OPINION AS MODIFIED ON REHEARING
Before WRIGHT,* WALD and MIKVA, Circuit Judges.
Opinion for the Court filed by Circuit Judge WALD.
WALD, Circuit Judge:In this action, eight Illinois municipalities 1 seek direct review of an order of the Federal Energy Regulatory Commission (“FERC” or “Commission”) approving a wholesale electric power tariff filed by Central Illinois Public Service Company (“CIP-SCO” or “Company”) under section 205 of the Federal Power Act, 16 U.S.C. § 824d. The petitioners, wholesale customers of CIPSCO, challenge the substantive reasonableness of the tariff, as well as the Commission’s decision to refrain from investigating CIPSCO’s high operation and maintenance costs. For the reasons stated below, we affirm the Commission’s actions.
I. THE FACTS
On December 1,1977, CIPSCO filed tariff sheets with the Commission proposing increases in three wholesale rates: W — 1, which governs sales to electric cooperatives, W-2, which covers sales to full-requirement 2 municipal customers, and W-3, which applies to sales to partial-requirement3 municipal customers. Though the electric cooperatives acquiesced in the new rates,4 several of the W-2 and W-3 customers formed the Illinois Municipal Group (“petitioners” or “Cities”) to protest the proposed rate increases. The Cities petitioned the Commission to intervene in the ratemaking proceeding as CIPSCO’s opponent on December 27, 1977.5 Three days later the Commission issued an order accepting the Company’s W-l filing and permitting it to become effective without suspension on January 1, 1978.6 In the same order, however, the Commission conditionally accepted the Company’s W-2 and W-3 tariffs pending determination of the issues raised by the Cities. The challenged rates went into effect subject to refund on January 2,1978, following a one-day suspension.7
Hearings were not held on the disputed issues until December, 1978.8 The presiding Administrative Law Judge (“ALJ”) issued his Initial Decision, based on the record of 77 physical exhibits and 1,337 pages of transcript as well as briefs and reply briefs submitted by both parties, on July 26, 1979.9 In his 72-page decision, the ALJ made some but not all of the adjustments to the Company’s cost-of-service figures requested by the Cities and the Commission staff *193(“Staff”),10 upheld the Company’s refusal to extend preferential rates contained in certain long-term contracts with one W-2 and one W-3 customer to the remaining customers in those rate categories; found that the Cities’ failed to present a prima facie case that a price squeeze inimical to the Company’s wholesale customers existed; and recommended that the Commission investigate the Company’s abnormally high operation and maintenance costs.
Both parties filed briefs on exceptions to the Initial Decision with the Commission. Six months later, on February 21, 1980, the Commission issued an order summarily affirming the ALJ’s opinion in all but two respects.11 The Commission reversed the ALJ’s recommendation of an investigation into the prudence of CIPSCO’s management 12 and his directive that CIPSCO be required to “flow through” its construction-related interest deductions on a current basis.13 The Cities petitioned for rehearing on twelve issues, including its contract discrimination and price squeeze arguments, eight rate base or rate of return disputes, and the Commission’s refusal to investigate CIP-SCO’s management.14 The Commission denied this rehearing request by taking no action within the statutorily prescribed period.15
The Cities have petitioned this court for review of the Commission’s order. CIPSCO moved for and was granted leave to intervene before this court in support of the Commission’s actions.
II. THE ISSUES
Though the Cities preserved twelve arguments by their petition for rehearing, they present only six in this petition for direct review.16 Of those six, three were argued by one of the petitioners, the village of Rantoul, before this court in a challenge to a FERC order approving a previous CIP-SCO tariff.17 This court affirmed the Commission’s actions in that case in an unpublished memorandum opinion.18 Though that opinion does not bind our decision in *194this case,19 our review of the record in the i;?stant proceeding gives us no cause to question the outcome in Rantoul or its applicability to this case. Accordingly, we summarily affirm the Commission in this case as to those issues decided in Rantoul,20 and discuss in this opinion only those arguments presented here for the first time.
Petitioners raise three novel claims of error. They challenge first the Commission’s decision to deny “price squeeze” relief to the Cities, contending that they established that a price squeeze, as defined by the Commission’s regulations, existed and thus that the Commission’s adoption of the ALJ’s finding that no price squeeze relief was warranted lacked a substantial eviden-tiary basis. Secondly, they claim the Commission erred in failing to normalize the excess generating capacity created by the opening of the new Newton power plant, thereby unreasonably inflating the rate base and corresponding rate assessments. Finally, they argue that the Commission’s reversal of the ALJ’s recommendation that FERC investigate CIPSCO’s management constituted reversible error, and they ask this court to direct the Commission to undertake the investigation.
III. ANALYSIS
A. Price Squeeze
The most serious of the Cities’ allegations is that the challenged rates are set too high in comparison to CIPSCO’s retail rates to allow the Cities to compete with CIPSCO for retail, especially high-volume retail,21 customers, and thus that FERC should have reduced those rates to “the lower end of the range of reasonableness.”22 The Cities *195trace FERC’s decision not to do so to the ALJ’s acceptance of faulty cost of service estimates and his reliance on an inappropriate conceptual framework for analyzing the significance of those estimates. FERC maintains that the ALJ’s analysis of petitioners’ price squeeze complaint was correct, and that the petitioners failed to establish a prima facie case for the existence of a price squeeze.
1. The Elements of a Prime Facie Case
Though a fixture of antitrust law since the 1945 decision in United States v. Aluminum Co. of America (“Alcoa”),23 price squeeze complaints are relatively new to the regulated industries field. FERC routinely disclaimed jurisdiction over such arguments until 1976, when the Supreme Court held that they could be considered when determining whether a rate meets the antidiscrimination requirement of section 205(b) of the Federal Power Act.24 Following that decision, FERC promulgated guidelines for dealing with price squeeze complaints. Those guidelines established the criteria necessary to make out a prima facie price squeeze case:
(1) Specification of the filing utility’s retail rate schedules with which the intervening wholesale customer is unable to compete due to purchased power costs;
(2) A showing that a competitive situation exists in that the wholesale customer competes in the same market as the filing utility;
(3) A showing that the retail rates are lower than the proposed wholesale rates for comparable service;
(4) The wholesale customer’s prospective rate for comparable retail service, i.e., the rate necessary to recover bulk power costs (at the proposed wholesale rate) and distribution costs; and
(5) An indication of the reduction in the wholesale rate necessary to eliminate the price squeeze alleged.25
It should be noted that unlike the ordinary situation where a party who establishes a prima facie case is, as a matter of law, entitled to relief unless the defending party rebuts the case, the establishment of a pri-ma facie price squeeze case means only that enough has been shown to warrant inquiry into the price squeeze allegations to ascertain whether price discrimination exists. It is, the Commission assures us, merely the threshold26 which must be crossed to put the question at issue.
Petitioners claim that they established a prima facie ease by comparing the wholesale rates W-2 and W-3 with the comparable27 retail rates 9 and 9-B in ef-*196feet on January 2, 1978. At that time, the wholesale rates were higher than the retail rates for which petitioners would have been eligible based on their demand characteristics.28 However, on April 14, 1978, a new retail tariff went into effect,29 raising the 9 and 9-B retail rates above the W-2 and W-3 wholesale rates. Whether or not these new rates were annualized,30 the ALJ found that petitioners would have paid more for their electric power purchases during the 1978 calendar year had they paid the retail rather than the wholesale rates. Therefore, he concluded that petitioners failed to establish the third element of a prima facie ease under the Commission’s guidelines.31
Petitioners challenge the ALJ’s reliance on the revised retail tariffs. They claim that the proper comparison is always between the wholesale and retail rates in effect on the date of implementation of the revised wholesale tariff because “CIPSCO had full discretion to wait until after a resolution of its retail rates before filing a wholesale increase.”32 However, we find that the Commission had the discretion to consider the April retail rate increases when evaluating petitioners’ price squeeze allegations. Petitioners overstate the utility’s control over the timing of its rate increases given the inevitable regulatory process delays entailed in finalizing such rates;33 we cannot ignore that lack of control when judging the reasonableness of the ALJ’s actions. Moreover, while annualizing a new rate as opposed to averaging the rates in effect weighted by their periods of incidence may in some cases lead to unacceptably unrepresentative information regarding the existence of a price squeeze,34 the ALJ’s decision to annualize the increases here was not unreasonable in light of the proportion of the year they were actually in effect35 and the fact that the outcome, under the Commission’s guidelines, would have been no different had the ALJ computed the costs to petitioners over the whole year based on the actual date of implementation of the new rates.36
Petitioners further challenge the ALJ’s finding that the wholesale rates were lower than the retail rates by arguing that a “true comparison” of the two rates requires that the wholesale rate be compared to an “adjusted” retail rate, that is, the retail *197rate minus expenses associated with the costs of distribution.37 This argument is a confusing variation38 on petitioners’ main argument: a frontal assault on FERC’s price squeeze guidelines 39 as insufficiently sensitive to price squeeze possibilities. Petitioners point out that if retail and wholesale rates are equal (thus preventing wholesale customers from making a prima facie case under the Commission’s guidelines), the wholesale customers will be unable to compete with the public utility for retail customers because they will inevitably incur some distribution costs, which will require them either to lose money or to increase their retail rate above that of the public utility — the classic price squeeze situation.40 Therefore petitioners advocate utilization of the transfer-price test adhered to in many antitrust cases: “[i]f a vertically integrated entity cannot purchase at its own wholesale rates and still realize a profit by selling at its own retail rates, then it can be concluded that the supplier has overcharged its wholesale customers.” 41 The petitioners in another utility rate case before the Commission described the operation of this test in the electric utility context:
They argue that a utility performs three analytically distinct functions: generation, transmission and distribution. Generation and transmission are company-wide costs and do not vary with the type of customers taking the power. Distribution, on the other hand, is largely a retail cost since the municipals provide their own distribution. Thus, the cities argue that we should compare the wholesale rate for transmission and generation with the average retail rate for serving all customer classes with transmission and generation. Any disparity would be implicitly unjustified since these costs should be uniform company-wide. Consequently, the difference would show the amount of the price squeeze.42
Petitioners attempted to show that under this test a price squeeze exists here by working the transfer price analysis forwards and backwards. First, petitioners subtracted costs they claim are associated only with the distribution function from the revenues that would have been collected from them had petitioners been charged for their electricity purchases under the retail rates 9 and 9-B. The difference, according to petitioners’ calculations, is less than the amounts actually collected from them under the wholesale rates.43 Petitioners claim this shows that CIPSCO charges its wholesale *198customers more for generation and transmission services than it does its retail customers. Secondly, petitioners’ applied the transfer-price test directly to show that CIPSCO could not make a profit if it owned the municipal operations, bought power from itself at its wholesale rates, distributed it through the municipal operations, and charged its customers at its prevailing retail rates.44
The ALJ rejected the Cities’ first attempt at proving that they were being charged more than retailers for generation and transmission services not by rejecting individually the proposed adjustments45 but by relying on a study prepared by the Staff purporting to show the Company’s comparative rate of return between its wholesale and 9 and 9-B retail customers. According to this study, the Company’s profit margin was higher for its retail service.46 Accepting as accurate the study’s assignment of costs,47 the ALJ concluded that the study rebutted petitioners’ argument that the Company’s rate differential was not justified by a difference in costs. He then went on to disparage, rather than refute, the remainder of petitioners’ analysis, calling the transfer price test a
concept ... so new that one doubts if it will ever become old for it would place in the hands of the party alleging discrimination the ability to create the discriminatory premises either willfully or simply through an inefficient municipal operation.48
He concluded with the parting shot:
it is not the policy of Conway to guarantee that all costs incurred in operating a municipal system (efficiently or otherwise) should be “subsidized by its investor-owned supplier.” 49
The price transfer test, as set forth in Alcoa, is a legitimate means of indicating a price squeeze. FERC itself has accepted that test.50 The ALJ here, however, appears to have misunderstood the test. He concluded that “it would place in the hands of the party alleging discrimination the ability to create the discriminatory premises either willfully or simply through an inefficient municipal operation, J.A. 416, thus misinterpreting the test as requiring an inquiry into whether a vertically integrated *199company could realize a profit by purchasing at its own wholesale rates and selling at the retail rates of the party alleging discrimination rather than at its own retail rates. Properly understood, the transfer price test is, in fact, quite similar to the method of analysis employed by the Staff study in this case and relied upon by the ALJ to show the absence of any price discrimination.51 Both depend upon a proper allocation of wholesale and retail costs, and both seek to test whether there is price discrimination52 between wholesale and retail customers.53 Thus, to use the transfer price test to attack the findings of the Staff study seems to us to be not materially different than a direct attack upon the accuracy of the Staff study. Since we have already rejected Cities’ direct attack upon the Staff study,54 and have acknowledged the agency’s discretion to utilize its choice of methodology,55 this indirect attack upon the Staff study must also be rejected. Furthermore, in this case, Cities itself has misapplied the transfer price test by assuming, without adequate proof, that CIPSCO’s retail costs would be the same as the retail costs of its wholesale customers, J.A. 139, an assumption which may explain the ALJ’s misunderstanding of the test he was being asked to apply.56
*200At any rate, having failed to make out a price squeeze case, Cities is not aided by FPC v. Conway, supra. The Supreme Court in Conway held that although the Commission lacked jurisdiction to fix retail rates, its jurisdiction to set wholesale rates allowed it to take retail rates into consideration where wholesale customers were price squeezed by dint either of the utility company’s desire to drive out retail competitors or of the interplay between federally and locally regulated rates.
The Commission must arrive at a rate level deemed by it to be just and reasonable, but in doing so it must consider the tendered allegations that the proposed rates are discriminatory and anticompeti-tive.57
The Supreme Court upheld a decision of this court, per Leventhal, J., and observed that this court was “quite correct” in concluding:
When the costs are fully allocated, both the retail rate and the proposed wholesale rate may fall within a zone of reasonableness, yet create a price squeeze between •themselves. There would, at the very least, be latitude in the FPC to put wholesale rates in the lower range of the zone of reasonableness, without concern that overall results would be impaired, in view of the utility’s own decision to depress certain retail revenues in order to curb the retail competition of its wholesale customers.58
We read that language as permitting the Commission to adjust wholesale rates , within a range of reasonableness to respond either to utility efforts to depress retail rates to meet competition, or to situations where the imperfections of regulation result in an unintended price squeeze. Since ratemaking is an inexact science, even bona fide allocations of costs between wholesale and retail operations may be imperfect or rates of return set by different regulators at the wholesale and retail levels may make it impossible for purchasing wholesalers, no matter how efficient, to compete at the retail level. In such cases, Conway acknowledges the Commission’s discretion to press wholesale rates to the lower end of the zone of reasonableness. The Conway doctrine is not, however, we emphasize, designed to subsidize particular retail competitors. Rather, the doctrine allows the Commission some leeway where it finds that the process of price setting by regulation, and not the superior efficiency of the utility, might result in retail competitors being driven from the market.
Here, the requisite proof that a price squeeze existed was not forthcoming and indeed was refuted by the Staff study. When costs were fully allocated between wholesale and retail operations, CIPSCO’s wholesale profit margin was sufficiently lower than its retail profit margin so that there was no occasion for the Commission to exercise its Conway discretion.
B. Normalization of Generating Capacity
In 1977, CIPSCO opened a new generating plant resulting in a sudden increase in the system’s generating reserves.59 Petitioners contend the Commission erred in failing to “normalize” this increase in generating capacity — that is, to exclude from *201the rate base all costs associated with reserve capacity in excess of 15 percent — despite the fact that they did not allege60 and made no attempt to prove61 that any portion of these costs were occasioned by managerial imprudence. We affirm the Commission’s refusal to normalize these costs.
While we recognize that rate treatment of excess generating capacity in the electric utility industry is controversial in this era of slow demand growth, we are also aware that it is notoriously difficult to project accurately the demand for electric power far enough in advance to coordinate need with the construction of new generating facilities,62 CIPSCO has argued that practical reasons often exist for constructing facilities with a capacity in excess of the recommended norm.63 The ALJ relied here upon the absence of any sort of managerial imprudence to support the inclusion of the total plant in the rate base.64 Regardless of whether that test is — or should be — the exclusive one in deciding whether a utility’s ratepayers must always bear the total costs of excess capacity by having such costs included in the rate base in the year the plant comes on line, we do not find the Commission’s action on the basis of the record in this case to be arbitrary or capricious.65
C. Refusal to Investigate
The ALJ found that CIPSCO’s operating and maintenance costs exceded the industry norm, and recommended that FERC investigate the causes of this phenomenon.66 FERC reversed this recommendation, deciding not to institute an investigation because “[t]he asserted need for the investigation requested by Cities has not been adequately established.”67 Cities contends FERC “ap*202proach [is] . . . substantially in error” as “F.E.R.C. should have routine reporting and evaluating functions to monitor such companies as CIPSCO.... [because the Cities] should not have to launch such an expensive inquiry, which is clearly beyond their resources ...”68 Cities then asks this court to order the Commission to undertake such an investigation.69
Petitioners’ argument is misdirected to this court. It is now settled law that, barring extreme circumstances, an agency’s refusal to institute an investigation is unreviewable.70
Affirmed.
Judge Wright did not participate in the rehearing.
. The eight municipalities are: Cairo, Flora, Metropolis, Roodhouse, Carmi, Bushnell, Marshall and Rantoul.
. Petitioners Cairo, Flora, Metropolis, Rood-house and Carmi are full-requirement customers of CIPSCO.
. Petitioners Bushnell, Marshall and Rantoul purchase only part of their electric power supply from CIPSCO.
. See Central Illinois Public Service Company Initial Decision of the Administrative Law Judge, No. ER78-80, at 2 (July 26, 1979) (“Initial Decision"), Joint Appendix (“J.A.”) 348.
. Initial Decision at 2-3, J.A. 348-49.
. Id at 3, J.A. 349.
. Id. CIPSCO filed an application for rehearing of the December 30th order, which was denied on February 28, 1978. Id.
. Formal and informal settlement negotiations as well as extensive discovery took place during the intervening nine months. Id.
. Id. at 3 — 4, J.A. 349-50. See also note 4 supra.
. CIPSCO stated that the proposed tariffs would increase its revenues by approximately $1,407,000. Initial Decision at 2, J.A. 348. While the Commission staff fully supported CIPSCO’s W-2 rate increase, they found $191,-000 of the W-3 increase to be unjustified. Id. at 4, J.A. 350. CIPSCO concurred in five of the Staffs proposed adjustments; the ALJ found in favor of the Staff on three of the four remaining ones. He also agreed with the Staff’s position on the allowable rate of return, contract discrimination, and price squeeze issues. The only issue on which the Cities prevailed before the ALJ without the support of the Staff was on their request for an investigation into CIP-SCO’s operating efficiency.
. See Opinion No. 75, Central Illinois Public Service Company Opinion and Order Affirming in Part and Modifying in Part Initial Decision, No. ER78-80, Feb. 21, 1980, J.A. 423.
. Opinion No. 75, supra note 11, at 2, J.A. 424.
. Id.
. The Commission now argues that the Cities failed to raise some of these issues in their petition for rehearing, and thus are barred from arguing them before this court, see 16 U.S.C. § 8251(b) (inclusion of issues in petition for rehearing is jurisdictional prerequisite to review in this court). However, our review of the petition for rehearing, see J.A. 427-464, reveals this argument to be groundless. Though the Cities did not argue each point at length in their petition for rehearing, they specifically (1) detailed each decision to which they were objecting and (2) incorporated by reference specific pages from documents previously filed with the Commission in this docket at which such arguments were contained.
. See 16 U.S.C. § 8251(a) (rehearing deemed denied for purposes of judicial review if FERC takes no action on petition within 30 days).
. See Initial Brief of the Petitioners at 4 — 5.
. See Brief for the Petitioner, Village of Rantoul, Illinois v. FERC, No. 80-1632 (D.C.Cir. July 10, 1981) (mem.) at 1-2 (listing arguments). The three arguments in common are: that CIPSCO should be required to offer each customer service at the rate offered to certain customers by virtue of pre-existing contractual requirements; that the Commission erred by ordering tax normalization subject to refund for rate-making purposes pending the outcome of a related rule-making proceeding; and that the Commission erred in failing to take proper account of the self-generating potential of the W-3 customers in allowing CIPSCO’s proposed demand allocations.
. Village of Rantoul, Illinois v. FERC, No. 80-1632 (D.C.Cir. July 10, 1981) (mem.).
. See General Rules of the United States Court of Appeals for the District of Columbia Circuit, tit. II, rule 8(f).
. At oral argument, counsel for the Cities contended that the contract discrimination issue in this case differs from that in Rantoul because, unlike that case, there is no evidence here that the petitioners were offered and rejected the now favorable contract terms. In this case, each city (with the exception of Carmi, Rood-house and Bushnell, which began purchasing power from CIPSCO after it had ceased offering long-term contracts to customers, see J.A. 150-06 (cross-examination of Carl Wall, CIP-SCO vice-president)) had had a contract similar to those still governing the sale of power to the cities of Marshall and Metropolis. Because they had entered into the fixed rate, fixed term contracts at a date earlier than had Marshall and Metropolis, their contracts expired at an earlier date. See J.A. 79 (cross-examination of Bruce Barnes, Jr., Cities’ witness); J.A. 105-06 (cross-examination of Carl Wall). We view this factual dissimilarity as a distinction without a (legal) difference. Petitioners have still failed to demonstrate more than that a rate differential stemming “from the fact that a public utility is free to file for unilateral rate increases with respect to some of its customers while the rate it charges to the remainder are frozen under the Mobile-Sierra doctrine” exists; as we have made clear in previous cases, more is necessary to establish a violation of section 205(b) of the Federal Power Act. See Boroughs of Chambersburg v. FERC, 580 F.2d 573, 578 (D.C.Cir.1978).
The Cities agreed at oral argument that their tax normalization and demand allocation argu-merits were identical in all essential respects to those made in Rantoul. Our review of the record reveals that petitioners in this action, like the petitioner in Rantoul, lack a substantial evidentiary basis for contending that the demand allocation method used failed to credit petitioners’ for their generating capacity. We likewise find without merit their argument that the Commission could not order rates collected subject to refund pending the outcome of a highly disputed rulemaking proceeding. The Commission clearly has the discretion to adopt interim procedures which, if found inappropriate for any reason, will be remedied on a retroactive basis.
. Petitioners claim they are unable to compete with CIPSCO’s 9 and 9-B rates. Retail rate 9 applies to sales to large light and power customers, primarily industrials. To qualify for rate 9-B, customers must guarantee “load,” that is, a fairly constant demand, as well as high demand. See J.A. 60 (additional testimony of Carl Wall).
. Petitioners in fact ask for a variety of remedies, which often conflict with one another, in their initial and reply briefs. See Initial Brief of the Petitioners at 31 (asking for wholesale rate set at 1977 retail rate); id. at 35 (asking for rate calculated on 7.07% return); id. at 37 (asking for rate low enough to allow petitioners to compete for 9 and 9-B customers); Reply Brief for the Petitioners at 11 (wholesale rate should be 1977 retail rate “adjusted for any cost savings and applicable tax law”). We have taken the liberty of rephrasing their argument to correspond with the price squeeze remedy accept*195ed by the Court in FPC v. Conway Corp., 426 U.S. 271, 278-79, 96 S.Ct. 1999, 2004, 48 L.Ed.2d 626 (1976).
. 148 F.2d 416 (2d Cir. 1945). The Commission summarized Alcoa’s mandate as follows:
In Alcoa, the court found that the company was engaged in two distinct activities: The production of “ingot” and the manufacturing of “sheet.” Alcoa had a monopoly over the production of ingot which is used to manufacture sheet, but had competitors for its manufacturing function. In applying the transfer price analysis the court assumed that the cost of manufacturing the sheet was the same for Alcoa and its competitors and found that since the competitor’s total costs were greater than Alcoa’s, Alcoa was charging more than a fair price of the ingot.
Opinion No. 62, Southern California Edison Co. Opinion and Order on Rate Increases, FERC Docket No. ER76-205, Aug. 22, 1979 at 30, J.A. 624.
. See FPC v. Conway Corp., 426 U.S. 271, 96 S.Ct. 1999, 48 L.Ed.2d 626 (1976). Section 205(b) of the Federal Power Act, 16 U.S.C. § 824d(b) provides:
No public utility shall, with respect to any transmission or sale subject to the jurisdiction of the Commission, (1) make or grant any undue preference or advantage to any person or subject any person to any undue prejudice or disadvantage, or (2) maintain any unreasonable difference in rates, charges, service, facilities, or in any other respect, either as between localities or as between classes of service.
. 18 C.F.R. § 2.17.
. Respondent’s Petition For Rehearing and Suggestion For Rehearing En Banc at 8 n.ll.
. Petitioners claim the W-2, W-3, 9 and 9-B rates are comparable because they are calculated for customers with the same load and de*196mand characteristics; if retail rather than wholesale purchasers, the petitioners would qualify for either the 9 or 9-B rate. J.A. 81 (redirect examination of Bruce Barnes, Jr.).
. See Exhibit 1A, J.A. 157; Initial Brief of the Petitioners at 27.
. Initial Decision at 67, J.A. 413.
. Id.; See also Exhibit 1, J.A. 156 (1978 retail rates annualized); Exhibit IB, J.A. 168 (actual retail rates in effect).
. Initial Decision at 67, J.A. 413.
. Initial Brief of the Petitioners at 28.
. Because both state regulatory agencies and FERC can suspend the effectiveness of proposed rates, prescribe temporary rates pending the outcome of investigations, and order refunds at the conclusion of lengthy (often three to four year) investigations, thus retroactively reducing rates, see Note, The Applicability of Antitrust Laws to Price Squeezes in the Electric Utility Industry, 54 St. John’s L.Rev. 103, 109-10 (1979); see also Brief for Intervenor CIPSCO at 37 (detailing procedure of Illinois Commerce Commission), a utility may be unsure of what rates are in effect at any given time until long after the fact. Ruling that they must wait out this lag time before proposing new increases, in this era of rising fuel costs and inflation, would be truly unreasonable.
. In Cities of Batavia v. FPC, 548 F.2d 1056 (D.C.Cir.1977), this court considered and remanded to the Commission for reconsideration a price squeeze complaint arising from a ten-month period in which wholesale prices exceeded comparable retail rates. Obviously, the longer an objectionable price differential exists, the more likely it is that anticompetitive effects will result and the more unwilling we will be to gloss over the fact that such a differential existed by relying on “annualized” rate tables.
. The retail rates were lower than the wholesale ones for one-quarter of the year.
. On average, the retail rates were higher than the comparable wholesale ones during the 1978 calendar year. See Exhibit 1B, J.A. 168; Initial Decision at 67, J.A. 413; Reply by Intervenor CIPSCO to “Response by Illinois Cities to Question Posed by Court at Oral Argument” at 2-4 (citing transcript of testimony of Bruce Barnes, Jr.).
. Initial Brief of the Petitioners at 28.
. See Initial Decision at 68-69, J.A. 414-15 (subtraction of costs from revenues inappropriate failing adjustment for all variables).
. To call 18 C.F.R. § 2.17 a “regulation,” as even FERC contends, is to overstate its importance. It was promulgated as a statement of general policy to guide the Staff in dealing with price squeeze cases in the immediate aftermath of Conway, the Commission has felt free to disregard its strictures when conducting price squeeze proceedings. See, e.g., Opinion No. 62, supra note 23, at 21-22, J.A. 615-16 (guideline “just a statement of the standards the Commission intends to apply in price squeeze cases”; “[ejither the parties or the Commission are free to argue that the standards thus set out are inappropriate”); Opinion No. 53, Boston Edison Co. Opinion and Order Affirming in Part and Modifying and Reversing in Part Initial Decision and Establishing Just and Reasonable Electric Rates, Docket No. E-8855, July 31, 1979 at 23, J.A. 593 (proposing different test for price squeeze allegation when wholesale rate less than comparable retail rate). Indeed, in this proceeding, the ALJ did not stop with a conclusion that petitioners failed to present a prima facie case under the section 2.17 criteria. Rather, giving petitioners the benefit of the doubt, he assumed arguendo that they had presented a prima facie case and evaluated evidence of CIPSCO’s relative rates of return as rebuttal testimony. Initial Decision at 67, J.A. 413 (Staffs presentation an affirmative defense).
. See Hjelmfelt, A Price Squeeze Theory for Implementation of Federal Power Commission v. Conway Corp., 50 U.Colo.L.Rev. 459, 459 n.2 (1979).
. Reply Brief for the Petitioners at 12; see note 23 supra (utilization of transfer price test in Alcoa); Hjelmfelt, supra note 40, at 459 n.2 (listing authorities).
. Opinion No. 62, supra note 23, at 30, J.A. 624 (footnote omitted).
. See Initial Brief of the Petitioners at 28-29; Initial Decision at 66, 68-69, J.A. 412, 414 — 415.
. See Exhibit 56, J.A. 315-19; J.A. 139-40 (cross-examination of James Bachman).
. He chose this strategy although evidence existed on the record from which he could have concluded at least some of the adjustments advocated by the Cities were unwarranted. See, e.g., J.A. 74 (prepared testimony of James Bachman) (distribution factors deducted by Cities’ witness duplicate “without justification” voltage discounts calculated into rates 9 and 9-B); J.A. 59 (prepared testimony of Carl Wall) (transmission loss credit developed by Cities’ witness inaccurate).
. See J.A. 16 (revised testimony of David Hed-berg, Staff witness) (CIPSCO earned 8.65% and 9.89% return on rates W-2 and W-3 and 15.99% and 12.96% return on rates 9 and 9-B respectively).
. Petitioners now challenge the accuracy of the study’s allocation of costs of service between retail and wholesale customers. However, the majority of their challenges stem from their dissatisfaction with the way FERC has defined the cost of wholesale service in this proceeding, a definition which we uphold in its entirety in this appeal. In addition, petitioners claim that the ratio of wholesale to retail costs may have changed in the two years since this study was conducted. However, the Cities have not indicated any reason aside from the passage of time why such ratios might have changed, see Initial Decision at 68, J.A. 414; in the absence of such evidence, we will not hold that the utilization of a two year old study for these purposes is per se unreasonable. Two years is not so long a time that substantial changes can be presumed to have taken place. Finally, petitioners allege that neither the ALJ nor the Staff examined the methodology of the study. This assertion is belied by the record. See Initial Decision at 68, J.A. 414; J.A. 13 (testimony of David Hedberg); J.A. 129-131 (cross-examination of James Bachman); Exhibit 9, J.A. 200-201 (Display A) (letter to Carl Wall from Arthur Anderson & Co. re cost of service study). In view of the total lack of evidence that extrapolation from this study would lead to distorted results, we accept the Commission’s decision to utilize it as a reasonable exercise of its discretion.
. Initial Decision at 70, J.A. 416.
. Id.
. Opinion No. 62, supra note 23, at 31, J.A. 625.
. See Opinion No. 62, Southern California Edison Co. Opinion and Order on Rate Increases, FERC Docket No. ER 76-205, Aug. 22, 1979 at 31, J.A. 625. Cities argues that the Staff study cost of service test was inferior to the price squeeze test because it merely sampled the relative rates of return in wholesale and retail markets. Response By Cities Of Bethany, Et Al. To Petition Of CIPSCO For Rehearing And Suggestion For Rehearing En Banc at 3-4, 7. But see slip op. at n.27. It is true that an exhaustive comparison would probably be more accurate than a survey, however, the survey indicated that there was no need to conduct a full comparison of wholesale and retail rates of return. Had the Staff study indicated otherwise, we assume that either a more exhaustive cost of service study would have been run or possibly a transfer price test would have been conducted.
. See 16 U.S.C. § 824d(b); see also FPC v. Conway Corp., supra.
. Under the price transfer test the vertically integrated company’s costs of retail operations must be added to the product purchased at its wholesale rates to determine profitability, so the usefulness of that test depends upon an accurate identification of retail costs. If the company fails to realize a profit under this test, Indications are that retail costs have been improperly transferred to the wholesale operations (or, less likely, retail operations are inefficient). Under the methodology of the Staff study, wholesale and retail revenues over their respective incurred costs are compared. This method of analysis also depends upon a proper identification of wholesale and retail costs. If, after wholesale and retail costs are fully allocated, the company’s wholesale profit margin is significantly greater than its retail profit margin, indications are that a price discrimination is occuring. Both methods of analysis are acceptable methods of establishing a price squeeze case. See J.A. at 625. Although the results of a cost of service study are phrased in terms of a profit differential, see Memorandum By The Cities of Bethany, Et Al. In Response To The “Memorandum Amicus Curiae” Filed By The Antitrust Division of the Department of Justice at 4, like the price squeeze test, the cost of service study also seeks to determine whether the vertically integrated company maintains artificially high wholesale prices in order to squeeze out its retail competition. If, after costs are fully allocated, it appears that the wholesale rate of return of a vertically integrated company is reasonable and, in fact, lower than retail rate of return, it would appear that wholesale customers are not being price squeezed — indeed, the higher the retail rate of return vis-a-vis the wholesale rate of return, the greater the wholesale customer’s discount.
. See supra at 198 n.47.
. Id. See also San Antonio, Texas Acting By and Through Its City Public Service Board v. United States and Interstate Commerce Commission, 631 F.2d 831, 837 (D.C.Cir.1980), slip op. at twelve (courts reluctant to interfere with agency’s choice of methodology); but see, id. at 841 (when errors in methodology apparent, courts available for relief).
. See Alcoa, 148 F.2d at 437:
The plaintiff’s theory is that “Alcoa” consistently sold ingot at so high a price that the “sheet rollers,” who were forced to buy from it, could not pay the expenses of “rolling” the “sheet” and making a living profit out of the price at which “Alcoa” itself sold “sheet.” To establish this the plaintiff asks us to take “Alcoa’s” costs of “rolling” as a fair measure of its competitors’ costs, and to assume that they had to meet “Alcoa’s” price for all grades of “sheet,” and could not buy ingot elsewhere.
We note that Alcoa’s own costs of rolling were used as a fair measure of its competitor’s costs. *200Here, competitor’s costs were offered as a fair measure of CIPSCO’s costs. Although it may be, as Cities asserts, that its retail costs were lower than CIPSCO’s, there was inadequate evidence in the record to prove that assumption. Petitioners draw our attention to J.A. 136-39. Although the evidence there correctly indicated that if CIPSCO owned each of the Cities’ distribution systems it would pay higher taxes, J.A. 136-38, the hypothetical itself merely assumed that CIPSCO would incur the same costs that Cities incurred.
. FPC v. Conway, supra, 426 U.S. at 279, 96 S.Ct. at 2004.
. FPC v. Conway, 510 F.2d 1264, 1274 (D.C.Cir.1975), aff’d 426 U.S. 271, 96 S.Ct. 1999, 48 L.Ed.2d 626 (1976).
. The addition of the Newton generating plant increased the anticipated reserve capacity for 1978 to 30.2%. J.A. 4 (testimony of Bruce Barnes, Jr.). CIPSCO’s average reserve capacity for the previous ten years had been 10.61%, id., below the Commission’s recommended norm of 15-20%. See Initial Decision at 28, J.A. 375.
. See Initial Decision at 28-29, J.A. 375-76.
. Id.
. See Villages of Chatham and Riverton v. FERC, 662 F.2d 23, 35 (D.C.Cir.1981); Initial Decision at 30, J.A. 376. See generally Treb-ing, Equity, Efficiency, and the Viability of Public Utility Regulation, in Application of Economic Principles in Public Utility Industries 17, 33-34 (W. Sichel & T. Gies, eds. 1981).
. For instance, in times of increasing demand, a utility would have to open new plants each year to maintain a constant reserve capacity. As pointed out by the Commission in another proceeding involving a similar question, it may be cheaper in the long run to take advantage of economies of scale by building larger plants at more infrequent intervals, thus resulting in excess capacity at portions of the building cycle. See Opinion No. 63, supra note 50, J.A. 632-34; see also Initial Decision at 29, J.A. 375.
. The magnitude of the mismatch is not so great that we can conclude that CIPSCO’s management was per se incompetent, or even that CIPSCO committed a gross error in judgment. Moreover, it is unclear to what extent the mismatch will burden the ratepayers with extra costs. The ALJ points out that CIPSCO is expected to sell most of its excess capacity to other utilities, and the revenues “would necessarily redound to the benefit of ratepayers.” See Initial Decision at 31, J.A. 377. One point troubles us, however. Because FERC sets rates in advance, if CIPSCO was expected to sell its excess generating capacity during the 1978 calendar year, the expected revenues should have redounded to the ratepayers’ benefit — that is, they should have been factored into the rates accepted by the Commission. The ALJ’s use of the future tense to describe the crediting of these revenues may have been a mere “slip of the pen.” However, on remand, the Commission should check to ensure that any expected sale revenues were in fact credited against the sums to be raised from the ratepayers.
. We note, however, that some state utility commissions as well as commentators have advanced a form of normalization, not unlike that which petitioners press here, as a more equitable way of sharing the risks of overfore-casting utility demand. This alternative involves charging maintenance and depreciation expenses on the excess capacity as above-the-line operating expenses to be paid by the ratepayers over a period of years until the capital investment has been repaid. Because the excess capacity is not included in the rate base, however, shareholders do not receive any return on their investment. See Trebing, supra note 77. It may be that different rate treatments of excess capacity are not only justified, but necessary, to attain a “just and reasonable” price in other circumstances.
. Initial Decision at 52, J.A. 398.
. Opinion No. 75, supra note 11, at 2, J.A. 424.
. Initial Brief of the Petitioners at 60.
. Reply Brief for the Petitioners at 22.
. See City of Chicago v. United States, 396 U.S. 162, 165, 90 S.Ct. 309, 311, 24 L.Ed.2d 340 (1969); North Carolina Util. Comm. v. FERC, 653 F.2d 655, 669 (D.C.Cir.1981); General Motors v. FERC, 613 F.2d 939, 944 (D.C.Cir.1979). Petitioners do not argue, and we see no support for the proposition, that FERC’s decision not to institute an investigation in this case overstepped a statutory directive. Cf. Nader v. CAB, 657 F.2d 453, 456 (D.C.Cir.1981) (allegation that suspension policy violated statutory directive reviewable). In their reply brief, the Cities argue for the first time that this decision not to investigate is reviewable because “the Commission simply did not consider all the relevant factors in a way which would allow this Court to determine that the agency has actually exercised its discretion.” Reply Brief for the Petitioners at 23. However, the record lends no support to this contention; FERC clearly considered all the relevant factors listed by petitioners when exercising its discretion to refuse to launch an investigation. See Opinion No. 75, supra note 11, at 2, J.A. 424.