263A(a)(1) (“Nondeductibility of Certain Direct and Indirect Costs”) sets out the general rule that when improving real property, certain costs must be capitalized instead of deducted from taxable income. “In the case of any property to which this section applies, any costs described in paragraph (2) ... shall be capitalized.” I.R.C. § 263A(a)(1). The text of that statutory provision refers to subsection paragraph (2) (“Allocable Costs”), which defines such costs as including both “direct costs” and “indirect costs.” “The costs described in this paragraph with respect to any property are (A) the direct costs of such property, and (B) such property's proper share of those indirect costs (including taxes) part or all of which are allocable to such property.” I.R.C. § 263A(a)(2) (emphasis added).
Next, interest is a type of “cost” as discussed in subsection (f) (“Special Rules for Allocation of Interest to Property Produced by the Taxpayer”). Specifically, subsection (f)(1) states the general rule that interest is a cost requiring capitalization when that cost is “allocable” to the property. “Subsection (a) shall only apply to interest costs which are (A) paid or incurred during the production period, and (B) allocable to property which is described in subsection (b)(1) and which has [other requirements not relevant here].” I.R.C. § 263A(f)(1) (emphasis added).
To determine what interest costs are “allocable” as mentioned in subsection (f)(1), subsection (f)(2) (“Allocation Rules”) states the general rule that interest is allocable “to the extent that the taxpayer's interest costs could have been reduced if production expenditures ... had not been incurred.” “In determining the amount of interest required to be capitalized under subsection (a) with respect to any property ... (ii) interest on any other indebtedness shall be assigned to such property to the extent that the taxpayer's interest costs could have been reduced if production expenditures (not attributable to indebtedness described in clause (i)) had not been incurred.” I.R.C. § 263A(f)(2) (emphasis added). This is the avoided-cost rule and implements Congress' concern with the avoided-cost principle:
The legislative history of amendments to section 189 indicates Congress' intention that the Treasury Department issue regulations allocating interest to expenditures for real property during construction consistent with the method prescribed by Financial Accounting Standards Board Statement Number 34 (FAS 34). Under FAS 34, the amount of interest to be capitalized is the portion of the total interest expense incurred during the construction period that could have been avoided if funds had not been expended for construction. Interest expense that could have been avoided includes interest costs incurred by reason of additional borrowings to finance construction, and interest costs incurred by reason of borrowings that could have been repaid with funds expended for construction.
S. Rep. No. 99-313, at 140, 144 (1986) (emphasis added); H.R. Rep. No. 99-426, at 625, 628 (1985) (same); Joint Committee on Taxation, JCS-10-87, 1987 WL 1364655 (1987) (same).
The term “production expenditures” mentioned in subsection (f)(2) is defined in subsection (f)(4)(C) to mean costs “required to be capitalized under subsection (a).” “The term “production expenditures” means the costs (whether or not incurred during the production period) required to be capitalized under subsection (a) with respect to the property.” I.R.C. § 263A(f)(4)(C). The text of that statutory provision refers to “subsection (a),” which means I.R.C. § 263A(a), the first statutory provision that began this discussion on the relevant statutory provisions.
The general formula to determine the amount of interest that must be capitalized is the amount of “production expenditures” multiplied by the weighted-average interest rate on the debt during the time the production occurs. In other words, the production expenditures represent the base amount and some fraction of that amount represents the interest that must be capitalized. A larger base will lead to more interest capitalized.
After a notice of proposed rulemaking in 1991, the Treasury published final regulations in 1994. The regulation at issue here defines what constitutes “production expenditures” (the base amount) and therefore determines the amount of interest capitalized. Treasury Regulation § 1.263A-11(e)(1) (“General Rule”) states:
If an improvement constitutes the production of designated property under § 1.263A-8(d)(3), accumulated production expenditures with respect to the improvement consist of (i) All direct and indirect costs required to be capitalized with respect to the improvement, (ii) In the case of an improvement to a unit of real property (A) An allocable portion of the cost of land, and (B) For any measurement period, the adjusted basis of any existing structure, common feature, or other property that is not placed in service or must be temporarily withdrawn from service to complete the improvement (associated property) during any part of the measurement period if the associated property directly benefits the property being improved, the associated property directly benefits from the improvement, or the improvement was incurred by reason of the associated property.
The parties agree that a certain amount of construction-period interest should be capitalized instead of deducted, but the extent of that capitalization requirement is the essence of this dispute. The parties agree that the Treasury regulation plainly defines production expenditures to include not only the amount spent on the improvement but also the adjusted basis of the entire unit being improved. For simplicity, adjusted basis can be considered as the original cost of the unit. The issue on appeal is whether that latter inclusion of the adjusted basis of the unit violates various statutory provisions. Because the regulation requires a larger base amount (by including the adjusted basis amount), it results in a larger amount of interest to be capitalized. Thus, the practical impact determines how much interest Dominion must capitalize instead of deduct from its taxable income as a result of burner improvements in its power plants.
The challenge to the regulation is only as applied to property “temporarily withdrawn from service” and not as applied to property that “is not placed in service.”
This court has jurisdiction under 28 U.S.C. § 1295(a)(3).
This court reviews the CFC's grant of summary judgment without deference. The validity of a Treasury regulation is analyzed under the Chevron two-step test. First, step one determines whether Congress has directly spoken to the precise question at issue. Chevron, U.S.A., Inc. v. NRDC, 467 U.S. 837, 842–43 (1984). If the statute is silent or ambiguous, then step two determines whether the agency's answer is based on a permissible construction of the statute.Id.
Last year, the Supreme Court confirmed that courts applyChevron deference to Treasury regulations.See Mayo Found. for Med. Educ. and Research v. United States , 131 S.Ct. 704, 711 [107 AFTR 2d 2011-341]–13 (2011).
As to Chevron step one, the CFC correctly recognized that the regulation does not contradict the text of the statute but only because the statute is opaque. See Dominion Res., 97 Fed. Cl. at 253. Subsection (f)(2) states the amount of interest to be capitalized is that amount that “could have been reduced if production expenditures ... had not been incurred.” Then subsection (f)(4)(C) defines “production expenditures” as the amount required according to the general rules. Regardless of the definition of “production expenditures,” the statute provides or assumes that sum would have been available to pay down the debt. The conclusion has been assumed in the premises, and therefore the statute is circular. As demonstrated in the discussion above, each rule or definition refers to another rule or definition in a progression such that the reader ends at the beginning. Subsection (a)(1) refers to (a)(2) refers to (f)(1) refers to (f)(2) refers to (f)(4)(C) refers back to (a). Thus, atChevron step one, this court determines that the statute is ambiguous. In such an instance, this court detects noChevron step one violation.
As to Chevron step two, however, Treasury Regulation § 1.263A-11(e)(1)(ii)(B) as applied to property temporarily withdrawn from service is not a reasonable interpretation of the avoided-cost rule set out in the statute at I.R.C. § 263A(f)(2)(A)(ii). Specifically, the regulation is unreasonable in defining “production expenditures” to include the adjusted basis of the entire unit.
The regulation directly contradicts the avoided-cost rule that Congress intended the statute to implement. The avoided-cost rule recognizes that if the improvement had not been made, those funds (an amount x equal to the cost of the improvement) could have been used to pay down the debt and therefore reduce interest that accrued on the debt. Because the improvement was made, however, that amount x was not used to pay down the debt; therefore, interest accrued on that amount x.
To determine the accrued interest resulting from making the improvement instead of paying down the debt, one would multiply the interest rate by the amount x paid for the improvement. One would not multiply the interest rate by the amount x paid for the improvement plus the adjusted basis of the entire unit. An amount equal to the adjusted basis of the unit would not have been available to pay down the debt had the improvement not been made. Those funds were expended at the time the property was acquired (before the decision to make the improvement) — and are not made available to pay down debt by forgoing the improvement.
For example, let's say an owner purchased real property for $100,000 by a loan with a 3% interest rate. A few years later, she made an improvement that cost $5,000. If she had used that $5,000 toward the debt instead of the improvement, she would have avoided accruing $150 in interest ($5,000 multiplied by 3%). The avoided-cost rule requires her to capitalize that $150 in interest. The Treasury regulation, however, requires her to capitalize $3,150 in interest ($100,000 + $5,000 then multiplied by 3%). That result makes no sense, because there is no way that she could have avoided accruing $3,150 in interest by not making the improvement, as she did not expend or incur an amount equal to $105,000 when making the improvement.
The House and Senate reports clarify the meaning of the statute. The regulation must implement the avoided-cost principle — in particular that the interest to be capitalized is the amount “that could have been avoided if funds had not been expended for construction.” S. Rep. No. 99-313, at 140, 144 (1986); H.R. Rep. No. 99-426, at 625, 628 (1985); Joint Committee on Taxation, JCS-10-87, 1987 WL 1364655 (1987). The adjusted basis does not represent such an “avoided” amount. A property owner does not expend funds in an amount equal to the adjusted basis when making the improvement. Instead, she expends funds in an amount equal to the cost of the improvement itself.
Indeed, the statute uses the term production “expenditures,” the plain meaning of which is an amount actually expended or spent — specifically, expended or spent on the improvement. Similarly, the statute states that the interest to be capitalized is an amount that could have been reduced if production expenditures had not been “incurred.” A property owner would not expend orincur an amount equal to the adjusted basis when making the improvement. Thus, the regulation unreasonably links the interest capitalized when making an improvement to the adjusted basis.
Further, the Treasury regulation leads to absurd results. Because the adjusted basis amount can have almost no relation to the improvement cost amount, the regulation can require capitalizing vastly different amounts of interest for the same improvements. Here, Dominion's two improvements had similar costs of $5.3 million and $6.7 million. Yet, because the adjusted bases of the two units are drastically different, the regulation leads to “production expenditures” of $15 million and $138 million, respectively. This court detects no reasonable basis for requiring such wildly disproportionate results for similar improvements. The law did not intend such an absurd result.
The only way that an amount equal to the adjusted basis could potentially satisfy the avoided-cost method is by assuming that the property owner would have sold the unit and used the sale proceeds to pay down the debt. The CFC correctly recognized that the United States' argument was “removed from reality because the associated property is being improved to continue in service.” Dominion Res., 97 Fed. Cl. at 258. In particular as to this industry, the United States' argument “is, of course, a fiction, given that the generating units in question are large, immobile parts of large power plants which could not realistically be sold individually.”Id. at 253. The CFC erred by concluding that the United States' fiction was a “policy choice” by the agency and thus permissible. See id. at 258. This court discerns no reasonable explanation that assumes that a property owner would have sold the same unit that it removed from service for the sole purpose of improving. Selling the unit obviates the very reason for the improvement. As discussed above, the Treasury regulation contradicts the avoided-cost rule that the law implemented. Thus, the regulation is not a reasonable interpretation of the statute.
Therefore, the associated-property rule in Treasury Regulation § 1.263A-11(e)(1)(ii)(B) as applied to property temporarily withdrawn from service is not a reasonable interpretation of I.R.C. § 263A(f)(2)(A)(ii) and is invalid.
The associated-property rule in Treasury Regulation § 1.263A-11(e)(1)(ii)(B) as applied to property temporarily withdrawn from service also violates the State Farm requirement that Treasury provide a reasoned explanation for adopting a regulation. State Farm requires that the Treasury “articulate a satisfactory explanation for its action, including a rational connection between the facts found and the choice made.” Motor Vehicles Mfrs. Ass'n of the United States, Inc. v. State Farm Mut. Auto. Ins. Co., 463 U.S. 29, 43 (1983). “We have frequently reiterated that an agency must cogently explain why it has exercised its discretion in a given manner.” Id. at 48. An agency rule is arbitrary and capricious if it “entirely failed to consider an important aspect of the problem, offered an explanation for its decision that runs counter to the evidence before the agency, or is so implausible that it could not be ascribed to a difference in view or the product of agency expertise.” Id. at 43.
In Notice 88-99, the IRS provided guidance on the upcoming regulations. 1988-36 I.R.B. 29, 1988-2 C.B. 422 (August 16, 1988). Nowhere did this guidance mention adjusted basis as part of the interest-capitalization method. In the notice of proposed rulemaking, however, the IRS first made mention that it would include adjusted basis in the calculation. See Capitalization of Interest, 56 Fed. Reg. 40,815-01 (proposed Aug. 16, 1991). Still, this notice provided no rationale other than the general statement that the regulations are intended to implement the avoided-cost method. See Dominion Res., 97 Fed. Cl. at 255 (“Although the preamble explains how to apply the associated-property rule, it does not give a reason for its inclusion.”), 258–59 (“Sections of Treasury's proposed regulation explained the avoided-cost method without reference to improvements to existing property.”). Similarly, the IRS provided no rationale in the final regulations. See Capitalization of Interest, 59 Fed. Reg. 67,187 (Dec. 29, 1994).
The Court of Federal Claims recognized that “it is a stretch to conclude that Treasury cogently explained why it has exercised its discretion in a given manner respecting capitalization of interest under the associated-property rule.” Dominion Res., 97 Fed. Cl. at 259. Yet, the trial court erroneously stretched to conclude that “the path that Treasury was taking in the rulemaking proceedings can be discerned, albeit somewhat murkily.” Id.
The notice provides no explanation for the way that use of an adjusted basis implements the avoided-cost rule. Indeed, it does not satisfy the avoided-cost rule. Though the CFC noted that “Treasury did, however, at least alert interested potential commentators that treatment of the basis of the property being improved was at issue insofar as capitalization was concerned,” id. at 259, this is not sufficient to satisfy the State Farm requirement that the regulation must articulate a satisfactory or cogent explanation.
Judge: Clevenger, Circuit Judge, concurring in part and concurring in the result. I agree with my colleagues that the government failed to shoulder the burden assigned to it by State Farm, and that Treasury Regulation § 1.263A-11(e)(1)(B), as it applies to property temporarily withdrawn from service, is accordingly unlawful. But I do not share the majority's view that this outcome also must derive from an irreconcilable incompatibility between I.R.C. § 263A's avoided cost principle and the regulation's policy of treating interest allocated to property withdrawn from service as an avoidable cost. The broad conclusion endorsed by the panel is unnecessary when this case can be resolved on narrower grounds.
There appears to be no dispute among the panel that the government has not articulated any rational explanation for many details of the regulation before us, from the regulation's first proposal in the mid-'90s up to the current date. Such a failure makes the regulation procedurally unlawful. I would reverse on the grounds set forth in part V of the majority opinion, but would give the government another chance to explain and justify its view that the adjusted basis of property temporarily withdrawn from service can be taken into account in determining production expenses.
Regarding the majority's Chevron review, I agree with its conclusion that section 263A(f)(4)(C) of the Tax Code leaves a sizeable “gap” concerning what is and is not to be considered a “production expenditure.” As perChevron step one, this gap invites the government to promulgate regulations more fully defining this term. Where I depart from the majority is in my conception of what exactly it means for an expense to be an “avoided cost.” In this appeal we are concerned with interest accumulating on a utility's operating debt. All agree that, when the utility takes on an improvement project, it is only fair that some of the interest accumulated during the project be capitalized, along with the project's direct costs, as “production expenditures.” A shorthand is to think of the utility borrowing money from itself to pay for the project. Both the “borrowed” principal and the accumulated interest are capitalized. So far, so good. The “avoided cost” rubric is another shorthand for the same concept. It says, if the utility had not taken on the improvement project, then the money it spent on the project would have been available to pay down the operating debt. As a result, both the money spent and the interest that accumulated on that money (because the debt was not paid down) are charged to the project. This appeal focuses on a subtle aspect of this regulatory approach. Say, as part of the hypothetical project, that the utility has to remove from service a plant that otherwise would have been running—generating electricity, employing workers, and generally creating value. Now that plant is non-operational, at least for the time it takes to complete the improvement. Is the plant's conversion from operational to (temporarily) non-operational status a “cost” of the improvement project? Can it be charged as a “production expenditure”? How? One vision of the associated-property rule in Treasury Regulation § 1.263A-11(e)(1)(B) is as an attempt to answer these questions. Yes, this approach says, the lost value associated with the withdrawal of property from service is a production expenditure that must be capitalized. But putting an actual number on the lost value presents administrative difficulties. Do we measure lost value in terms of revenue generated? Workers employed? Electricity produced? Some combination of these? What about other, less tangible forms of value? For convenience, one might propose to elide such an administratively-difficult assessment and simply assume that the lost value more or less equals the interest that accrued on the value wrapped up in the plant (i.e., the taxpayer's basis) while the plant was inoperative. So, this approach says, the taxpayer is instructed to take whatever basis it has in the plant and compute the interest that accrued on that basis during the period of non-operation. The majority is correct that fitting this approach into the “avoided cost” rubric requires adoption of a surprising fiction. If the interest accrued on the plant is to be thought of as an “avoided cost,” one must assume (counterfactually) that the taxpayer could have liquidated the plant and used the proceeds to pay down its operating debt. The majority views this fiction as such a departure from economic reality as to render the associated-property rule invalid. But, mindful that the avoided cost rule is in its entire concept a fiction, I am not convinced that its application here is so surprising as to merit overturning the regulation under step two ofChevron , if Treasury can properly explain its reasoning, which it has yet to do. The “avoided cost” rule is full of surprising fictional assumptions. Take, for example, a utility instructed by regulators to make some token improvement to its plant or have the plant shut down. Assuming this modest improvement would not require the plant to be withdrawn from service, no party to this appeal would dispute that the improvement costs (including associated interest) would be capitalized according to the “avoided cost” principle. But in no real sense are these “avoided costs,” as no rational utility would embrace regulatory shutdown merely to avoid the cost of a token improvement. It seems to me no more or less strange to say that a utility might “avoid” the cost of a plant being temporarily inoperable by selling it. The avoided cost rule is one of theoretical possibility, not economic reality. If we accept that there are real costs incurred in shutting down a property while improvements are made on it, such costs (assuming they can be quantified rationally) could have been avoided if funds had not been expended for the improvements. In reality, the cost of shutting down a property is no less “incurred” or “expended” than is the amount of money spent on the improvements. In other words, treating the cost of shutting down a plant temporarily while improvements are made as a “production expenditure” does not necessarily contradict the avoided cost rule that underlies section 263A. Indeed, I do not understand the majority to hold that the costs incurred in shutting down a property for improvements can never be treated as production expenditures. Instead, I read the majority to hold more pointedly that the adjusted basis of the property withdrawn from service cannot ever be treated as a production expenditure. Further, it seems to me that the associated property rule might be separately justified as necessary to fully implement aspects of the tax regime that every party acknowledges as lawful. It is important to keep in mind that this dispute concerns only the use of basis in underlying property that has been removed from service. No one here disputes that the full basis of underlying property can and should be taken into account for interest capitalization where the property never went into service in the first place. For example, if a taxpayer acquires a factory, then improves it before bringing the factory online, there is no dispute that the taxpayer's full basis in the factory (i.e., his purchase price, presumably) is taken into account in computing the amount of interest to be capitalized as an expense associated with acquiring the improved factory. In its briefing, the appellant acknowledged such an assessment as appropriate, but characterized it as involving “purchased property” and not invoking the associated property rule at issue here, even though Treasury Regulation § 1.263A-11(e)(1)(B) on its face applies to property “that is not placed in service.”See Appellant Br. 19; see also Treas. Reg. § 1.263A-1(e)(2)(i) (cited by appellant as the “purchased property rule”). Irrespective of how it is couched, the larger point is this: a taxpayer's basis in property into which improvements are incorporated does, in some cases, affect the amount of interest he must capitalize as a cost of that improvement. Such is undisputed. But, says the government, it creates an easy opportunity for tax evasion. A clever taxpayer, one who seeks to acquire the same improved factory just described but with smaller tax consequences, could structure his operations so that he falls outside the purchased property rule. Rather than buy a factory and then immediately set to work improving it, he puts the factory into service for some minimal period of time. His basis depreciates accordingly, albeit minimally. Then he takes the factory out of service and starts improving it. Because he now falls outside the strict text of the purchased property rule, his basis in the factory no longer contributes to the amount of interest he must capitalize, and so his tax burden drops substantially—even though the taxpayer's activities had no economically meaningful purpose other than tax avoidance. The associated-property rule would plug this regulatory gap; the majority's reasoning reopens it. To me that is a step that should be taken with caution, and is neither fully justified nor necessary here. Finally, I note that in this case the government has had substantial difficulty coming to grips with its own arguments. It apparently convinced the Court of Federal Claims that the associated property rule could be justified as an attempt to capture the opportunity costs associated with withdrawing property from service, even though it failed to explain as much in promulgating its rule. But its appellate briefing was, to say the least, opaque. And at oral argument, the government specifically rejected the “opportunity costs” approach, only to reverse its position in a post-argument letter. A litigant can, of course, be held to the consequences of its own failure to cogently present its case, particularly where that litigant is the federal government. But in such circumstances, the best approach is to tailor those consequences to the facts of the case. The outcome of this case can and should extend fromState Farm . The government's failure to justify its regulation ab initio left open the question of whether the avoided cost principle necessarily undermines any rationale that could justify treating an adjusted basis of property withdrawn from service for improvement as a production expenditure, for purposes of calculating interest to be capitalized. Such reaffirms my conclusion that this appeal does not present an appropriate vehicle for deciding theChevron question. It is therefore a more discreet approach to leave that question aside. The approach taken by the majority goes too far in that it creates a binding rule (at least in this circuit) that the government can never re-promulgate its associated-property rule for property temporarily withdrawn from service, no matter how well-formed its reasoning. I therefore concur in the result, and in Part V, of the majority's opinion, but otherwise respectfully dissent.