This petition for review takes us once again into the labyrinthine depths of ratemaking regulation in the electric utility industry. Specifically, we are called upon to determine whether Kentucky Utilities Company, an electric utility subject to the jurisdiction of the Federal Energy Regulatory Commission (FERC or the Commission), may include in its rate base capitalized carrying charges on investment in a new power plant to the extent that such carrying charges were incurred after the plant went into service but prior to the effective date of the new rates. In the Commission's Initial Decision, the Administrative Law Judge (ALJ) held that such charges were not recoverable. The Commission affirmed the ALJ's decision without opinion. After its petition for rehearing was summarily denied, Kentucky Utilities Company filed a petition for review in this court. For the reasons that follow, we deny the petition.

* Kentucky Utilities Company (KU) is an investor-owned electric utility. It generates and supplies electricity to a number of retail and wholesale customers; among the latter are the intervenors, various small communities or entities served by KU, referred to collectively as the Kentucky Municipalities. At a cost of over $250 million, KU constructed its Ghent 3 generating plant, a 500 megawatt coal-fired unit. The plant was placed in service on May 31, 1981; as of that date, just over $28 million of the construction cost represented accumulated "AFUDC."1 AFUDC, which is shorthand for "allowance for funds used during construction," is a method of deferring, in a capital account, costs associated with plants under construction for inclusion in a utility's rate base once the plant is put into service.2 In March 1981, shortly before the Ghent 3 plant was to go into service, KU submitted to FERC a revised schedule for its wholesale rates. The rate filing, designated WPS-81, carried a proposed effective date of June 1, 1981. The new rates reflected, for the first time, inclusion of the Ghent 3 unit in KU's rate base.

Had FERC allowed KU's proposed new rates to become effective on the date sought by the utility, the present case presumably would not have arisen; instead, there would have been only a one-day delay between the time the Ghent 3 facility went into service (May 31) and the time the new rates reflecting the new plant became effective (June 1). FERC's initial examination of the filing, however, led the Commission to conclude that KU's proposed rate increase was likely to be more than ten percent in excess of the "just and reasonable" amount permitted by the Federal Power Act ("the Act"), 16 U.S.C. Sec. 791a et seq. (1975), and applicable Commission regulations. Acting pursuant to its authority under section 205(e) of the Act, 16 U.S.C. Sec. 824d(e), FERC suspended the proposed rates for the maximum period allowable, namely five months. In consequence, instead of becoming effective on June 1, 1981, the new rates did not go into effect until November 21, 1981. The result of this FERC-imposed suspension period was to preclude KU from enjoying, as to its jurisdictional rate base, a return on its investment in the completed Ghent 3 unit for that five-month period.

The present case arises from KU's effort, in a later rate filing, to recoup the AFUDC it accrued during the five-month period when its earlier WPS-81 filing was suspended. On July 23, 1982, KU filed with FERC a proposed revision of its wholesale rates. Included within the jurisdictional rate base was a sum--$1,568,000--representing the AFUDC claimed by KU to have accrued during the earlier suspension period. Once again, the Commission suspended the proposed new rates, which eventually became effective in a two-step process on September 23, 1982 and February 22, 1983. Certain of KU's wholesale customers, including the Kentucky Municipalities, challenged a number of items contained in the new filing, but all issues save for the present one were resolved in a compromise settlement.

The remaining issue--KU's claimed right to accrue AFUDC after the Ghent 3 plant had gone into service--was litigated before an Administrative Law Judge. In August 1983, the ALJ issued an Initial Decision, rejecting KU's contentions that it should be allowed to recover the "lost" AFUDC. The ALJ set forth three legal grounds on which KU's claim, in her view, failed. First, the ALJ concluded that no AFUDC accrual could be permitted for ratemaking purposes once a plant leaves the status of construction work in progress (CWIP) and becomes operational. See supra n. 2. Under this approach, a utility irrevocably loses the right to accrue AFUDC on the date the new facility actually enters service. Second, in the ALJ's view, the cost of service not recovered during a suspension period simply could not, under settled ratemaking principles, be recovered in a subsequent rate filing. Finally, the ALJ concluded that, as a general rule, the Commission adheres to its accounting principles in ratemaking, and that those principles preclude AFUDC accrual for post-CWIP periods. Finding KU's petition without support in law, the ALJ treated it as a petition for relief in equity, but denied relief on the equitable ground that KU must bear the consequences of its own decisions with respect to the timing and magnitude of its rate increase request. Had KU's rate increase been less ambitious, the ALJ reasoned, the Commission would not have been moved to invoke the full extent of its suspension powers.

We agree with the branch of the ALJ's holding, affirmed without opinion by the Commission, that the accrual of AFUDC ceases when a new plant comes into service. In consequence, we need not and do not decide whether the ALJ was correct in holding alternatively that "[t]he cost of service not recovered during a suspension period cannot be recouped in a subsequent filing." Initial Decision at 7, Joint Appendix at 11.3 We do, however, address the third ground of the agency's decision, namely the propriety of the Commission's adherence to certain accounting principles in the setting of ratemaking, to the extent that FERC's decision in this respect bears upon our determination that as a fundamental principle of ratemaking, accrual of AFUDC ceases as soon as the new plant becomes operational.

We turn first to the question whether the Commission was correct in holding that, pursuant to longstanding Commission policy, accrual of AFUDC must stop when a new plant becomes used and useful.4 Following consideration of that issue, we will examine the question whether, under the circumstances present in this case, it was appropriate for the Commission to adhere, in a ratemaking context, to its general principles of accounting.

* No less than fifty years of agency precedent provide support for the Commission's refusal to permit KU to continue to accrue AFUDC once its Ghent 3 unit became operational. In a case of considerable vintage, FERC's predecessor, the Federal Power Commission (FPC), embraced the principle that once a new plant or facility has commenced operation, carrying charges on that new investment cannot be charged to a capital account. Chelan Electric Co., 1 F.P.C. 91, 97 (1933). Two years later, in Safe Harbor Water Power Corp., Licensee, 1 F.P.C. 230, 249-51, 254 (1935), the Commission held that costs associated with the construction of a new plant are to be computed up to the date the plant goes into commercial operation.

If the FPC somehow failed to make this rule clear in Chelan and Safe Harbor,5 the Supreme Court subsequently explained, more broadly, that an inability to obtain a return on investment through operating revenues does not provide justification for capitalizing the loss for inclusion in a later rate base. See FPC v. Natural Gas Pipeline Co., 315 U.S. 575, 590, 62 S. Ct. 736, 745, 86 L. Ed. 1037 (1942). The rationale for the Supreme Court's conclusion--and that of the Commission here--is that the risk of failure to earn a return on investment remains at all times on the utility and its investors, not on the ratepayers. This risk is, indeed, built into the procedures established by the Commission for processing requests for new and increased rate bases. This bedrock principle of risk allocation was restated by the Court in FPC v. Tennessee Gas Transmission Co., 371 U.S. 145, 83 S. Ct. 211, 9 L. Ed. 2d 199 (1962), where the Court observed that a utility6 "initiating an increase in rates under Sec. 4(d) assumes the hazards involved in that procedure. It bears the burden of establishing its rate schedule as being 'just and reasonable.' " 371 U.S. at 152, 83 S.Ct. at 215.

Undaunted by all this, KU argues that the Commission's refusal to allow it, in effect, to "recoup its losses by making retroactive the higher rate subsequently allowed," Tennessee Gas, supra, 371 U.S. at 153, 83 S. Ct. at 215, represents a departure from FERC's policy of cost-based rates and that the Commission is therefore required to give a reasoned explanation of what KU deems to be the Commission's novel approach. We cannot, however, fault the Commission's affirmance of the ALJ's holding on this point. The fundamental principle that there can be no capitalization of carrying charges incurred on funds used during construction once construction has ceased is simply too well established to merit extensive elaboration. Ample precedent supports the Commission's position, and that precedent was expressly relied upon by the ALJ.

KU argues further that the Commission's refusal to allow the utility to recover carrying charges incurred during the period of suspension of its WPS-81 filing is contrary to the uniform position of the four state utility commissions that have considered the issue. While KU concedes that FERC is not required to follow even a unanimous practice among the several States, KU argues that the Commission is obliged to explain why the reasoning of those commissions was not followed.

It is elementary, of course, that FERC enjoys exclusive jurisdiction over wholesale utility rates. The Commission is not required, in order to engage in reasoned decision-making, to address every state utility commission "precedent" which is at odds with FERC's practice. This is all the more true when, as here, the "precedent" is plainly distinguishable in fundamental respects. As the Kentucky Municipalities point out, the four state utility commissions in question operate under statutes and procedures quite different from FERC's operative statute and regulations. The Ohio Public Utility Commission (PUC) decision relied upon by KU provides a good illustration of this point. As the Ohio PUC observed:

[T]he accounting treatment utilized for AFUDC by the FERC may be appropriate under their ratemaking procedures but may work a hardship on a utility under Ohio law. Under the system adopted by the FERC, the company, with proper planning, can insure the collection through new rates of carrying costs associated with new facilities going into service. Under those circumstances, cessation of AFUDC at the in-service date would be appropriate. Under Ohio law this is not possible.

In the Matter of Dayton Power & Light Co., Ohio Public Utilities Commission, Case No. 82-858-EL-AAM, Opinion and Order of August 25, 1982, at 5-6. Indeed, the applicable Ohio statute establishes an unavoidable suspension period of either 275 days or until such time as the state commission approves the new rate, whichever is earlier. See 49 Ohio Code Sec. 4909.18 (1983 Supp.).7 Under governing state law, it simply is not within the utility's power to avoid a substantial gap between the time it ceases accruing AFUDC and the time the new rates reflecting the new investment become effective. Furthermore, the Ohio PUC declined to embrace a general rule that accrual of AFUDC may continue during a suspension period; instead the state commission expressly opted in favor of a case-by-case approach to the issue. See Dayton Power, supra at 8. The Ohio Commission carved out an exception for the utility because, in its view, severe consequences would flow from disallowance of the utility's request to recover suspension-period carrying charges in a subsequent rate filing. Id. at 4.

In Northern Indiana Public Service Co., Indiana Public Service Commission, Cause No. 37129, Opinion and Order of April 20, 1983, also relied on by KU, the Indiana PSC permitted a departure from the Uniform System of Accounts, allowing accrual of AFUDC during a suspension period by virtue of the "financial ramifications" which otherwise could flow. Id. at 3 (possible erosion of approximately $6 million in earnings per month, and a further erosion of $1.8 million per month due to commencement of depreciation). Similarly, the decision of the Pennsylvania Public Utility Commission in Petition of Pennsylvania Power & Light Corp., P-82-367, Opinion and Order of July 1, 1982, to allow accrual of AFUDC beyond the new plant's in-service date turned expressly upon the possibility that "the potential financial consequences could be dire" were the Commission to refuse to create an exception. Penn Power at 4. As if more were needed, the Pennsylvania PUC made it clear that it was approving only a procedural modification, reserving for another time approval of a rate change for Penn Power. Id. at 4-5.

Finally, in a decision by the Arkansas Public Service Commission, likewise invoked by KU, the Arkansas PSC was operating under a statute similar to the operative Ohio measure. In Arkansas, as in Ohio, a nine-month delay between a request for a rate increase and the new rate's effective date is ordinarily borne by the applicant unless a "utility demonstrates immediate and impelling necessity, pursuant to Ark.Stat. 73-217." In the Matter of the Application of Arkansas Power & Light Co., Arkansas Public Service Commission, Docket No. 82-335-TF, Advisory Opinion, January 14, 1983, at 10.

Thus, the state decisions invoked by KU are simply not, as KU would have us believe, squarely on point, involving as they do distinct statutory schemes or severe financial consequences. But even more fundamentally, FERC is not, as we noted above, bound to follow a state commission's considered judgment with respect to either accounting or ratemaking. FERC must, rather, follow its own precedents or, alternatively, provide a reasoned explanation for a material departure therefrom. See, e.g., Greater Boston Television Corp. v. FCC, 444 F.2d 841, 851 (D.C.Cir.1970). In our view, the Commission has complied fully with this salutary principle of administrative law, as our review of pertinent FERC precedents demonstrates. In a word, the Commission has in fact followed its own governing precedents. Indeed, in Metropolitan Edison Co., 11 F.E.R.C. p 61,027, amended on reconsideration, 13 F.E.R.C. p 61,142 (1980) ("Met Ed "), the Commission recently reaffirmed its position with respect to forbidding accrual of AFUDC on a plant which had already entered into service.8

KU takes issue with the Commission's decision in Met Ed, arguing that the Commission failed to explain its reason for not adopting the approach taken by a state utility commission, the Pennsylvania PUC. In Met Ed, the Commission observed that it had "no quarrel" with the Pennsylvania PUC's decision to allow continued accrual of AFUDC beyond the new plant's in-service date for ratemaking purposes, although not for accounting purposes. Met Ed, supra, 11 F.E.R.C. p 61,027, at 61,042. KU contends that this statement by the Commission obligated FERC to explain why it was not following an approach in the instant proceeding that it found unobjectionable in Met Ed. KU Brief at 24. Furthermore, KU argues that, by virtue of the "no quarrel" language in Met Ed, the Commission was likewise obligated to explain, in the present case, its reason for not adhering to the position taken by the four state utility commissions discussed previously.

In observing that it had "no quarrel" with the Pennsylvania PUC's approach, however, FERC was, in our view, quite properly recognizing the state commission's undisputed right to establish its own ratemaking principles. At the same time, FERC clearly insisted upon its authority--and indeed its responsibility--to enforce "a uniform accounting system, not subject to variations resulting from ratemaking determinations of state commissions." 11 F.E.R.C. p 61,027, at 61,042. In our reading, the Commission in Met Ed was simply not, as KU would have it, implicitly establishing a ratemaking principle which would allow accrual of AFUDC beyond the date a new plant or facility becomes operational. To the contrary, the Commission stressed that it was addressing at that time only a principle of accounting and that it was not establishing a ratemaking principle with respect to Met Ed or any other utility. Id.9C

As an additional ground for its challenge, KU asserts that to allow accounting rules to govern ratemaking is a classic example of unreasoned decisionmaking. KU Brief at 24, citing the dissent in Florida Gas Transmission Co., 24 FERC p 61,005, at 61,043 (1983). In Florida Gas,10 the Commission majority acknowledged the validity of the dissent's contention there that "[a]ccounting regulations do not make policy judgments; that is the job of Commissioners."11 Id. The Commission did not, however, rely here solely upon an accounting principle; instead, it relied upon a substantive agency policy. As this court observed in Tennessee Gas Pipeline Co. v. FPC, 561 F.2d 955 (D.C.Cir.1977), "[p]etitioners persist in failing to recognize that even though an accounting rule does not govern ratemaking, the underlying policy upon which the rule rests does have direct bearing." 561 F.2d at 957. A principal underpinning of Commission policy with respect to either ratemaking or accounting is the determination in each instance whether the utility and its investors should bear a certain risk or cost or whether that risk or cost should be borne by the utility's ratepayers.

In the context of suspension-period carrying charges, the Commission had already determined that the utility and its investors should bear the risk that the decision regarding when to file a rate increase request will not always result in a coincidence between the plant's in-service date and the date the rates reflecting the new plant become effective. This lack of coincidence is due in part to the Commission's practice of suspending for one day rates that on their face are ten percent or less in excess of a just and reasonable rate12 and for five months rates that on their face are more than ten percent excessive. If the newly filed rate is eventually determined to be excessive, the Commission enjoys authority, of course, to order the utility to refund the excess portion to its ratepayers. See, e.g., Opinion No. 184, 24 F.E.R.C. p 61,158 (1983). The refund procedure is intended to encourage utilities to file rate increase requests as proximate to just and reasonable rates as the utility can determine without, of course, being able to divine fully in advance what the Commission will ultimately decide constitutes a just and reasonable increase. The Commission's adherence to its accounting regulation regarding the cessation of AFUDC at the point a new plant goes into operation is entirely consistent with its substantive policy of allocating the risk of suspension or refund requirements to the utility. The AFUDC accounting principle, applied in this ratemaking context, merely makes costs associated with that risk likewise fall on the utility.

As an additional basis for its contention that the Commission's decision was not the product of reasoned decisionmaking, KU argues that the costs disallowed by the Commission were legitimately incurred and not different, in any important respect, from costs which the Commission has allowed to be allocated to a different time period. As an example of past costs which the Commission has allowed to be included in future rate base, KU points to the Commission's policy allowing utilities to recover for nuclear fuel disposal costs in future rate filings. KU also relies upon FERC's practice of amortizing, in future rates, expenses associated with abandoned plant. Other examples, KU argues, abound of the Commission's permitting an allocation of costs to earlier time periods.

Even aside from the absence of a suspension period question, these illustrations are inapposite. Costs associated with disposal of spent nuclear fuel, for example, are deferred until they are actually incurred.13 But in any event, there simply is, in KU's various examples, no comparable circumstance to FERC's long-standing policy against permitting accrual of AFUDC once a plant goes into service. As its very name suggests, AFUDC is permitted while a facility is under construction; once the construction period ends and service begins, the "allowance for funds used during construction" must also cease.

KU is thus left with only a policy argument, namely, that carrying charges incurred during a suspension period are legitimately incurred costs and that failure to allow the cost to go into rate base contravenes the Commission's cost-based rate policy. This argument, upon analysis, fails. While the Commission ordinarily attempts to establish rates on the basis of cost, cost is not always the decisive factor in determining a just and reasonable rate. At times, considerations such as the allocation of costs between past and present ratepayers--or between ratepayers and utility investors--constitute an overriding concern. Here, the Commission determined that principles of risk and cost allocation did not mandate the imposition on ratepayers of costs incurred as the result of the utility's decision regarding when to file and in what amount, a burden more properly borne by the utility and its investors. We cannot overturn that determination, supported as it is by weighty and consistently applied policy considerations.

In light of well-established principles of utility rate regulation, we discern no error in the Commission's decision. Application of the settled accounting principle that accrual of AFUDC must cease once the new plant is put into service furthers the substantive policy with respect to risk allocation which informs the principles governing the establishment of new wholesale rates.

Denied.