Transwestern Pipeline Company, Petitioner, v. Federal Energy Regulatory Commission, Respondent, Conoco,inc., et al., Intervenors, 59 F.3d 222 (D.C. Cir. 1995)

Annotate this Case
US Court of Appeals for the District of Columbia Circuit - 59 F.3d 222 (D.C. Cir. 1995) Argued May 8, 1995. Decided July 14, 1995

On Petition for Review of Orders of the Federal Energy Regulatory Commission.

Harvey Y. Morris, argued the cause for petitioner Public Utilities Com'n of the State of California and intervenors. With him on the briefs were Edward W. O'Neill and Arocles Aguilar for petitioner, Randall R. Morrow for intervenor Southern California Gas Co., Bruce A. Connell for intervenor Conoco, Inc.

Steve Stojic, argued the cause for petitioner Transwestern Pipeline Co., with him on the briefs were Frank X. Kelly, Joseph R. Hartsoe and Maria K. Pavlou.

Joel M. Cockrell, Atty., F.E.R.C., argued the cause for respondent, with him on the briefs were Jerome M. Feit, Sol., and Joseph S. Davies, Jr., Deputy Sol., F.E.R.C.

Martin J. Bregman entered an appearance for intervenor Western Resources, Inc. Gregory Grady, John H. Cary and Lois M. Henry entered appearances for intervenor Williams Natural Gas Co. David W. Anderson and Patrick G. Golden entered appearances for intervenor Pacific Gas and Elec. Co. F. Nan Todd Wagoner and Richard J. Kruse, Jr. entered appearances for intervenor Texas Eastern Transmission Corp.

Before: WILLIAMS, GINSBURG and RANDOLPH, Circuit Judges.

STEPHEN F. WILLIAMS, Circuit Judge:


On May 11, 1988 the Federal Energy Regulatory Commission approved a certificate filed by Transwestern Pipeline Company under which Transwestern would switch its method for recovering its costs of gas sold from a Purchased Gas Adjustment ("PGA") to a Gas Inventory Charge. The certificate provided among other things that if both of Transwestern's customers nominated purchases of zero under the new certificate, Transwestern could bill them for any gas sales costs it had not recovered under the PGA system. Under the Commission's accounting regulations, these costs had been collected in Transwestern's "Account No. 191". By the time Transwestern's PGA system ended on September 30, 1989, both customers had in fact nominated zero purchases under the new system, thus triggering liability under the so-called "direct bill". In Transwestern Pipeline Co. v. FERC, 897 F.2d 570 (D.C. Cir. 1990) ("Transwestern I "), we held that this direct bill of the final Account No. 191 balance violated the filed rate doctrine insofar as it included gas costs that had accrued before the Commission's approval of the certificate on May 11, 1988. Id. at 576-81. Accordingly we remanded the case to FERC for it to "determine what portion of the Account No. 191 deficiency accrued" after May 11, 1988. Id. at 581.

The Commission thus set out on remand to calculate any amounts in Transwestern's Account No. 191 on September 30, 1989 that could not be billed under our decision in Transwestern I. This involved two operations. First, the Commission had to remove from the September 30 balance any PGA costs "accrued" before May 11, 1988. In this process the Commission excluded about $6.4 million (plus interest) for amounts Transwestern had paid after May 11 to settle pricing disputes with suppliers related to gas purchased before May 11, as well as other adjustments to Account No. 191 made after May 11 but related to pre-May 11 purchases. Transwestern Pipeline Co., 64 FERC p 61,217 at 62,622-25 (1993); 66 FERC p 61,350 at 62,171-74 (1994). This reduced the September 30, 1989 deficiency to about $34.9 million. See 64 FERC at 62,628. Transwestern objects to these exclusions, claiming that the costs the Commission removed were eligible for recovery under our order in Transwestern I (even though they related to pre-May 11 purchases) because these amounts had been "booked" into the PGA accounts after May 11, 1988.

Second, the Commission had to choose some metric for matching amounts Transwestern had recovered in its PGA between May 11, 1988 and September 30, 1989 with Transwestern's accumulated costs; the more Transwestern's PGA collections in that period were allocable to its May 11, 1988 deficiency in Account No. 191, the less the outstanding September 30, 1989 balance would be attributable to pre-May 11 sales. The Commission chose to adopt a FIFO rule--first in, first out--under which each dollar Transwestern recovered through its PGA after May 11, 1988 would retire the oldest dollar in the account as of that date. The Commission found that Transwestern's May 11, 1989 Account No. 191 balance was about $38.7 million, and its collections between May 11, 1988 and September 30, 1989 were about $38.2 million. Applying FIFO, the Commission removed the difference, about $500,000, from the $34.9 million that was left in Account No. 191 after the operation described above, and allowed Transwestern to direct bill the resulting sum, about $34.4 million. The customers--the California Public Utility Commission, as representative of the ultimate customers, supported by several intervenors--object that this latter exclusion was far too small. Because of the mechanics of Account No. 191 recovery as previously carried out under the Commission's regulations (mechanics that involve a complex array of sub-accounts that we will describe below), they say that FIFO misdescribes the historic process of PGA recovery. Using the historic procedure, they say, the Commission should have found that $8.4 million of pre-May 11 deficiency could not have been recovered as of September 30, 1989; thus $8.4 million, rather than $500,000, should have been subtracted from the $34.9 million to compute the sum that could lawfully be direct billed.

Except for a portion not yet final (which we therefore do not review), the Commission's ultimate order, Transwestern Pipeline Co., 66 FERC p 61,350 (1994), is consistent with our mandate on remand, and not otherwise unlawful. Accordingly, we deny the petitions.

The Accrual Issue

We first address Transwestern's claim that sums "booked" to Account No. 191 after May 11, 1988 but relating to pre-May 11 gas purchases should be classified as having "accrued" after May 11 as we used the term in Transwestern I. The Commission found, "in light of the court's overall decision and the court's view of the filed rate doctrine", that "accrued" embodies a general principle of matching costs with the purchase of gas to which those costs relate. 64 FERC p 61,217 at 62,623-24 (1993). We find no error in the Commission's choice.

As an accounting concept, the core purpose of accrual is appropriate matching. In the usual context, the goal is to produce an accurate statement of net income, so that expenses associated with specific revenue (e.g., cost of goods sold) are reflected in the books in the same period where the specific revenue is reflected, regardless of when the firm actually paid the expenses. See, e.g., Glenn L. Johnson & James A. Gentry, Jr., Finney & Miller's Principles of Accounting 104-05 (8th ed. 1980).

The goal of the computation for which we remanded to the Commission, of course, was not to derive an accurate statement of Transwestern's income. But our repeated use of the term "accrued" obviously reflected the idea that the task before the Commission was not a mechanical one, and might well require adjustments to the figures produced by the Commission's historic methods for making adjustments to Account No. 191. Our decision in Transwestern I, dividing the permissible from the impermissible portions of the direct bill, was explicitly based on the notice afforded by the Commission's May 11, 1988 approval of Transwestern's certificate. See 897 F.2d at 577-80. And although we recognized that pipeline customers often have very limited options in responding to notice of an applicable rate, id. at 579, we still insisted on notice as a requirement for the application of a rate consistent with the filed rate doctrine. As a general matter, the customer seems likely to be most concerned with notice of gas costs at the time of specific decisions to take gas: if at the time of any such decision the pipeline was buying gas at higher prices than the price reflected in its current cost of gas sold, the customer would know that down the pike there would be an upward adjustment to cover the discrepancy. Moreover, matching of this sort is a normal aspect of application of the filed rate doctrine. Thus, in Associated Gas Distributors v. FERC, 893 F.2d 349 (D.C. Cir. 1989), we found a violation of the filed rate doctrine in a direct bill that forced customers "to pay a surcharge, over and above the rates on file at the time of sale, for gas they had already purchased." Id. at 355-56 (citation omitted). In reaching that result we pointed out that it was necessary under the filed rate doctrine "to identify the purchase decisions to which the costs are attached." Id. at 355. See also Public Utilities Comm'n of California v. FERC, 988 F.2d 154, 160 (D.C. Cir. 1993).

Accordingly, we think the Commission reasonably sought to fulfill both the letter and the spirit of Transwestern I when it excluded the costs at issue here--the pricing dispute settlement costs, prior period adjustments, and associated interest--on the ground that they "were incurred in connection with Transwestern's ... service before [May 11, 1988], not its service after that date.... Customers' takes after May 11, 1988, had nothing to do with the costs." 64 FERC at 62,624.

Transwestern makes much of the fact that we equated the word "incurred" with the word "accrued" in footnote 7 of the Transwestern I opinion, 897 F.2d at 580 n. 7 ("Given our conclusion that the filed rate doctrine is no bar to Transwestern's collection of unrecovered purchased gas costs incurred after May 11, 1988, this argument only applies to costs in Account No. 191 which were incurred prior to that date.") (emphasis added) . Transwestern's basic theory is that because in some contexts "incurred" means "booked", the fact that these costs were not booked to Account No. 191 until after May 11 (and under Commission rules, could not have been booked until after that date), meant that we had endorsed a principle that required classifying these sums as post-May 11. But "incurred" seems to us quite consistent with a focus on the gas purchase giving rise to the liability, even though the liability may only later have been reduced to numbers firm enough for posting to Account No. 191.

* * *

After resolving this issue against Transwestern, the Commission stated that it remained "open" whether Transwestern could seek to recover these sums "through a charge that does not constitute a surcharge for the past sales service". 64 FERC at 62,630. Accordingly, it said it would "permit Transwestern to propose a prospective charge to recover these costs from its current open-access transportation customers", noting that it would "consider issues concerning the filed rate doctrine and the rule against retroactive ratemaking in the context of the specific mechanism proposed by Transwestern". Id. at 62,631. In fact, Transwestern has filed a tariff to implement such a charge, the Commission has given it initial approval, see Transwestern Pipeline Co., 67 FERC p 61,237 (1994), and petitions for rehearing that decision are pending. The customers here claim that the Commission's prior statements on the issue, made in the context of this proceeding, constitute a final decision approving such a charge and are therefore reviewable at this time. We think not.

Although Sec. 19(b) of the Natural Gas Act allows " [a]ny party ... aggrieved by an order issued by the Commission" to seek review before us, 15 U.S.C. § 717r(b) (1988), we have long held that we have jurisdiction to review only final orders of the Commission. See, e.g., Public Utilities Comm'n of California v. FERC, 894 F.2d 1372, 1376-77 (D.C. Cir. 1990). An order is considered "final" when it "imposes an obligation, denies a right, or fixes some legal relationship, usually at the consummation of an administrative process." State of Alaska v. FERC, 980 F.2d 761, 763 (D.C. Cir. 1992) (citations omitted). As the disputed remarks explicitly left open the legal issues raised by the proposed charge, it seems fantastic for the customers to argue that they satisfy finality requirements.

This drives the customers to rely on the special significance of our power to enforce our mandate. In City of Cleveland v. Federal Power Comm'n, 561 F.2d 344 (D.C. Cir. 1977), we intervened to guide a Commission investigation and adjudication that we had mandated in a prior opinion, even though the investigation was pending and had yielded no final order in the usual sense of the word. So here, the customers argue, the Commission's allusion to the possibility of alternative modes of recovery strays from the mandate and requires similar corrective guidance. Thus, on the customers' analysis, when Transwestern I ruled out recovery of the pre-May 11 costs by a direct bill it necessarily also ruled out recovery by imposition of a charge on users of transportation service provided long after May 11, 1988 and completely unrelated to Transwestern's pre-May 11 gas costs.

The judicial intervention in City of Cleveland protected the parties from conducting a truncated inquiry that, as the court saw it, would simply have to be redone on a broader basis in order to comply with the mandate. The entire exercise was being carried on at the direction of the court. Here, instead, we have ongoing Commission consideration of a proposal that Transwestern would have been free to make had the Commission never uttered its allusion to the possibility that a transportation surcharge might prove distinguishable, in terms of the filed rate doctrine and the rule against retroactive ratemaking, from the forbidden direct bill. Because the currently ongoing Commission activity is therefore essentially independent of our mandate, and because resolution of such issues may turn on technical matters with which judges have little familiarity, we do not think this is an occasion for us to intrude on the Commission's pending efforts to resolve the issue.

The Accounting Issue

The customers claim that the Commission erred in using FIFO to decide what part of the PGA charges collected between May 11, 1988 and September 30, 1989 could properly be attributed to pre-May 11 PGA costs. (Recall that under FIFO the Commission found that $38.2 million of the $38.7 million collected were attributable to such pre-May 11 costs.) The customers argue that had the Commission applied its own PGA regulations to this issue, it would have had to find that $8.4 million in pre-May 11, 1988 costs were never recovered under the regular PGA system. Thus, they say, the Commission's use of FIFO effectively allowed Transwestern to direct bill $7.9 million in pre-May 11, 1988 costs, in violation of this court's mandate in Transwestern I, the filed rate doctrine, and the rule against retroactive rulemaking.

We start with a description of FERC's standard PGA accounting, as the parties have described it to us. Originally, pipelines had included in their commodity charge a sum for gas costs, per unit of gas taken, based on a projection of past experience into the future. But because of the volatility of and rapid increases in wellhead rates, the Commission in the early 1970s devised the PGA system as a means of catch-up (or, in the case of declining prices, restoration to customers). The difference between actual and projected costs over an initial accounting period would be accumulated in a subaccount ("A") of Account No. 191. At the end of the first period, a new subaccount ("B") would be opened to accumulate the next period's costs. Beginning four months after the end of the first period, the unrecovered balance in "A" would be recovered through a commodity rate surcharge (the "purchased gas adjustment"), imposed over a predetermined amortization period. When the amortization period ended, "A" would be closed, "B" would cease its accumulations and would queue up to start the recovery process, and any unrecovered balance from "A" would be transferred to become the initial balance of a new subaccount, "C". "B" would then be recovered while "C" accumulated, "C" recovered while "D" accumulated, and so on, ad infinitum. These leapfrogging subaccounts were designed to be perpetual, and the regulations governing them never in any way addressed the specific issue of what to do in the event gas sales came to a halt. Although no party vouchsafes any reason for these accounting devices, they appear aimed simply at achieving a certain orderliness; at any moment in time, one subaccount is being used to reflect new underages as they occur, one to reflect recoveries of past differentials, and the other two are in a state of transition (and, presumably, computation).

As of May 11, 1988, cost differentials in Transwestern's Account No. 191 had accumulated in subaccounts A, B, and C. "A" then ran a full course of amortization and was actually overamortized; it thus carried a surplus of $338,388 into "C". The latter, "C", was actually accumulating costs on May 11, 1988; about 14.4% of its total ($3.68 million of $25.5 million) accumulated before May 11. The customers apparently concede that the surplus should be counted entirely against the $3.68 million, leaving $3.3 million in pre-May 11 costs before amortization. "C" was in the process of being amortized on September 30, 1989; only 25% of its costs had been recovered as of that date. The customers argue for pro rata amortization of costs, saying that under the regulations about $2.4 million of the $3.3 million had not been amortized at the end of the PGA on September 30, 1989. (The customers do not, incidentally, explain where they get the idea of pro rata amortization; they point to no prior practice governing the order in which fungible amounts within a subaccount would be considered amortized, and indeed we have no reason to think the Commission ever had occasion to address the issue.)

"B" 's amortization also ran for its full period, but its balance was not fully amortized, leaving a deficit of almost $6 million that was carried into "D". As of September 30, 1989 this sum was languishing in "D", which was then still busy accumulating costs; it was never amortized at all. Thus in the customers' view, this $6 million, indisputably attributable to pre-May 11 purchases, could not ever have been collected in the May 1988-September 1989 period.

Accordingly, the customers argued before the Commission that as of September 30, 1989 subaccounts "C" and "D" contained unamortized pre-May 11 costs totalling $8.4 million. This $8.4 million should have been removed from the adjusted September 30, 1989 balance of $34.9 million in order to assure that the direct bill recovered only post-May 11 PGA costs.

The Commission, however, decided that the pre-existing sub-account system was "unsuited" to the task we charged it with in Transwestern I, because the subaccounts were "premised on an open-ended PGA, in which there is little significance to the matter of cost recovery as to whether the effective PGA rates are recovering past or more current cost accruals. Costs not recovered in one period are always recoverable in a later period." 64 FERC at 62,627.

Instead, the Commission decided to "apply the common-sense business practice of treating oldest bills as being recovered first, the first-in-first-out (FIFO) procedure". Id. It found FIFO "consistent with the general thrust of the PGA surcharges" and held that, in light of its own legal error, applying FIFO was "the best accommodation we can make given the problems resulting from after-the-fact shifting from one system that reflects the recovery of past unrecovered costs (the PGA), to another system that, based on a later court decision, strictly prohibits recovery of costs related to the period before notice of the new procedure (the direct billing procedure)." Id. at 62,627-28.

The customers object that the Commission's use of FIFO rather than the pre-existing regulatory system itself violates the filed rate doctrine and the rule against retroactive rulemaking because it retroactively subjects them to greater PGA liability than they had reason to anticipate when they were making purchases between May 11, 1988 and September 30, 1989. Moreover, they say, they would have made different purchasing decisions had they known of this switch.

It is quite true that the Commission's adoption of the FIFO rule did not occur until August 6, 1993, in the course of the remand proceedings, long after the end of the key period of May 11, 1988 to September 30, 1989. See 64 FERC at 62,625-28. But the customers overlook the perspective from which their expectations must be examined. On May 11, 1988 the Commission accepted Transwestern's proposed direct bill. Had that been fully legal, it would, of course, have constituted the basis for any expectations. "The expectations of those who act in anticipation of the right rate are protected, and they would seem presumptively the most deserving." Transcontinental Gas Pipe Line Corp. v. FERC, 54 F.3d 893, 899 (D.C. Cir. 1995) ("Transco"). But the direct bill approved May 11, 1988 proved to be only partly lawful--it was unlawful as to PGA costs already accrued. Thus any customer forming expectations as to the consequences of its purchase decisions would have to estimate how the Commission would (lawfully) compute the permissible direct bill--i.e., how to break out pre-May 11 from post-May 11 costs, and how to attribute PGA sums paid in the period running up to September 30, 1989. Unless there was some other constraint preventing the Commission from using FIFO to answer the latter question, its doing so could defeat no legitimate customer expectations. The Commission's approval of the direct bill, coupled with the legal challenge, put all customers on notice that they would be subject to as much of the direct bill as ultimately proved lawful. It thus changed "what would be purely retroactive ratemaking into a functionally prospective process by placing the relevant audience on notice at the outset that the rates being promulgated are provisional only and subject to later revision." Public Utilities Comm'n of California v. FERC, 988 F.2d 154, 164 (D.C. Cir. 1993); Transco, 54 F.3d at 898.

The customers claim to find such an external constraint in Public Utilities Comm'n of California v. FERC, 894 F.2d 1372 (D.C. Cir. 1990), where we held that (in considering whether a rate violates the rule against retroactive ratemaking) we must "take as given, as the baseline for reasoning, the accounting system that the Commission happened to employ for regulatory purposes in the period of historic costing ... for to do otherwise would open the door to endless retroactive ratemaking." Id. at 1382. But our reasoning there responded to a claim by the customers that the pipeline would enjoy a "windfall" unless tax savings that it had accumulated under the Commission's "normalization" method of accounting for special tax advantages on gas sold before the pipeline adopted Natural Gas Policy Act pricing were used to reduce its price on sales after its adoption of NGPA pricing. Id. We based our response on the grounds that the Commission had previously and lawfully found the pre-NGPA rate "just and reasonable", and that it could not justify a lower post-NGPA rate on the basis of another, inconsistent accounting method. Cf. Town of Norwood v. FERC, 53 F.3d 377, 382 (D.C. Cir. 1995) (contrasting impermissible use of surcharge to recover from future customers for shortfalls in prior rates, with permissible use of surcharge to collect from future customers costs of a type that would have been collected from future customers under old system and that, under new system, would have been collected from prior customers). Here, however, there is no inconsistency between the Commission's use of its subaccounts as to pre-May 11 sales and FIFO for post-May 11 sales. That the customers only discovered the application of FIFO to the post-May 11 sales after they made their purchases was due to the time consumed in working out the legal limits on the direct bill approved May 11. For the reasons already stated, that approval and the ensuing dispute undermine any claim that the delay in the unfolding of precise information violated the filed rate doctrine or the rule against retroactive ratemaking.

Finally, we see no inconsistency between the use of FIFO and our mandate in Transwestern I. Indeed, the accrual principle adopted in that decision strongly suggests use of FIFO rather than the pre-existing system, under which PGA costs for relatively new gas were often collected before PGAs on older gas, for no apparent reason other than administrative convenience.

* * *

In closing, we note that there is a certain schizophrenia in the positions of the parties on both the accrual and the accounting issues. As to "accrual" Transwestern would have been better off under the PGA system; as to the accounting problem, the customers would have been. Thus, as petitioners, both sides argue that the Commission's fiddling with its own accounting rules is arbitrary and capricious. As intervenors, however, seeking to back up the Commission's rejection of their adversaries' claims, the two sides turn around and argue that because the PGA system did not establish rules governing its own demise, the Commission's solution is a permissible and reasonable interpretation of our mandate.

In both cases, we agree with intervenors (and FERC) in holding that in the particular circumstances presented by this case the Commission could reasonably find that the PGA rules were irrelevant to the questions posed by the transition. Use of the matching principle and FIFO were completely reasonable methods of complying with our direction to limit the direct bill to PGAs accrued after May 11, 1988.

Accordingly, the petitions for review are

Denied.