Andrea Wolfman, Atty., F.E.R.C., Washington, D.C., with whom Stephen R. Melton, Acting Gen. Counsel, and Barbara J. Weller, Deputy Sol., F.E.R.C., Washington, D.C., were on brief, for respondent. Robert F. Shapiro, Atty., F.E.R.C., Washington, D.C., also entered an appearance.

John H. Pickering, Washington, D.C., with whom Giles D.H. Snyder, Charleston, W.Va., Timothy N. Black, and James E. Coleman, Jr., Washington, D.C., were on brief, for intervenors Columbia Gas Transmission Corp. and Columbia Gulf Transmission Co.

Dale A. Wright, James T. McManus, and John H. Cheatham, III, Washington, D.C., were on brief, for intervenor Interstate Natural Gas Ass'n of America.

Before BORK and SCALIA, Circuit Judges, and GASCH,* District Judge for the United States District Court for the District of Columbia.

Opinion for the Court filed by Circuit Judge SCALIA.

Each year since 1947, Columbia Gas System, Inc. ("Columbia System"), a public utility holding company, has filed a consolidated federal income tax return on behalf of itself and its subsidiaries, pursuant to the Internal Revenue Code of 1954, 26 U.S.C. Sec. 1501 (1982). The subsidiaries include two interstate natural gas pipelines subject to the jurisdiction of the Federal Energy Regulatory Commission ("FERC"): Columbia Gas Transmission Corp. ("Columbia Gas") and Columbia Gulf Transmission Co. ("Columbia Gulf"). For the second time, see City of Charlottesville, Va. v. FERC, 661 F.2d 945 (D.C.Cir.1981) ("Charlottesville I "), the City of Charlottesville, Virginia, a customer of Columbia Gas, much of whose gas is transported by Columbia Gulf, petitions this court for review of the Commission's decision to use a "stand-alone" methodology to determine the tax allowances included in the costs of services, and hence the rates, of the two pipelines. Charlottesville contends that FERC must require the pipelines to share with their ratepayers the tax savings resulting from the use of tax losses of the system's gas supply subsidiaries in the consolidated return; and that the Commission's methodology fails to accord to ratepayers, as the Commission agrees it should, a proportionate share of the tax savings resulting from the parent company's interest expense deduction from consolidated income.

* The Natural Gas Act requires the Commission to insure that the rates of pipelines subject to its jurisdiction are "just and reasonable." 15 U.S.C. Sec. 717c(a) (1982). Under cost-of-service ratemaking principles, this means rates yielding sufficient revenue to cover all proper costs, including federal income taxes, plus a specified return on invested capital. See Public Service Co. of New Mexico v. FERC, 653 F.2d 681, 683 (D.C.Cir.1981). When a utility is an unaffiliated company and files its own tax return, calculation of the appropriate tax allowance is in principle no different from determination of any other operating or maintenance expense, although differences in the timing of recognition of expenses for tax and ratemaking purposes often complicate the calculation. See Public Systems v. FERC, 709 F.2d 73, 75 (D.C.Cir.1983). When the utility is part of a consolidated group, however, only the group files a return and pays taxes, computed on the cumulated revenues and deductions of all affiliates and the parent. The Commission must nonetheless determine separate tax allowances for the affiliated utilities it regulates, which it has done over the years by using one or the other of two basic methodologies.

The more venerable, the so-called "flow-through" methodology, immediately reflects in rates all the tax savings resulting from use of a consolidated return. These savings are twofold. First, unless all affiliates with tax losses would be able to take advantage of carry-forward or carry-back provisions of the tax code allowing them to use their losses to reduce their own future taxes or collect refunds on past payments, see 26 U.S.C. Sec. 172, aggregating them in a consolidated return with income-producing affiliates results in less taxable income for the group than there would be if each affiliate filed separately. Second, even if each affiliate could make use of its losses in future periods, the group gains the time value of money by using the losses immediately. Although the group in fact pays taxes on its net aggregate income at the statutorily prescribed rate, it can be thought of as, in a sense, paying at a "lower" tax rate than it would if all of its affiliates filed separate returns, because of the lower total taxable income that results from consolidation. A flow-through methodology applies this hypothetical lower (or "effective") tax rate to the earnings of the (presumably) profitable utilities to yield the tax allowances. For instance, the form of flow-through recommended in this case by Charlottesville would derive an effective tax rate by determining the ratio of each pipeline's taxable income to the total taxable income of all affiliates, multiply this fraction by the group's consolidated tax liability, and divide this figure by the pipeline's taxable income.

The second basic methodology, used by the Commission only in recent years, seeks, for tax purposes, to segregate the affiliated utility from the rest of the consolidated group. With minor adjustments not pertinent here, the calculation proceeds as follows: The utility's tax base is determined by identifying the taxable income and deductions of the consolidated group specifically attributable to the utility's jurisdictional activities. The statutory tax rate (which, in the case of regulated utilities, will almost always be the maximum rate) is then applied to the tax base to yield the stand-alone tax allowance. Usually, as in the present case, a stand-alone calculation will produce a higher tax allowance than a flow-through methodology, because the tax base, and hence the effective tax rate, is not reduced by the tax losses of other affiliates.

Until 1972, the Commission (then the Federal Power Commission) generally required regulated companies to flow through consolidated tax savings to their ratepayers. In United Gas Pipe Line Co. v. FPC, 357 F.2d 230 (5th Cir.1966), and Cities Service Gas Co. v. FPC, 337 F.2d 97 (10th Cir.1964), two circuit courts held that it was beyond the Commission's statutory authority to consider nonjurisdictional losses of members of a consolidated group when determining the proper tax allowance for jurisdictional companies. The Supreme Court reversed these decisions, holding that "in the proper circumstances the Commission has the power to reduce cost of service, and hence rates, based on the application of nonjurisdictional losses to jurisdictional income." FPC v. United Gas Pipe Line Co., 386 U.S. 237, 245, 87 S. Ct. 1003, 1008, 18 L. Ed. 2d 18 (1967). The Court thus authorized, though it did not require, the general use of flow-through methodology to calculate tax allowances.

Five years later, perceiving a need to encourage gas exploration, the Commission decided to switch to stand-alone methodology, announcing that

if a case were before us where an affiliated, regulated or non-regulated, producer of oil or gas showed a tax loss, and this loss company were joined in a consolidated return with a pipeline, the United or Cities Service cases, in our opinion, would no longer be persuasive authority, for to reduce the rates of a regulated pipeline because of such affiliated exploration and development activities would be discouraging to the very enterprise we now want to encourage.

Florida Gas Transmission Co., 47 F.P.C. 341, 362 (1972). More generally, the Commission felt that "as the pipeline business develops with numerous subsidiaries we should avoid regulating one company on the basis of the activities of others in the affiliated group." Id. at 363. It concluded that "a utility should be regulated on the basis of its being an independent entity; that is a utility should be considered as nearly as possible on its own merits and not on those of its affiliates." Id.

The Commission further explained the stand-alone methodology in Southern California Edison Co., 59 F.P.C. 2167 (1977). Edison, an electric rather than a gas utility, and a member of a consolidated group, had calculated its tax allowance on a stand-alone basis, which practice a Commission administrative law judge had upheld on the strength of the "incentive" rationale of Florida Gas. The Commission acknowledged that the Florida Gas policy of encouraging exploration was based on the special circumstances of the gas industry, but nonetheless found the stand-alone policy applicable to Edison (and all other utilities) because "as an alternative ground for decision in [Florida Gas ] ..., the Commission determined that utilities should henceforth be considered as separate entities for the ratemaking purpose of determining their costs of providing jurisdictional service." 59 F.P.C. at 2173. This is to be done by considering the source of the consolidated tax savings, which normally

will be attributable to business activities which are totally unrelated to the providing of electric utility service.... Certainly, if this were the case, it would be difficult to justify the appropriation of any tax savings attributable to [such activities] ... by the jurisdictional consumers when these same consumers did not pay the expenses which created the deductions for tax purposes.

Id. at 2174 (emphasis added). The standard set out in the emphasized passage is often referred to as the "benefits/burdens" test: the benefits of consolidated tax savings are given to ratepayers (by reducing the jurisdictional affiliate's tax allowance) if they bore the burden of paying the deductible expenses that generated the savings.

The present chapter in the ongoing stand-alone saga arose out of February 28, 1973 rate filings by the two Columbia pipelines. Previously the pipelines had, in accordance with the Commission policy affirmed by the Supreme Court in United, claimed tax allowances calculated on the basis of the consolidated tax liability of the affiliated group. These were consistently less than they would have been had each pipeline filed a separate return, since every year a number of their sister companies and their parent generated considerable tax losses. The sister companies' losses were almost entirely the result of gas exploration and development activities, which in the early stages often generate substantial tax deductions and little income. The parent, Columbia Gas System, Inc., always shows a year-end tax loss on a separate return basis,1 because the interest income it receives from its subsidiaries (plus a small amount of interest earned on short-term outside investments) is always less than the interest it pays to outside creditors and claims as a deduction, see 26 U.S.C. Sec. 163(a). The gap between interest income and interest expense is attributable in small measure to the fact that the parent, which does all of the outside borrowing for the system, lends capital to its subsidiaries at a slightly lower interest rate than it pays,2 but principally to the fact that the parent's capital structure is more heavily debt-oriented than are the capital structures that it maintains for its subsidiaries.

In the Columbia pipelines' 1973 rate case, their first rate case after Florida Gas, the pipelines ceased reflecting consolidated tax savings in their proposed tax allowances. Since settlement was reached on all other issues, the propriety of that tax treatment was reserved for separate hearing in the rate proceeding involving the pipelines' subsequent May 30, 1975 rate filings. See Columbia Gas Transmission Corp., 54 F.P.C. 2578, 2579-80 (1975).3 Following this hearing, the administrative law judge held that the stand-alone policy was "fatally defective as applied," Columbia Gulf Transmission Co., 8 F.E.R.C. (CCH) p 63,009, at 65,145 (July 7, 1977), and that "the tax savings resulting from filing a consolidated return have to be reflected in the respective costs of service of the Columbia pipelines." Id. The ALJ found that because the latest nationwide rates for gas producers included a stand-alone tax allowance to encourage exploration and development, the inclusion of a second-level stand-alone allowance for pipelines would be double-counting. He further found that much of the money collected as tax allowances by the Columbia pipelines in 1973-75 went to "general corporate purposes" rather than to exploration and development affiliates as contemplated by the Florida Gas rationale, and that even when exploration took place, there was no assurance that the Columbia pipelines' ratepayers would benefit from it. He strongly urged FERC to reassess generally the policy of Florida Gas. Id. at 65,148, 65,150.

In Opinion No. 47, reversing the ALJ's decision, FERC reaffirmed Florida Gas, holding that "[n]othing brought to our attention invalidates its reasoning." Columbia Gulf Transmission Co., 8 F.E.R.C. (CCH) p 61,002, at 61,009 (July 2, 1979) (footnote omitted). The Commission emphasized that the need to encourage gas exploration had not abated since 1972, and emphatically disagreed with the ALJ's assessment of the character and effect of the 1976 nationwide rate orders. Id. at 61,007-08. It concluded that "[t]o require Columbia to pay its pipeline customers the tax savings from losses incurred in the search for gas supplies will operate as a disincentive to continued gas supply development activities by pipelines and other regulated utilities." Id. at 61,008. With respect to losses generated by the parent company's method of financing its subsidiaries, the Commission noted that the pipelines' allowable rates of return were calculated using the capital structure and interest costs of the parent because the capital structures and interest costs of the wholly-owned pipelines themselves, which issue no securities to the public, are determined solely by the parent rather than the marketplace. In such cases, the Commission said, it must either use the capital structure and interest costs of the parent or construct a hypothetical structure and costs that attempt to approximate what market judgments would produce.4 In this case, all parties agreed that the former was preferable.5 In calculating the pipelines' tax allowances pursuant to the methodology described earlier, see supra page 1207, the Commission allowed as deductions interest expenses based, as the pipelines' rates were, upon the parent's structure and costs. FERC maintained that this procedure already gave "the pipelines' customers ... a proportionate share of the tax deductions related to Columbia's interest costs," and that the approach adopted by the ALJ of applying in addition the parent's effective tax rate "would give pipeline customers the benefits of the same tax deductions twice." 8 F.E.R.C. at 61,008 (footnote omitted). Accordingly, over the dissent of two commissioners, id. at 61,011-17, FERC held the tax allowances claimed by the pipelines just and reasonable. Id. at 61,009.

Charlottesville's petition for rehearing was denied by the Commission in Opinion No. 47-A, Columbia Gulf Transmission Co., 9 F.E.R.C. (CCH) p 61,355 (Dec. 20, 1979), which generally restated the reasoning of Opinion No. 47. The Commission added, however, that it would depart from its view that pipelines with loss affiliates need not charge lower rates than unaffiliated pipelines "when the customers of a regulated entity contributed to the expenses which created the loss deductions of the affiliate in the consolidated tax group." Id. at 61,746 (citing Southern California Edison ). The Commission found this exception inapplicable. Id. at 61,746-47.

Charlottesville petitioned this court for review of Opinions Nos. 47 and 47-A. The panel upheld an aspect of the decisions not at issue here, see Charlottesville I, 661 F.2d at 952, but remanded the question of treatment of the above described tax losses to the Commission. While finding that the Commission had the statutory authority to employ the stand-alone method, id. at 953, 954, the panel majority did not believe that the Commission had provided adequate evidentiary support for its basic premise that stand-alone tax allowances would in this case encourage exploration and development. Id.; id. at 954-55 (Wald, J., concurring).6 With respect to the tax losses of the parent company, the court unanimously found the record "incomplete on the relationship between the tax costs and capital costs included in the rate base." Id. at 952.

On remand, in Opinion No. 173, the Commission reiterated its adherence to the stand-alone method, but for reasons different from those in Opinion No. 47. FERC said:

In reviewing this record we have reconsidered the policy question from the ground up. Though we reach the same result on that question as did the Commission in Opinion No. 47 and Opinion No. 47-A, we do so for considerably different reasons. Specifically, we place no reliance on what has come to be called the "incentive rationale" that the Commission relied on in those Opinions. We do so because the policy the Commission applied is broader and more fundamental than one needed to encourage gas supplies. The policy involves a basic decision about the proper way to set cost-based rates. As such, the policy applies not only when a pipeline joins with an exploration affiliate in filing a consolidated return but also when any jurisdictional company, be it a pipeline or an electric utility, joins with other businesses in filing a consolidated return.

Columbia Gulf Transmission Co., 23 F.E.R.C. (CCH) p 61,396, at 61,870 n. 8 (June 22, 1983). In place of the incentive rationale of Florida Gas, the Commission relied on the benefits/burdens rationale set forth in Southern California Edison, which it described in Opinion No. 173 as the principle of "cost responsibility or cost incurrence," 23 F.E.R.C. at 61,850, i.e., the allocation of the consolidated company's costs (all costs and not merely taxes) to the regulated subsidiary "on the basis of a causal link," id., between those costs and the services the subsidiary provides to its ratepayers. In the case of taxes, this allocation is to be made by, first, estimating the taxable income the company is expected to receive from providing jurisdictional service (a relatively easy task) and, second, identifying the deductions of the group fairly traceable to the provision of service. This latter process is conducted as follows:

Because deductions are given for expenses incurred in producing income, the necessary causal link between the ratepayers and the deductions is the expenses the company incurs in providing service. Accordingly, the proper way to allocate deductions is to match the deductions with the expenses included in the cost of service. Thus, when an expense is included in the cost of service, the corresponding tax deduction is also allocated to the ratepayers. In this way any tax reducing benefits, or savings, the company realizes in providing the service are recognized in calculating the tax allowance for the benefit of the ratepayers.

The corollary to this is that when an expense is not included in the cost of service (because the company did not incur that expense in providing service), the deduction created by that expense is not allocated to the ratepayers. To do otherwise would result in the tax savings the company realizes from expenses incurred in providing services to other groups and periods or for its own benefit being used to reduce rates for a particular group of ratepayers. The tax allowance would then be lower or higher than is warranted by the profit each group provides the company. Since the amount of profit to be provided is the measure of the tax cost the company will incur in providing service, none of the rates for the groups would be cost-justified. Subsidization would inevitably result. One group would bear the burden, but another group would gain the benefit.

Id. at 61,851.

Applying this rationale to the tax losses generated by the activities of the exploration and development affiliates, the Commission found that there was an insufficient connection between the burdens imposed on the pipelines' ratepayers and the tax benefits enjoyed by the consolidated group to warrant allocating a share of the savings to the ratepayers. Id. at 61,861-63. With respect to the interest-expense losses of the parent, the Commission again found, as it had in Opinion No. 47, that by using the parent's capital structure and interest costs to determine the appropriate interest deductions in the pipelines' tax allowances, the stand-alone method did in fact flow through a proportionate share of these consolidated tax savings, and to flow them through again (e.g., by use of an effective tax rate) would be double counting. Id. at 61,866-68.

FERC denied without comment Charlottesville's petition for rehearing, Columbia Gulf Transmission Co., 24 F.E.R.C. (CCH) p 61,258 (Aug. 22, 1983), and Charlottesville once again petitioned for review in this court pursuant to 15 U.S.C. Sec. 717r(b) (1982).

Charlottesville raises several threshold procedural objections. It first contends that the Commission improperly prejudged the outcome of the remand proceeding, citing the language used by the Commission when it remanded the case to the ALJ following Charlottesville I:

The other remanded questions involve our general policy on consolidated taxes. Our policy in the past has been that "a utility should be regulated on the basis of its being an independent entity; that is a utility should be considered as nearly as possible on its own merits and not on those of its affiliates." Our present view remains the same, based on our conviction about the proper way to set rates for a regulated company. Evidence of a particular company's circumstances is not needed to make this policy determination. In light of the remand, however, the parties should address the validity of our "stand alone" policy in their presentations to the law judge.

Columbia Gulf Transmission Co., 20 F.E.R.C. p 61,036, at 61,069 (July 8, 1982) (quoting Florida Gas, 47 F.P.C. at 363). In order to establish improper prejudgment of a case, it must appear to " 'a disinterested observer ... that [the agency] has in some measure adjudged the facts as well as the law of a particular case in advance of hearing it.' " Cinderella Career and Finishing Schools, Inc. v. FTC, 425 F.2d 583, 591 (D.C.Cir.1970) (quoting Gilligan, Will & Co. v. SEC, 267 F.2d 461, 469 (2d Cir.), cert. denied, 361 U.S. 896, 80 S. Ct. 200, 4 L. Ed. 2d 152 (1959)). The above-quoted passage from the Commission's order may establish that FERC "prejudged" the law as to whether the stand-alone policy is consistent with its statutory mandate (on which point it had in any event been upheld in Charlottesville I, see 661 F.2d at 953, 954), and as to whether the nature of that policy requires case-by-case demonstration of incentive effect on exploration and development. But preconceptions regarding the law no more invalidate agency action than they do the action of a court. See FTC v. Cement Institute, 333 U.S. 683, 700-03, 68 S. Ct. 793, 803-04, 92 L. Ed. 1010 (1948). The "facts" that the Commission would have had to prejudge for Charlottesville to prevail on the present argument are whether the stand-alone policy, assuming its nature and validity, sustains the rate order--i.e., whether the Columbia pipelines' ratepayers could satisfy the benefits/burdens test. There is no indication of any such prejudgment in this record.

Charlottesville further contends that FERC exceeded the scope of the remand of Charlottesville I by basing its decision in Opinion No. 173 on a different rationale (i.e., the benefits/burdens principle) than that advanced in Opinions Nos. 47 and 47-A. See 23 F.E.R.C. at 61,870 n. 8. Charlottesville argues that the remand in Charlottesville I was for the purpose of allowing the Commission to develop an adequate record to support the incentive rationale, "not to permit the Commission to seize upon whatever new theories had struck its fancy since the issuance of Opinion Nos. 47 and 47-A, as if it were dining at an evidentiary smorgasbord." Brief for Petitioners at 41. We do not agree that the mandate of Charlottesville I was so narrow. The court found the record on the incentive rationale insufficient to support the Commission's order, and remanded the case "for proceedings not inconsistent with this Opinion." 661 F.2d at 954. As a general proposition, "[an] administrative tribunal is free on remand to reach the same result on different grounds." Chicago and North Western Transportation Co. v. United States, 574 F.2d 926, 930 n. 5 (7th Cir.1978). See SEC v. Chenery Corp., 332 U.S. 194, 200-01, 67 S. Ct. 1575, 1579, 91 L. Ed. 1995 (1947). Nothing in the mandate of Charlottesville I forbids application of this rule, and the Commission was therefore free to reaffirm the stand-alone policy on different grounds.

Charlottesville's last procedural contention--that the benefits/burdens principle relied on in Opinion No. 173 was tacitly rejected by the decision in Charlottesville I and thus cannot support the Commission's action on remand--is too insubstantial to bear discussion. See 661 F.2d at 949, 951.

Charlottesville contends that the stand-alone methodology is per se unlawful because the Commission can properly consider only "actual taxes paid." It might suffice to say that Charlottesville I explicitly found the stand-alone policy consistent with the Commission's mandate, see 661 F.2d at 953-54, and explicitly rejected the "actual taxes paid" objection, see id. at 952 & n. 38. Since discussion of the objection in that earlier appeal was brief, however, and since the "actual taxes paid" concept is relevant not only to Charlottesville's attack upon stand-alone in principle, but to one of its subsequent challenges to the Commission's manner of applying stand-alone in the present case, it warrants some extended discussion.

The earliest case we have found applying the "actual taxes paid" principle by name is El Paso Natural Gas Co. v. FPC, 281 F.2d 567 (5th Cir.1960), cert. denied, 366 U.S. 912, 81 S. Ct. 1901, 6 L. Ed. 2d 1247 (1961), which ordered the Commission to require an integrated gas company to flow through to ratepayers tax savings resulting from depletion allowances, see 26 U.S.C. Secs. 611-613A, and the deduction of intangible drilling expenses, see 26 U.S.C. Sec. 263(c). The Fifth Circuit held:

We think, in short, that there is no statutory authority for the Commission to treat actual savings in taxes to which natural gas companies are entitled any differently than savings in any other cost of service. It is the obligation of all regulated public utilities to operate with all reasonable economies. This applies to tax savings as well as economies of management. The net result of this, of course, is that such savings as are effected are passed on to the consuming public. This we consider to be the natural and necessary consequence of rate regulation.

281 F.2d at 573. The court did not, however, require the flow-through of tax benefits resulting from accelerated depreciation, see 26 U.S.C. Sec. 167, allowing the utility instead to "normalize" those deductions7 on the ground that they resulted only in a tax deferral rather than a tax saving. See 281 F.2d at 573-74.8 See also Cities of Lexington, etc., Ky. v. FPC, 295 F.2d 109, 111-19 (4th Cir.1961) (following El Paso on these issues).

Subsequently to El Paso, the Commission determined that because the total depreciable plant of utilities was likely to grow or remain stable in the foreseeable future, meaning that excess depreciation deductions would be continuously available, accelerated depreciation would realistically result in tax savings rather than merely deferral, and it changed its policy to require flow-through. That decision was also upheld by the Fifth Circuit, see Alabama-Tennessee Natural Gas Co. v. FPC, 359 F.2d 318, 336-39 (5th Cir.), cert. denied, 385 U.S. 847, 87 S. Ct. 390, 17 L. Ed. 2d 310 (1966), and by this court, see City of Chicago, Ill. v. FPC, 385 F.2d 629, 633-34 (D.C.Cir.1967), cert. denied, 390 U.S. 945, 88 S. Ct. 1028, 19 L. Ed. 2d 1133 (1968). In support of our affirmance we alluded to the "general rule followed by the Commission, and indeed most if not all regulatory commissions, ... that the 'consumers should be charged for only the actual liability for Federal income taxes.' " 385 F.2d at 633 (quoting United Fuel Gas Co., 12 F.P.C. 251, 265 (1953)).

In a later case, however, we rejected a challenge to the Commission's use of area rates9 grounded squarely on the "actual taxes paid" principle, because the propriety of area-wide calculation of costs in general was not subject to question, see Permian Basin Area Rate Cases, 390 U.S. 747, 88 S. Ct. 1344, 20 L. Ed. 2d 312 (1968), and there was no reason to treat taxes differently from other costs. See City of Chicago, Ill. v. FPC, 458 F.2d 731, 754-56 (D.C.Cir.1971), cert. denied, 405 U.S. 1074, 92 S. Ct. 1495, 31 L. Ed. 2d 808 (1972). We said in dictum, however, that "[w]hen a company's rates are set under individual cost-of-service ..., only actual costs--tax and other--can be included." 458 F.2d at 756 (emphasis added).

If taken literally, the dictum would have utterly invalidated normalization. Such a conclusion was recognized and rejected by this court in Memphis Light, Gas & Water Division v. FPC, 500 F.2d 798 (D.C.Cir.1974). There Texas Gas Transmission Corp. sought and was granted a rate increase based on its change from flow-through to normalization of certain tax deductions. We found "no conflict between the rate increase ... and the 'actual expenditures theory,' " id. at 807, because "while Texas Gas [was] not currently paying taxes at straight-line rates, it [was] incurring the liability [then] to pay increased taxes in the future." Id. (emphasis added). This was held not to constitute "an increase of rates based on fictitious taxes," id., for if it was, then normalization, the legality of which was well established, "would always violate the actual taxes paid principle." Id. n. 61.

This court's next foray into actual taxes paid arose out of the Commission's nationwide rate orders covering gas produced from wells commenced after January 1, 1975, and prescribing new prices for gas produced from wells commenced after January 1, 1973. Consumers challenged the orders, which included tax components calculated on a nationwide rather than individual basis, on the ground that, by failing to require the use of tax returns to demonstrate actual liability, "the Commission has departed from the well-settled principle of regulation that rates provided to cover tax costs must be based on 'actual taxes paid.' " American Public Gas Ass'n v. FPC, 567 F.2d 1016, 1038 (D.C.Cir.1977), cert. denied, 435 U.S. 907, 98 S. Ct. 1456, 55 L. Ed. 2d 499 (1978). We held this to be a "misunderstanding of the case law dealing with that principle," 567 F.2d at 1038, explaining:

In rate regulation there is no mystique requiring that expenses be actually "paid." Regulated companies are routinely permitted to set up reserves against the probable expenses of obligations undertaken now yet falling due in the future even when the amount of obligation is subject to revision.... [T]here is no historical basis for petitioners' simplistic interpretation of the "actual taxes paid" principle.

Id. On the heels of this case came Tenneco Oil Co. v. FERC, 571 F.2d 834 (5th Cir.), cert. dismissed, 439 U.S. 801, 99 S. Ct. 43, 58 L. Ed. 2d 94 (1978), which upheld the nationwide rate orders for gas produced from wells commenced before January 1, 1973, including their use of a nationwide tax allowance. Although finding that "[t]he Commission ha[d] not shown that its income tax component [was] strictly cost-justified," 571 F.2d at 844, the court approved the component as "a reasonable effort to construct a rate element that would confront the reality of the tax laws rather than pretend they do not apply." Id. More recently, this court has approved the Commission's decision generally to employ normalization in calculating tax allowances, deferring to its judgment that normalization will not result in tax savings. See Public Systems v. FERC, 709 F.2d 73, 81-82 (D.C.Cir.1983).

We conclude from these and from other cases10 that the imprecision of the "actual taxes paid" formulation is exceeded only by the name of the Holy Roman Empire: two out of the three words are wrong. Taxes, yes. But not necessarily actual taxes, since inexact estimations are often allowed, e.g., a nationwide tax allowance applied to all individual utilities, see Tenneco Oil, 571 F.2d at 844. (In point of fact, of course, even the use of actual taxes for the test year by which future rates are calculated rarely succeeds in the objective of precisely estimating the actual taxes in the years when the rates are paid.) And not necessarily taxes paid, since tax liability incurred by current activities but in fact not paid currently can be charged to present ratepayers, e.g., taxes deferred by reason of accelerated depreciation but passed on to current ratepayers through normalization, see Public Systems, 709 F.2d at 81-82. So the principle should be expressed "actual or estimated taxes paid or incurred"--whereupon it ceases to constrain the Commission with regard to taxes any more than the Commission is constrained with regard to its treatment of other expenses. Which is as it should be.

Only one factor makes application of this modest principle to consolidated returns distinctive: The theoretical tax liability for the pipeline's net income may never be reflected on anyone 's actual tax bill, since it may be offset by losses of the parent and other affiliates. To take the harshest hypothetical case, if stand-alone is allowed the pipeline ratepayers may be assessed with a tax expense when the consolidated company in fact pays no taxes.11 We find that no obstacle to the Commission's action, neither in theory nor in economic reality. It is certainly reasonable to regard as no less real an economic detriment than cash payment of tax assessments, the consumption of otherwise usable tax losses to forestall those assessments. Had the affiliates reporting the losses filed separately, the losses could have been carried forward and used to offset their own income the following fifteen years (or carried back to yield refunds for the previous three); or they could have been used to offset the profits of some other company with which (had the pipeline not been in the consolidated group) the parent would have chosen to affiliate; instead, the parent used them, in place of cash as it were, to eliminate tax liability. Economic detriment to the parent is undeniable, even if, since the losses fully "pay" the tax, the tax bill is zero. One might properly say that the consumption of allowable tax losses cannot accurately be considered a "tax expense" since, though it is related to the assessment of tax liabilities, it does not pay an assessed tax, but rather prevents the assessment of one. But that is no more than a definitional quibble; if it is correct, then the Commission should call this allowable expense a "tax-related expense" instead of a "tax expense"--hardly reversible error. The point is that it is an expense of the consolidated company caused by the jurisdictional activities and thus properly assessed against the ratepayers.

We do not assert, of course, that this is the only rational or permissible view of things. In fact, in arguing that the pipelines' tax expense cannot exceed the consolidated tax liability Charlottesville does no more than place the Commission's own now-disavowed child on its doorstep. That is the principle the Commission explicitly recited in its 1963 Cities Service decision, imposing a "flow-through" rather than a "stand-alone" methodology. "[T]he amount of the consolidated tax payment," it said, "is the only real cost which was incurred by Gas Company in conjunction with the other Cities Service affiliates"; and the purpose of tax allowances is "not to insure that other components of a complex corporate system are enabled to 'cash-in' on their tax losses." Cities Service Gas Co., 30 F.P.C. 158, 162 (1963). In upholding the Commission's action, the Supreme Court appeared to adopt this argument as not permitting but requiring flow-through, see FPC v. United Gas Pipe Line Co., 386 U.S. 237, 243-44, 87 S. Ct. 1003, 1007, 18 L. Ed. 2d 18 (1967); id. at 254, 87 S.Ct. at 1012 (Harlan, J., dissenting), but as the Fifth Circuit explicitly said in Tenneco, see 571 F.2d at 843, and as we plainly implied in Charlottesville I, see 661 F.2d at 952, those expressions were dicta.

It can be argued that the parent's consumption of tax loss allowances is not necessarily an economic detriment, since those allowances might have been valueless in any event. The Internal Revenue Code's carry-back provisions generally extend only three years, see 26 U.S.C. Sec. 172(b)(1)(A), and the carry-forward provisions fifteen years, see Sec. 172(b)(1)(B), so that if the system were deprived of the pipelines' income as a current set-off, it could accumulate masses of "excess" tax losses that would prove useless because unmatched by other income within the permissible periods. Such speculation, however, involves a number of imponderables. For example, if confronted with the prospect of such "excess" losses the system could choose to capitalize rather than expense certain exploration deductions, which would have the effect of stretching out the tax benefits, making it more likely that the deductions could be used in future years. Or, if "excess" losses were unavoidable and stand-alone treatment of the pipeline was not permissible, the parent might (and rationally would) choose to remove the pipelines from the consolidated group, replacing them with other enterprises that would permit the shareholders (rather than ratepayers) to benefit from the set-off. In an era when enterprises are frequently the objects of acquisition or merger because of their temporary tax-loss benefits, such a prospect is far from fanciful. This speculation whether consumption of the tax losses represents a real economic detriment is reminiscent of the dispute, in the context of normalization, over whether the taxes deferred by reason of accelerated depreciation will in fact ever be paid, or will as a practical matter be postponed forever. See supra pages 1213-14. Just as the courts have left that call to the Commission, permitting it to conclude either way--first allowing normalization and later disallowing it because of indefinite postponement of tax liability, compare El Paso, 281 F.2d at 573-74, with Alabama-Tennessee, 359 F.2d at 336-39, and City of Chicago, 385 F.2d at 633-34--so also we think this matter is one for the Commission's judgment.

In sum, we are persuaded that the Commission's stand-alone approach is reasonable enough to survive our highly deferential review, see Permian Basin Area Rate Cases, 390 U.S. at 791, and reaffirm the holding of Charlottesville I, 661 F.2d at 949-50, that it is lawful.IV

It remains to be considered whether the Commission's methodology for allocating tax deductions for stand-alone purposes is reasonable, and whether it has been applied correctly to the facts of this case. The methodology in question is the so-called benefits/burdens test, which the Commission urged as an alternative ground for sustaining its action in Charlottesville I, but which we did not reach. That test, first set forth in substance in Florida Gas, 47 F.P.C. at 360-63, see also Southern California Edison, 59 F.P.C. at 2172-74, assigns to the pipelines those tax benefits (deductions) attributable to expenses whose burden was borne by the pipelines' ratepayers.12 The object of inquiry is whether "the customers of a regulated entity contributed to the expenses which created the loss deductions of the affiliate in the consolidated tax group." 9 F.E.R.C. at 61,746. This is on its face eminently reasonable, and we have no difficulty sustaining it in principle.

Despite Opinion No. 173's frequent references to the benefits/burdens test, see, e.g., 23 F.E.R.C. at 61,851, 61,861-62, Charlottesville asserts that the Commission was in fact not using it, but was inexplicably and without warning departing from its prior precedent to apply a "cost of service" test. This, according to Charlottesville, is "substantively different" and more stringent towards the ratepayers, Reply Brief for Petitioner at 28, since "it is no longer enough that 'the expenses which created the deductions used to achieve the tax savings were paid by the jurisdictional customers'; the expenses must now be paid by the jurisdictional customers in the pipeline's cost of service." Request of the City of Charlottesville, Virginia, for Rehearing of Opinion No. 173, Docket Nos. RP75-105 & RP75-106 (Consolidated Taxes) at 25 (filed July 22, 1983) (quoting Louisiana Power & Light Co., 14 F.E.R.C. (CCH) p 61,075, at 61,124 (Jan. 28, 1981)). We find this argument not merely unpersuasive but totally incomprehensible. Opinion No. 173 does, to be sure, expressly say that "the test is whether the expenses are included in the relevant cost of service," 23 F.E.R.C. at 61,853; but that is plainly an iteration rather than an alteration of the benefits/burdens test. A "cost of service" is a document filed by a rate-regulated utility showing all of the items requiring reimbursement from the ratepayers--its bottom line, in other words, is the "R" or "revenue requirement" element of the classic ratemaking formula R (revenue requirement) = C (operating costs) + Ir (invested capital or rate base times rate of return on capital). See T. MORGAN, ECONOMIC REGULATION OF BUSINESS 219 (1976). See also E. GELLHORN & R. PIERCE, JR., REGULATED INDUSTRIES IN A NUTSHELL 97 n. 1 (1982) (citing substantively identical formulae). That being so, it is beyond imagination how the ratepayers could possibly pay an expense without doing so "in the pipeline's cost of service." Charlottesville's argument to the contrary seems to rest upon the astonishing misconception that the phrase "cost of service" refers to "operating costs," i.e., the "C" rather than the "R" in the ratemaking formula. Thus, Charlottesville's brief says that "[t]he Commission fulfills [its duty to ensure 'just and reasonable' rates] in part by examining a pipeline's 'cost of service,' i.e., the various operating expenses it incurs in providing service to its customers...." Brief for Petitioner at 14. It suffices to say that this is appallingly wrong. We think it plain, in short, that the cost-of-service test and the benefits/burdens test are one and the same.

We turn, then, to the application of the benefits/burdens test in the present case. When the Commission specifically considered the burdens that Charlottesville maintained justified a sharing of consolidated tax savings, it concluded that "the burdens Charlottesville ... point[s] to (to the extent they are burdens at all) are simply too far removed from the tax benefits to justify allocating the benefits to the ratepayers." 23 F.E.R.C. at 61,861. This conclusion survives review, though as to some of the items on technical grounds.

Charlottesville suggested first that use of the pipelines' internally generated funds to finance the activities of the gas supply affiliates entitled the ratepayers to a share of the tax savings resulting from those activities. The Commission disagreed because use of the funds (consisting of profits, depreciation and deferred taxes) did not in its view burden the ratepayers, i.e., did not affect their rates. "[W]hat the pipelines' shareholders do with this cash is largely their own business," the Commission said. "They may reinvest it in the pipelines or they may invest it in other business ventures." Id. Assuming that to be true (which petitioner has not contested)13 the Commission's conclusion represents a reasonable application of the benefits/burdens test.

The Commission made essentially the same response to Charlottesville's contention that the Columbia System's practice of "funnelling" money from the pipelines to the gas supply affiliates burdens the ratepayers. This funnelling results from the procedures the System uses to allocate internally its consolidated tax payments. The parent collects money from its subsidiaries to discharge the System's consolidated tax liability. During the relevant time period, however, rather than collecting a portion based on the ratio of each subsidiary's taxable income to the consolidated tax liability, the System, pursuant to an exemption from S.E.C. Rule 45(b)(6), 17 C.F.R. Sec. 250.45(b)(6) (1980),14 excluded from the consolidated tax liability the savings (deductions) attributable to the gas supply development activities. The other subsidiaries thus paid to the parent more than was needed to discharge the current year's tax liability, and the excess was "funnelled" to the exploration and development companies, allowing them to convert their tax-deferring credits into immediate cash for use in the exploration and development efforts. In future years, when the gas supply subsidiaries finally become profitable, they will be allocated additional portions of the consolidated tax liability to offset this early-period funnelling effect. As far as the pipelines' ratepayers are concerned, however, this "funnelling" is somebody else's business. As the existence of this litigation testifies, what the ratepayers contribute to the consolidated tax liability is determined not by the amount of the parent's allocation, but by the amount that the Commission allows as an operating expense. To the extent that the former exceeds the latter, it comes out of the pockets of the shareholders of the pipelines (i.e., the parent itself). We agree with the Commission that, since the internal allocation in no way affects rates, there is, under the Commission's test, no burden entitling ratepayers to corresponding tax benefits. See 23 F.E.R.C. at 61,861-62.

The Commission acknowledged that the remaining assertions of "burden" advanced by Charlottesville had "some merit," id. at 61,862. One of these was that Columbia Gas made advance payments to Columbia Gas Development Corp. (and to gas supply companies not affiliated with the Columbia System), pursuant to a Commission program designed to increase gas exploration. See Accounting and Rate Treatment of Advance Payments to Suppliers for Gas and Amending F.P.C. Form No. 2, 44 F.P.C. 1142 (1970). The advances were to serve essentially as interest-free loans to the producers. To encourage such payments, the Commission allowed pipelines to include them in their rate base and thus to collect a return on the advanced money from the ratepayers.15 In the present proceeding, the Commission necessarily agreed that this constituted a burden on the ratepayers, but did not consider the burden sufficiently linked with a corresponding tax benefit. "[The] advance payments," it said, "did not create any tax benefits for either the pipeline or the gas supply company. So there is no relationship between this burden and the tax benefits to be allocated." 23 F.E.R.C. at 61,862. The Commission evidently believes that merely contributing the cost of capital for activities that generate tax-deductible expenses, as opposed to contributing to the expenses themselves, does not establish sufficient connection between the burden and the benefit. It means quite narrowly, in other words, its requirement that ratepayers must pay "the expenses which created the deductions used to achieve the tax savings," id. at 61,871 n. 17 (quoting Louisiana Power & Light Co., 14 F.E.R.C. at 61,124). We must decline to review the propriety of this approach, because we conclude that the issue is not properly before us. The Natural Gas Act provides that, on a petition for review, "[n]o objection to the order of the Commission shall be considered by the court unless such objection shall have been urged before the Commission in the application for rehearing unless there is reasonable ground for failure so to do." 15 U.S.C. Sec. 717r(b). Nothing resembling an objection to the Commission's holding on this point was argued to the Commission. The rehearing petition assigned as error the Commission's "holding that the payment by the Columbia pipelines of advance payments to Columbia Gas Development Corporation did not constitute enough of a burden to the Columbia pipelines' ratepayers to justify the sharing of consolidated tax savings with those ratepayers." Request of the City of Charlottesville, Virginia, for Rehearing of Opinion No. 173 at 5. As we have described, however, FERC did not contest the reality or substantiality of the burden, but rather held that the corresponding benefit was too remotely related. Nor can we find the proper issue fairly raised by the rehearing petition's argument, which simply protested the Commission's alleged substitution of the cost-of-service test for the "contribution to expenses" test of Opinion No. 47-A and the "payment of expenses" test of Louisiana Power & Light Co., 14 F.E.R.C. at 61,124. See Request of the City of Charlottesville, Virginia, for Rehearing of Opinion No. 173 at 25, 28.

Charlottesville's other assertion of burden acknowledged by the Commission to have had "some merit" suffers from the same infirmity, as far as this appeal is concerned, of having been rejected by the Commission on a basis that Charlottesville did not contest on rehearing. In calculating the pipelines' rates of return, the Commission, with the agreement of all parties including Charlottesville, had used the capital structure, cost of debt, and cost of equity of the parent, the only company in the system that sells securities to the market. Charlottesville contends that because the gas supply companies increase the parent's cost of debt and equity (financing for exploration and production being more expensive than for other, less risky system activities), the rates of return for the pipelines reflect a "risk premium" attributable to the gas supply companies' activities--and that the ratepayers should share in the tax benefits this burden generates. The Commission agreed that there was a "risk premium" burden on the ratepayers as to debt costs, though not as to equity costs, see 23 F.E.R.C. at 61,863, 61,875 n. 83, and said that "[t]his evidence might justify, on purely equitable grounds, some sharing of the consolidated tax savings if there were no other way to protect the ratepayers from this increased cost of capital," id. at 61,863.16 The Commission contended, however, that a superior way to protect the ratepayers is to adjust the rate of return to take account of these concerns. It catalogued a number of cases in which adjustments to the capital structure and capital costs used in the rate of return calculation were made--or endorsed in principle--in similar situations. Id. Hence, "nothing prevented the parties from raising their concerns that the return was too high when that issue was presented to the Commission for decision. But they did not. Instead, they settled." Id. (footnote omitted). FERC thus declined to use the tax allowance calculation to correct a rate of return calculation better dealt with directly during the settlement process (or in litigation over the proper rate of return). Id. Despite this explicit statement of the Commission's reasons, Charlottesville's assignment of error in its rehearing petition was that:

The Commission erred in holding that the use of the parent company's cost of raising debt and equity capital and the parent's capital structure (which reflect the increased business risks of the Columbia system's unregulated subsidiaries) in setting the Columbia pipelines' rate of return did not constitute a burden to the ratepayers that would justify the sharing of consolidated tax savings.

Request of the City of Charlottesville, Virginia, for Rehearing of Opinion No. 173 at 6 (footnote omitted). This did not place the Commission's actual reasoning in issue, and hence we must decline the request to review it here. We might add, with regard to both of the last two points, that Charlottesville not only utterly failed to address the Commission's actual holdings in its rehearing petition and argument, but only by the most sympathetic interpretation can be said to have addressed them before us. Particularly in a case as complex, for both the Commission and the courts, as this ratemaking proceeding, our refusal to rescue litigants from the consequences of their confusing imprecision is not only a required course but a fair and efficient one.

Finally, Charlottesville makes a more broad-based attack on the Commission's application of the benefits/burdens methodology. That application represents, Charlottesville maintains, a one-sided view of the tax consolidation process, since it implicitly assumes that the benefits of that process are attributable exclusively to the availability of "excess" deductions--i.e., deductions that would (if separate filing were made) exceed an affiliate's taxable income. Under the Commission's approach, only when a pipeline contributes to production of those deductions is it regarded as contributing to the tax benefits. In fact, however, excess deductions are valueless (or, if there are carry-forward possibilities, of lesser and uncertain value) unless there is past or current taxable income to offset. The tax benefits of consolidation are, in other words, a synergistic product that requires both income and deductions. Hence, Charlottesville argues, the Commission should make some allowance to the pipelines' ratepayers for the fact that the steady stream of taxable income they provide to the consolidated group helps give immediate and certain value to the deductions of the loss companies. The argument is simply a rerun, in the context of methodology, of the fundamental objection to stand-alone expensing that we discussed earlier. No more in this context than there is it a proper basis for overturning the Commission's judgment. The object of the present exercise is not theoretically accurate assignment of causality for the tax advantages of consolidated filing, but estimation of the tax liability attributable to operations of a regulated company that happens to be an affiliate. There are a number of plausible ways to make that estimation--ranging, perhaps, from an approach that would give the utility's ratepayers the benefit of all tax deductions of the consolidated group offset against the utility's income (since the deductions would have been worthless without the income) to an approach that would give ratepayers the benefit of none of them (since the utility would have had no deductions on its own). Within certain rational limits that have clearly not been exceeded here, which approach to choose is more a matter of regulatory policy than of logic. The approach the Commission has chosen, allowing those deductions made possible by charges to the ratepayers, is an entirely reasonable one, beyond our authority to upset. See FPC v. United Gas Pipe Line Co., 386 U.S. 237, 246, 87 S. Ct. 1003, 1008, 18 L. Ed. 2d 18 (1967); Colorado Interstate Gas Co. v. FPC, 324 U.S. 581, 589, 65 S. Ct. 829, 833, 89 L. Ed. 1206 (1945); Charlottesville I, 661 F.2d at 952.

As noted earlier, the Commission, with the agreement of all parties, uses the capital structure and capital costs of the parent, Columbia Gas System, Inc., to determine the pipelines' rates of return. Since the parent has more debt in its capital structure than the pipelines, and pays outside creditors of the system a marginally higher interest rate than it charges its affiliates, the effect of this treatment is to increase the pipelines' debt costs,17 which are charged to the ratepayers in the cost of service. The Commission agrees with Charlottesville that since the ratepayers bore the "burden" of those higher costs, they should enjoy the tax "benefits" attributable to them. The difference between the parties is that the Commission contends, and Charlottesville denies, that the stand-alone method in fact achieves this result.

At first blush, the Commission would seem to be correct, since, when calculating stand-alone tax allowances, the Commission permits interest-expense deductions based, like the rates, upon the parent 's interest costs and not upon the lower costs of the pipelines. Charlottesville points out, however, that this is required ratemaking procedure even when no consolidated tax return is filed, because it "synchronizes" the interest costs used in calculating the rate of return and tax allowance; and that it alone therefore does not pass through to ratepayers a share of the consolidated tax savings gained by having the parent join in a consolidated return.

But for the fact that we took this argument seriously in Charlottesville I, we would be inclined to say that it makes no sense. Consolidated tax savings cannot be passed through unless consolidated tax savings exist, and we are here talking about a situation in which there concededly are none--i.e., the portion of the interest expense deduction on the consolidated return allocable to each pipeline's operations, and the interest expense deduction on each pipeline's hypothetical stand-alone return are precisely the same. That should be the end of the matter. It is no more rational to require this deduction to be run through the stand-alone return a second time than it would be to accord similar double-counting treatment to salaries, rents, and many other deductions that are reflected on both the consolidated and the stand-alone returns.

Charlottesville's response is, in effect, that the interest expense deduction differs from the other deductions on the stand-alone return in that it does not reflect a proper expense of the pipeline, but is improperly allowed to be deducted against taxes only because it was improperly allowed to be charged to ratepayers; and that it would undo this two-wrongs-make-a-right adjustment to refuse to give the pipelines the benefit of the real expense reflected on the consolidated return. Charlottesville does not put the point this way, but that is what it comes to. The premise of the argument is that the parent's capital structure and costs, which are attributed to the pipelines for ratemaking purposes, do not accurately reflect the pipelines' cost of capital; and that allowing the parent's interest expense to be deducted on the stand-alone returns does no more than compensate for this inaccuracy. The Commission's response to this was in part as follows: "[I]n this case the parties, including Charlottesville, have agreed to use the parent's capital structure and costs. That agreement is final. So it is too late to complain now that the return is too high." 23 F.E.R.C. at 61,869. As an original matter, that seems to us entirely correct. We do not think an agreement that imputed capital structure and costs are proper can be distorted into an agreement that imputed capital structure and costs, though improper, are compensated for by an improper tax allowance--especially since the tax allowance would only recover a percentage of the interest charges improperly assessed against the ratepayers. We think the terms of our remand in Charlottesville I, however, preclude this response. The mandate of that decision was for the Commission to "consider whether the capital distribution tax benefits are in fact passed-through to ratepayers or whether the pass-through is vitiated by rates premised on a rate base18 which is inflated by the parent's cost of capital." 661 F.2d at 952. If this means anything, it is that Charlottesville was entitled to raise, and the Commission required to address, the relationship between the tax allowance and the rate of return. The Commission addressed that point as well:

[D]espite its statements to the contrary, Charlottesville's position is that using the parent's capital structure and costs has inflated the return. Is this so? Nothing in the record shows that a return based on the pipelines' own capital structure and costs would be lower. But even if there were, we could not conclude that the return is inflated. As we have said previously, where a subsidiary issues no securities to the public, its capital structure and costs should not be used. Those elements have been determined by the parent, not the market. The parent can manipulate those elements for its own advantage. That cannot be permitted.

23 F.E.R.C. at 61,869 (footnotes omitted).

We agree with the Commission on both counts. Nothing in the record establishes that use of the pipelines' own capital structure and costs would have produced a lower rate of return. Charlottesville prepared an exhibit purporting to show that if the pipelines' capital structures and costs19 were used, the rate would have been 9.06 percent instead of the settlement rate based upon the parent's capital structure and costs,20 of 9.49 percent. The Commission questioned the usefulness of the exhibit because "the 9.06% rate of return was not based on a test year estimate of interest expense. It was based on amounts recorded on the pipelines' books." Id. at 61,876 n. 121. That was clearly a valid criticism. If one takes the test year interest (7.29 percent), rounds it down to the nearest one-tenth of one percent as the parent regularly does for intra-system loans (7.20 percent), and substitutes this figure for the 6.40 percent used in Charlottesville's exhibit, the "actual" rate of return is 9.51 percent, which is higher than the settlement rate. It is true, as Charlottesville notes, that the pipelines' lower debt/equity ratio would, all else being equal, reduce their costs of equity (by reducing risk), but the record contains nothing to establish (1) that all else is equal and (2) that the adjustment in the costs of equity would push the "actual" return below the settlement figure of 9.49 percent.

More fundamentally, however, we also agree with the Commission that even if the pipelines' "actual" capital structures and debt costs yielded a lower return, the Commission was under no obligation to use them. The Commission quite reasonably determined that the capital structures and intra-system borrowing costs of wholly owned subsidiaries reflect the convenience of the parent, which may or may not accord with market reality. That being so, the only options are to use the structure and costs of the parent (which are, or at least should be, determined by market reality) or to create hypothetical structures and costs. Thus, if Charlottesville wished to argue that the rates of return calculated using the former were inflated (producing a distortion which the stand-alone tax deduction, even though less than the distortion, was designed to set off), the proper comparison was with rates calculated not on the basis of the pipelines' artificial "actual" structures and costs, but on the basis of hypothetical arrangements reflecting market realities. We find nothing in the record suggesting that the latter would have been lower than the settlement rate of return. Moreover, in light of the Commission's recent rejection of the use of hypothetical capital structures, see supra page 1210 note 5, it may be that it is no longer possible to establish a capital-inflated rate of return except by reference to the parent's capital structure--i.e., by showing that the parent's capital costs are excessive or that its debt/equity ratio is too low.

* * *

* * *

For the reasons stated, the petition for review must be

Denied.