Public Service Co. of New Mexico v. Federal Energy Regulatory Commission , 832 F.2d 1201 ( 1987 )

McKAY, Circuit Judge.

Public Service Company of New Mexico (PNM) is an electric utility selling electricity both at retail and at wholesale. PNM’s wholesale sales of electric power in interstate commerce subject it to the jurisdiction of the Federal Energy Regulatory Commission (FERC or Commission) pursuant to the Federal Power Act, 16 U.S.C. §§ 824-828c (1982). The city of Gallup, New Mexico, is a full-requirements wholesale customer of PNM. The present appeals and cross appeals deal with PNM’s' wholesale rates to Gallup established by FERC after extensive hearings before an administrative law judge (AU).¹

I. General Principles

The Federal Power Act empowers the Commission to examine and establish PNM’s wholesale rates and practices to ensure that they comply with the statutory standard of “just and reasonable.” 16 U.S.C. § 824d(a). In determining a just and reasonable rate, the Commission must consider several factors, including operating expenses, depreciation expenses, taxes, and a reasonable return to the utility’s investors. The return on investment is computed by multiplying the rate base— the value (net of depreciation) of the shareholder’s investment in the enterprise which is allocated to public use — by the “overall” rate of return found to be just and reasonable by the Commission.² Although ringing of mathematical precision, the calculation of a just and reasonable rate is less a science than an art.

The Commission’s findings of fact used in arriving at an appropriate rate, if supported by substantial evidence,³ are conclusive under the statute. 16 U.S.C. § 825i (b). The Supreme Court has gone further and stated that if the actual rates set by the Commission, as opposed to the rates’ component parts, are not unjust or unreasonable in their consequences, they may not be overturned. See Permian Basin Area Rate Cases, 390 U.S. 747, 767, 88 S.Ct. 1344, 1360, 20 L.Ed.2d 312 (1968).⁴

Under the statutory standard of “just and reasonable” it is the result reached not the method employed which is controlling. It is not theory but the impact of the rate order which counts. If the total effect of the rate order cannot be said to be unjust and unreasonable, judicial inquiry under the Act is at an end. The fact that the method employed to reach that result may contain infirmities is not then important.

Federal Power Comm’n v. Hope Natural Gas Co., 320 U.S. 591, 602, 64 S.Ct. 281, 287-88, 88 L.Ed. 333 (1944) (citations omitted).

Moreover, pragmatism demands we recognize that no rate is challengeable merely because we believe an iota higher or lower would be more just or more reasonable. The economic and equitable components of a just and reasonable rate are inherently imprecise. Because “neither law nor economics has yet devised generally accepted standards for the evaluation of rate-making orders,” Permian Basin, 390 U.S. at 790, 88 S.Ct. at 1372, no single rate could ever be the indisputably “correct” one in the view of all analysts. Consequently, we are “without authority to set aside any rate selected by the Commission which is within a ‘zone of reasonableness.’ ” Id. at 797, 88 S.Ct. at 1376. We must

determine whether the order may reasonably be expected to maintain financial integrity, attract necessary capital, and fairly compensate investors for the risks they have assumed, and yet provide appropriate protection to the relevant public interest, both existing and foreseeable. The court’s responsibility is not to supplant the Commission’s balance of these interests with one more nearly to its liking, but instead to assure itself that the Commission has given reasoned consideration to each of the pertinent factors.

Id. at 792, 88 S.Ct. at 1373.

Finally, those who challenge the Commission’s order carry the burden of proof as to its unjustness or unreasonableness, and that burden is heavy.⁵ “A presumption of validity ... attaches to each exercise of the Commission’s expertise, and those who would overturn the Commission’s judgment undertake ‘the heavy burden of making a convincing showing that it is invalid because it is unjust and unreasonable in its consequences.’ ” Id. at 767, 88 S.Ct. at 1360 (quoting Hope Natural Gas Co., 320 U.S. at 602, 64 S.Ct. at 288). “Therefore, if the challenger fails to produce evidence that a rate is not just and reasonable, the Commission should be affirmed even if it produces no evidence of justness or reasonableness.” Union Elec. Co. v. FERC, 668 F.2d 389, 393 (8th Cir.1981).

II. Price Squeeze

In Case One, Gallup charged in part that PNM’s proposed wholesale rate increase would result in a “price squeeze.”⁶ The AU divided the proceedings into two phases, a relatively common practice. See, e.g., Boroughs of Ellwood City, Grove City, New Wilmington, Wampum, and Zelienople v. FERC, 731 F.2d 959, 962 (D.C.Cir.1984). In phase I, the ALJ addressed the appropriate rate of return on equity. He considered the allegations of price squeeze in phase II, along with several other issues, and concluded that Gallup failed to establish a prima facie case of price squeeze. See 18 C.F.R. § 2.17(a) (1987).

The Commission affirmed this conclusion but noted that Gallup was held to an “impossible standard.” In evaluating allegations of price squeeze, the two relevant rates to be compared are the newly established just and reasonable (but for price squeeze) wholesale rate and the retail rates in effect at the time the new wholesale rate becomes effective. FERC explained:

Since the price squeeze issue was not phased for determination until all other cost-of-service issues were resolved, the parties could not know at the time of the hearing what the Commission would ultimately approve as just and reasonable (but for price squeeze) wholesale rates. Nor could they know when increased wholesale rates would become effective and what the retail rates would be at that time. Gallup excepts to the judge’s conclusion on the grounds that it was held to an impossible standard, namely, to establish a -prima facie case of price squeeze when both of the relevant rates for comparison were unknown and unknowable at the time of the hearing. ... Until the just and reasonable wholesale rates are approved and implemented, Gallup has no basis for alleging that such rates in relation to certain retail rates will create a price squeeze. Prior to that time Gallup can only indicate its concern that a price squeeze may develop because of the magnitude of PNM’s proposed wholesale rates.

Record, Case One, joint app. vol. 2 at 471.

Accordingly, the Commission kept the issue open, allowing Gallup to file price-squeeze allegations with respect to the then-current retail and the new wholesale rates within thirty days of FERC approval of the new wholesale rates. The new wholesale rates would be provisionally effective, subject to refund should the allegations prove meritorious. The District of Columbia Circuit has approved such a procedure as a reasonable exercise of the Commission’s equitable powers under the statute. See Kansas Cities v. FERC, 723 F.2d 82, 90-96 (D.C.Cir.1983). PNM, however, claims that the Commission abused its discretion in keeping the price squeeze issue alive once Gallup failed to make the requisite prima facie showing.

After PNM made the required compliance filings detailing the newly established wholesale rates, Gallup filed new allegations of price squeeze pursuant to FERC’s order. Within sixty days, FERC issued an opinion, again concluding that these allegations were without merit. As part of the consolidated appeals in Case One, Gallup initially appealed this adverse ruling, but eventually abandoned the issue. See Initial Brief of the City of Gallup at 60-61. But, even if it had not, we conclude that PNM’s complaint is now moot.⁷

III. Late-Payment Charge

PNM’s wholesale rate schedules have historically imposed additional charges in the event of delinquent payment. Gallup’s rate schedule contained PNM’s standard one-percent-per-month late charge. In its filing, PNM left this provision intact without objection from Gallup. The AU nevertheless eliminated the late-payment provision upon Commission staff recommendation, and FERC affirmed. PNM argues that because its rate-increase filing proposed no change in Gallup’s late-payment charge, FERC erred in deleting it without affirmatively finding that it was unjust and unreasonable.

We have held:

By filing the rate increase, a gas company assumes the risk of having to justify its entire rate structure, including integral provisions of that structure which the company does not propose to change. A provision which serves as a charging mechanism is considered an integral part of a rate increase and a gas company’s rate structure.

Colorado Interstate Gas Co. v. FERC, 791 F.2d 803, 807 (10th Cir.1986) (citations omitted), cert. denied, — U.S. -, 107 S.Ct. 907, 93 L.Ed.2d 857 (1987). However, when FERC seeks to impose a change not proposed by the company, the statute provides that the Commission must first find the existing provision unjust or unreasonable. See ANR Pipeline Co. v. FERC, 771 F.2d 507, 513-14 (D.C.Cir.1985) (per cu-riam). That FERC never challenged the provision in prior rate-increase filings is of no moment. Rate making is not a static methodology but rather an ever-evolving art. These principles are equally applicable to the regulation of electric utilities.

In this case, the ALT commented:

Late payment charges generally serve two purposes: to compensate the creditor for the additional costs caused by late payments (such as contacting the debtor in order to collect the account) and to provide an incentive to the debtor to pay on time. In Connecticut Light & Power Co., [55 FPC 1986 (1976),] the Commission held that it would permit the inclusion of late payment charges only where shown to be justified by cost. Specifically declining to accept such provisions for the purpose of penalizing late payors (i.e., providing an incentive to pay on time), the Commission stated:

... we do not believe as a matter of policy that this Commission should assume the functions of a bill collection enforcement agency except in otherwise hopeless situations.

In the instant proceeding, no evidence was introduced supporting the inclusion of the late payment provisions on cost grounds, nor was there any evidence indicating that Gallup or any of PNM’s wholesale customers is chronically delinquent in its payments. In fact, PNM’s witness conceded that the cost of financing the accounts receivable for this customer class does not vary from customer to customer, and that no particular resale customer has needed any incentive to date. Witness Bedford explicitly stated that the Company’s intent regarding the late payment provision was “merely to ensure prompt payment.” As indicated, the Commission has held this to be improper grounds for including a late charge in an electric utility’s rate schedules. Accordingly, Staff’s recommendation that the late payment provision be deleted from PNM’s wholesale rate schedules is adopted as consistent with prevailing Commission precedent.

Record, Case One, joint app. vol. 2 at 399-400 (footnotes omitted). Thus, the AU relied on the Commission’s policy that, as a general matter, late payment charges not justified by cost tend to be unjust and unreasonable. FERC may legitimately articulate its general policy in a judicial proceeding as it did in Connecticut Light & Power, supra, rather than in an administrative rule-making and use such general analysis of an issue in disposing of individual cases. See Kansas Gas & Electric Co. v. FERC, 758 F.2d 713, 719 (D.C.Cir.1985); Florida Power & Light Co. v. FERC, 617 F.2d 809, 816 (D.C.Cir.1980).

We conclude that (1) the AU’s opinion, adopted by the Commission, did rely upon the Commission’s policy that late charges not justified by cost or delinquent payments by a particular customer are improper; (2) the Commission’s policy is reasonable; and (3) its application in this case is supported by substantial evidence provided by PNM’s own witness. We therefore affirm the Commission’s deletion of the late-payment charge in PNM’s wholesale rate schedule for Gallup.⁸

IV. Rate of Return

In Case One, the AU established 12.6 percent to 14.4 percent as the zone of reasonableness for an appropriate rate of return on equity, selecting 13.25 percent as the rate to apply within that zone. In arriving at this number, the AU used in part a discounted cash flow (DCF) analysis, which calculates the current cost of equity capital as the sum of the current dividend yield plus the expected rate of growth in the value of the stock. The Commission staff witness recommended using a growth rate of 4 percent in the DCF calculation, but the AU concluded the record supported a growth rate of 4.5 percent. The AU also considered a comparable earnings analysis in establishing the zone of reasonableness as well as other factors affecting PNM’s ability to attract capital, including PNM’s quality of earnings, need for external financing, market risks, debt/equity ratio, growth rate in sales, and involvement in a major construction program. The Commission raised the rate of return within the zone of reasonableness from 13.25 percent to 14.4 percent because of the impact of the volatile money-market conditions of the preceding year, the rise in the cost of capital in this capital-intensive industry, and PNM’s involvement in a major construction program.

Gallup first argues that the record cannot support a cost of equity greater than 11.75 percent because the growth component of the DCF study should have been between 2 percent and 3 percent rather than the 4.5 percent finally adopted by the AU. Gallup does not challenge the other data and other studies relied upon by the AU.

Determination of the expected growth-rate component of a DCF analysis, which is a forward-looking projection, requires seasoned judgment and entails choices among conflicting assumptions and data. At bottom, Gallup argues that its choices should have been adopted. In so doing, it “loses sight of its burden, arguing the weight of the evidence, by attacking the accuracy of testimony and associated FERC-adopted findings of the AU, rather than showing a lack of evidence to support those findings.” Nepco Mun. Rate Comm’n v. FERC, 668 F.2d 1327, 1340 (D.C.Cir.1981), cert. denied, 457 U.S. 1117, 102 S.Ct. 2928, 73 L.Ed.2d 1329 (1982).⁹ The record satisfies us that the zone of reasonableness established by the AU, after considering PNM’s ability to attract capital and the projected 4.5 percent growth rate, is supported by substantial evidence.

Gallup also challenges the Commission’s action in increasing the rate of return from the 13.25 percent recommended by the AU to 14.4 percent. Because the revised figure was still within the zone of reasonableness, we are without power to question it. See Permian Basin, 390 U.S. at 790, 88 S.Ct. at 1372. We agreed that “one who challenges a rate must show that the result, and not the method employed, is unreasonable.” Union Elec. Co., 668 F.2d at 396. Therefore, we do not address the propriety of the Commission’s consideration of the money-market conditions arising after the AU established an appropriate rate of return.

In Case Two, PNM requested a 14.6 percent rate of return and submitted supporting data and testimony demonstrating its financial risk. The Commission’s staff submitted a DCF analysis and a risk analysis comparing PNM with analogous utilities and recommended a rate of return of 14.8 percent. Gallup introduced no evidence on this issue, and the AU concluded that PNM had satisfied its burden of justifying the 14.6 percent rate of return. The Commission affirmed that rate.

Gallup attacks that rate on two grounds. First, it argues that PNM’s construction costs were imprudently incurred because they resulted in planned excess capacity and therefore could not be considered in the cost of capital computation. The AU reasoned: “If Gallup wished to argue the prudency of PNM’s construction program, it should have raised the point as a rate base issue and supported the claim with competent evidence.” Record, Case Two, joint app. at 215.

The AU’s observation is dispositive. While unnecessary costs may not be included in PNM’s cost of service, prudency of investments is an issue appropriately considered when establishing the rate base— the component to which the rate of return is applied.¹⁰ Construction expenses, whether prudently incurred or not, must be considered in determining the overall rate of return because they undeniably affect the rate of return on common equity.

Second, Gallup argues that PNM’s construction program will benefit only its nonjurisdictional retail customers, not Gallup and other wholesale customers. Consequently, Gallup asserts PNM is required to allocate the increased equity risk associated with the construction program to its retail customers; i.e., Gallup’s rates should not reflect the increased equity risk. FERC and PNM disagree both with the premise that PNM’s wholesale customers will not benefit from the construction program and with Gallup’s argument regarding allocation of equity risks between jurisdictional and nonjurisdictional customers.

The Commission’s pronouncements in this area are admittedly not uniform. Compare Minnesota Power & Light Co., 12 F.E.R.C. (CCH) ¶ 61,264 (1980), aff'd in revelant part sub nom. Cities of Aitkin et al. v. FERC, 704 F.2d 1254, (D.C.Cir.1982) (per curiam) with Order No. 442, Generic Determination of Rate of Return on Common Equity for Public Utilities, FERC Statutes and Regulations, 1130,677 at 30,084-85 (1986). However, we need not enter this morass for it is not our prerogative to require the Commission to use what we perceive to be the “best” methodology. We are to ensure only that the methodology employed was reasonable and produced reasonable rates. See Cities of Bethany et al. v. FERC, 727 F.2d 1131, 1136 (D.C.Cir.), cert. denied, 469 U.S. 917, 105 S.Ct. 293, 83 L.Ed.2d 229 (1984). Gallup’s arguments regarding its preferred methodology of risk allocation between wholesale and retail customers were explicitly presented to both the ALT and FERC and rejected. We do not find it unreasonable that the Commission declined to allocate risks between two classes of customers when determining an appropriate rate of return, and that is our sole inquiry on review. A capital expansion program increases the financial risk to the entire utility; a potential investor in the capital market does not distinguish between a utility’s retail and wholesale sales. FERC’s approval of a 14.6 percent rate of return was based on substantial evidence produced by both PNM and the Commission’s staff and thus is affirmed.

In Case Three, FERC concurred in the AU’s determination that a 15.44 percent rate of return on equity was just and reasonable. PNM challenges this rate as too low to prevent dilution of its stock and thus insufficient “to assure confidence in the financial integrity of the enterprise, so as to maintain its credit and to attract capital.” Hope Natural Gas Co., 320 U.S. at 603, 64 S.Ct. at 288. We note, however, that to raise sufficient capital for its construction program, PNM has already sold enough equity shares below book value to result in a nearly $22 million dilution¹¹ in the equity held by investors and apparently anticipates doing so in the future to meet capital demands. The AU adopted the Commission staff’s recommendation that this practice of dilution be considered when determining the growth-rate projection in the DCF analysis. PNM opposes this decision, claiming that it is contrary to the Hope Natural Gas Co. standard.

We affirm the Commission’s rate of return based on the Commission staff’s DCF analysis. The growth-rate component of a DCF analysis must reflect the investor’s expectation of dilution if it is to be realistic. Moreover, the AU cogently reasoned that if dilution were not factored into the growth-rate component, then the dividend-yield component must be altered to reflect dilution effects.

Dilution is the reality for PNM stock issuances.... To properly gauge the growth expectations of an investor one must examine the historical performance and future expectations of a stock based upon all relevant considerations. No one argues that investors in PNM stock do not expect dilution. To ignore this aspect of investor expectation is to forecast an unreasonable growth rate. PNM argues that to recognize the existence of dilution expectations is to somehow violate “regulatory equity” by keeping utilities from earning a fair rate of re-turn_ This argument is unpersuasive. Regulators should not ignore important market information that relates to the expectations and requirements of investors and substitute an unrealistic model that selects only data that increase rate of return.

Moreover, ... dilution cannot be ignored in the growth rate of the DCF equation without making a corresponding correction in the dividend yield. As [the Commission staff witness] factored out anticipated dilution due to projected rising stock prices, he also correctly lowered the current dividend'yield to reflect a higher current price. Any current dividend yield encompasses investor requirements based upon the Company’s financial condition. If dilution is not expected, the current price would be higher, reflecting the investors [sic] more favorable expectations, and the resultant current dividend yield would be lower. Accordingly, PNM’s objection both fails to reflect the true expectations of investors and fails to make the requisite current price adjustment described above. Therefore, the use of a DCF methodology considering all investor expectations, including dilution, is properly employed in determining a forward looking market oriented rate of return on common equity.

Record, Case Three, joint app. vol. 2 at 260. Finally, the Commission staff witness refuted PNM’s assertion that a 15.44 percent rate of return would perpetuate dilution, testifying instead that such a return should not result in future dilution. See id., joint app. vol. 1 at 209-10. The rate of return adopted was thus reasonable and based on substantial record evidence.

Gallup challenges the 15.44 percent rate of return as too high, arguing that its modified DCF analysis justifies only a 12.-59 percent rate of return. It also challenges FERC’s refusal to reduce the rate due to money-market changes occurring after the close of the record that reduced the cost of equity, claiming such refusal is inconsistent with FERC’s analogous increase in the rate of return in the first case discussed above and is thus arbitrary.

The AU’s reliance on traditional DCF analysis and rejection of Gallup’s “deductive elaboration of the DCF theory” was not unreasonable. Again, the choice of methodology is a matter for the Commission, and the result obtained under its chosen means was reasonable and based on substantial record evidence. Gallup’s other rate-of-return arguments in Case Three, including its challenges to both the capital structure used and the data used, do not dissuade us from this view.

Finally, the Commission’s refusal to take official notice of money-market conditions and reduce the rate of return from 15.44 percent to approximately 11.5 percent (even lower than the 12.59 percent originally argued before the AU and well below the 13.45 percent lower end of the zone of reasonableness selected in this case) was not unreasonable or arbitrary. In Case One, the Commission exercised its discretion to adjust the rate of return within the zone of reasonableness. To reduce the rate two percentage points below the zone of reasonableness would have required reopening the record and taking new evidence to establish a new zone of reasonableness. The Commission’s decision to decline reopening the record in the interest of finality was not unreasonable. Moreover, the Commission specifically found that even if Gallup’s money-market evidence were considered, the adopted rate of return remained just and reasonable. See record, Case Three, joint app. vol. 2 at 405. We conclude that FERC’s action was not arbitrary, but reasoned.

V. PNM’s Coal Costs

Both the AU and Commission concluded that the contract price PNM paid to its fifty-percent-owned affiliate, Western Coal Company, for its coal was reasonable and thus fully includable in its cost of service for determining Gallup’s rates. The AU and Commission determined that “the reasonableness of the cost of coal purchased from an affiliate should be determined by a comparison to the prices of coal available from non-affiliated suppliers," record, Case One, joint app. vol. 2 at 469, and that PNM's coal costs satisfied this standard.

In all three cases before us, Gallup argues, precisely as it did before the ALT and Commission, that a cost-of-service standard rather than a market price standard must be used to judge the reasonableness of PNM's coal costs as a matter of law. In addition to asserting that precedent requires use of a cost standard, Gallup argues that the Commission's market price standard has unreasonable consequences because it allows Western to earn excessive profits on extremely rich but "commercially worthless" coal.¹²

Gallup's claim that Western's excessive profits created unreasonable consequences was rejected by the ALT:

{N]o showing has been made ... that Western's earned rate of return is unjust and unreasonable.... No witness presented any credible evidence which would indicate that Western's returns exceed those earned by companies of comparable risk or the level necessary for Western to maintain its credit standing, preserve its financial integrity, and attract new capital.

Id. at 394 (footnotes omitted). The Commission affirmed the ALT's decision. Our own review of the record convinces us that the AU's finding on this point is founded upon substantial evidence; we must therefore decline Gallup's invitation to disturb it.

Moreover, our review of the precedent cited by Gallup does not convince us that the cost-standard for evaluating the reasonableness of PNM's coal costs is required by law. See Western Distrib. Co. v. Public Serv. Comm'n, 285 U.S. 119, 52 S.Ct. 283, 76 L.Ed. 655 (1932); Smith v. Illinois Bell Tel. Co., 282 U.S. 133, 51 S.Ct. 65, 75 L.Ed. 255 (1930); Mississippi River Fuel Corp. v. Federal Power Comm'n, 252 F.2d 619 (D.C.Cir.), cert. denied, 355 U.S. 904, 78 S.Ct. 331, 2 L.Ed.2d 260 (1957). These cases merely caution that a firm may control an affiliated company to such an extent that it is able to dictate the terms of agreements between the two. Where inter-affiliate pricing schedules are established under such a circumstance, we agree that wholesale rates based on the schedules must be examined closely to ensure they are just and reasonable. We are not convinced, however, that the case law dictates cost-based pricing.

In this case, as in Public Service Co. v. FERC, 628 F.2d 1267 (10th Cir.1980), cert. denied, 451 U.S. 907, 101 S.Ct. 1974, 68 L.Ed.2d 295 (1981), there is no record evidence that PNM is in a position to control its fifty-percent-owned affiliate.¹³ The cases cited by Gallup deal with wholly-owned subsidiaries, companies in a legal relationship more suspect than the one between Western and PNM. The ALT explicitly found that Western's coal contract was negotiated at arm's length and found "no evidence suggesting anything improper, such as fraud or duress, on the part of PNM...." Record, Case One, joint app. vol. 2 at 392. Even though convincing evidence of improper influence over Western by PNM was lacking, the Commission concluded that special scrutiny was nevertheless given in this case and that rejection of the price-comparison standard in favor of the cost-of-service standard was not required. See id. at 469.

Moreover, FERC's articulated reasons for choosing the market price standard are reasonable, including its lack of jurisdiction over rate-making in the coal industry, its lack of expertise in the field, and the perceived benefits of allowing the competitive market to determine coal prices. Unlike the situation in Electricity Consumers Resource Council v. FERC, 747 F.2d 1511 (D.C.Cir.1984), cited by Gallup, here the Commission based its conclusions on record evidence and reasoned decision making.

Finally, a government study cited by both parties in support of their respective positions demonstrates the lack of agreement in the industry as to which method of calculation yields the more reasonable result. As the study makes clear, there are strengths and weaknesses in both. See United States Department of Justice, Competition in the Coal Industry (1978). In affirming the Commission on this point, we do not presume to settle the matter by declaring the market price standard superi- or in every instance. We simply accept the Commission’s chosen methodology because we find substantial justification for its use in the record before us. “[Rjesolution of conflicting economic issues is precisely within the expertise of FERC.” Nepco Mun. Bate Comm’n, 668 F.2d at 1342.

In short, our task here is limited to determining whether a utility’s fuel purchase contracts impose an excessive financial burden on the utility's ratepayers. The identity of the fuel supplier (and its legal relationship to the utility) is pertinent to our inquiry only if record evidence suggests that dealings between business affiliates have corrupted the negotiations on coal prices. Since the record indicates that PNM purchased coal at a price comparable to that prevailing on the competitive market, we affirm the Commission’s decision that PNM’s fuel purchase arrangements with Western Coal Company are not responsible for an unjust or unreasonable rate.

In Case Two and Case Three, Gallup argues that even if the market price standard is used, PNM’s coal costs do not survive scrutiny. It contends that the comparison data upon which the AU relied were fatally flawed because the prices quoted included transportation costs. Because of PNM’s mine-mouth generating plant, see supra note 12, Gallup asserts that no transportation costs should be included in the reasonable price for Western’s coal. Those data that did purport to consider this peculiarity were predicated on impossible assumptions, Gallup argues.

In essence, Gallup maintains that its price data should have been adopted rather than data submitted by PNM and the Commission staff. Gallup based its price calculations on the operations of a single neighboring mine, while PNM and FERC analyzed prices in a multi-state, southwestern market. After reviewing the record, we cannot conclude that the AU’s decision to reject Gallup’s data was unreasonable. Evaluation of conflicting evidence is peculiarly within the province of the trier of fact — the agency. See Colorado Interstate Gas Co. v. Fed. Power Comm’n, 142 F.2d 943, 961 (10th Cir.1944), aff'd in relevant part, 324 U.S. 626, 65 S.Ct. 850, 89 L.Ed. 1235 (1945). The ALJ’s finding, adopted by FERC, that Western’s coal prices were comparable to the prices charged by non-affiliated concerns in the relevant market, was based on substantial evidence in the record. It need not be based on conclusive evidence or even be supported by the weight of the evidence. See supra note 3.

The other arguments Gallup advances in Case Two and Case Three with respect to the coal-cost issue are not really challenges to the data used but rather veiled attacks on the market price standard. Once the market price standard is adopted, the only relevant inquiry is whether PNM purchased Western’s coal at a price comparable to what it would have to pay to a non-affiliated supplier. Intimations about Western’s putatively excessive profits and statements that Western’s mine produces coal at a cost approximately one-tenth of the national average are only relevant to a cost-of-service standard, rejected by the Commission. We thus affirm FERC’s conclusion that PNM’s total coal costs could be included in the price of its service to Gallup.

VI. EEI Contributions

Consistent with its prior practice, in Case One and Case Two FERC permitted PNM to include in its rates for Gallup a portion of the contributions PNM made to the research and development activities of the Edison Electric Institution (EEl). The EEl is not open to participation by wholesale customers like Gallup, see Illinois Power Company, 15 F.E.R.C. (CCH) ¶ 61,-050 (1981), yet such customers ultimately benefit from the Institution's research and development projects. Thus, it is FERC's position that wholesale customers should bear their proportionate share of these costs. However, FERC also maintains that research and development contributions made to entities which do accept contributions from wholesale customers may not be similarly allocated. In the Commission's view, an assignment of costs in that instance would destroy the wholesale customer's incentive to contribute voluntarily and participate in the program. FERC explained:

[C]ontributions based on the volume of retail sales should not be partially recovered through wholesale rates to customers who may also make such contributions based on their own retail sales. If the wholesale supplier recovers its contribution from its retail customers, and its wholesale customers, in turn, contribute and recover their contributions from their own retail customers, each group of retail customers pays an equitable proportion of the H & D expense. However, if the wholesale supplier is allowed to recover a portion of its contributions from its wholesale customers, the retail customers of these wholesale customers (that also contribute) would be paying a greater share of the R & D expense than the retail customers of the wholesale supplier.

Record, Case One, joint app. vol. 2 at 474.

Gallup challenges FERC's decision regarding the EEl contributions, arguing that it is "inequitable and unlawful" to essentially force Gallup to contribute to a program that does not permit it a voice in determining how the contribution will be spent. Gallup further argues that PNM failed to demonstrate that Gallup would obtain "specific quantified benefits" from these contributions. See Initial Brief of the City of Gallup, Case One at 54-56.

Research and Development expenditures that benefit FERC-jurisdictional ratepayers (wholesale customers) in a general sense may be properly included in rates. See Public Util. Comm'n v. FERC, 660 F.2d 821 (D.C.Cir.1981), cert. denied, 456 U.S. 944, 102 S.Ct. 2009, 72 L.Ed.2d 466 (1982). The Commission's policy is not inequitable and unlawful merely because Gallup cannot direct how EEl spends such funds in its research. Nor is it void if PNM fails to show a "specific quantified benefit" to Gallup from EEl's research. That burden would be impossible to sustain. Research and development programs do not translate into direct and quantifiable benefits that can be immediately allocated among customers in proportion to their financial contributions, as Gallup's argument implies. Nonetheless, the wholesale customers of an electric utility undoubtedly benefit from the refined technologies resulting from research and development and ought to bear a proportionate share of their costs.

PNM does shoulder the burden of demonstrating that the research and development contributions it seeks to allocate were reasonable and prudent. Although Gallup does not challenge the expenditures, PNM must demonstrate that they were spent on activities that will benefit, if only indirectly, customers like Gallup. See Delmarva Power & Light Co., 22 F.E.R.C. (CCH) ¶ 63,052 (1983), aff'd in relevant part, 25 F.E.R.C. (CCH) ¶ 61,022 (1988) (disallowing EEl contributions because utility failed to disclose the extent to which monies were spent for lobbying activities, an expenditure for which it is prohibited from seeking reimbursement). Because PNM provided such an analysis in these cases, we conclude that the Commission's decision was reasonable in its consequences.

VII. New Mexico State Tax

In each of the three cases, FERC permitted PNM to include in its cost of service to Gallup an inspection and supervision tax paid to the State of New Mexico on its intrastate, but nevertheless FERC-juris-dictional, sales to Gallup pursuant to New Mexico Statutes Ann. Section 62-8-8 (1978). Under the New Mexico statute, all utilities doing business in the state, including those engaged in interstate business, must pay a 0.5 percent tax on gross receipts from intrastate business.¹⁴ Gallup challenges inclusion of this tax in PNM’S rate structure. In response to Gallup’s objection, the AU relied on the reasoning in Public Service Co. of New Mexico, 10 F.E.R.C. (CCH) ¶ 63,020 (1980), a case which also involved PNM:

[T]o remove this tax expense entirely from PNM’s cost of service treatment and consequently spread it among its retail customer class is also highly inequitable. New Mexico did impose and PNM did pay this tax upon wholesale sales to a wholesale customer. Thus PNM properly must reflect this tax cost as a wholesale cost.

Id. at 65,140.¹⁵

State taxes, “like other necessary operating expenses, may properly be included as rate components,” City of Cleveland v. Federal Power Comm’n, 525 F.2d 845, 850 n. 37 (D.C.Cir.1976), so long as the utility incurs the state tax in providing the service for which the rates are being set. The service for which the rates are being set in this case is intrastate provision of electrical power to Gallup, the same service measuring the state tax imposed. Thus, judicial precedent sanctions the Commission’s decision in this case. Commission precedent in Cincinnati Gas & Electric Co., 59 FPC 2147 (1977), cited by Gallup, does not dictate a contrary result. The Commission’s action was reasonable in its consequences and is thus affirmed.

VIII. Economy Energy Sales

Economy Energy Sales are sales of excess electrical power to other utilities that find it more economical to purchase the energy than to generate it themselves. The revenue earned from PNM’s economy energy sales offsets its own purchased power expenses, thus reducing the cost PNM must recover from its wholesale customers. See Indiana Mun. Elec. Ass’n v. FERC, 629 F.2d 480, 481 (7th Cir.1980). PNM prices most of its economy energy sales using a “split-the-savings” calculation: PNM’s incremental cost of generation plus one-half the difference between PNM’s incremental cost and the cost of generation avoided by the buyer (its decremental cost). If the decremental cost is indeterminable, PNM charges its incremental cost plus an “adder” of fifteen percent of the cost of fuel and operating and maintenance (0 & M) costs. The incentive component of such sales, which Gallup challenges, consists of either the “split-the-savings” revenue in excess of PNM’s incremental and 0 & M costs or the fifteen percent adder.

In Case One, PNM estimated its 1978 economy energy sales revenue using a fifteen percent incentive component. The actual 1978 revenue of $375,000 approximated the $248,000 estimate. Nevertheless, Gallup argues that PNM failed to meet its initial burden of proof of establishing that an estimate using a fifteen percent incentive component was reasonable at the time PNM calculated its estimate. Gallup originally asserted that a 52.1 percent incentive component should have been used, but now supports a 41.4 percent figure based on an average of the actual incentive components realized between 1975 and 1979, inclusive,¹⁶ resulting in estimated revenues approximately twice that actually earned.

The Commission’s current method of rate making requires the use of both historical data for the most recent twelve months (Period I) and projected data (Period II or “test year”). FERC adopted the test-year approach because it found the use of exclusively historical information too rigid to result in just and reasonable rates. See Papago Tribal Util. Auth. v. FERC, 773 F.2d 1056, 1059 (9th Cir.1985), cert. denied, 475 U.S. 1108, 106 S.Ct. 1515, 89 L.Ed.2d 913 (1986).

The utility must demonstrate that its estimates were reasonable when made, either by explaining the chain of reasoning that it employed to arrive at its projections, or by comparing its estimates with actual data, and so establishing their accuracy. If the utility proves that its estimates were reasonable when made even though at variance with actual results, the Commission may nonetheless credit those estimates, a decision further substantiated by evidence indicating that the actual figures are based upon unusual or unique occurrences.

Villages of Chatham and Riverton v. FERC, 662 F.2d 23, 29-30 (D.C.Cir.1981). In most cases, actual Period II data are usually available by the time rate-making orders are reviewed. Nevertheless, the Period II estimates should not be rejected merely because they vary from the actual data, since “that would supplant period II ratemaking with historic-cost ratemaking contrary to Commission regulations.” Id. at 29; see also Indiana Mun. Elec. Ass’n, 629 F.2d at 483. Relying on estimates over the actual results is particularly appropriate when the actual data are aberrational. In such a case, sound estimates may well result in more rational rates over the long run. Thus, estimates that are reasonable when made are accepted unless “subsequent events indicate that to use them as a basis for future projections would yield unreasonable results.” Id. at 485.

In this case, the AU and Commission did not blindly accept PNM’s estimate without requiring any foundation, as Gallup suggests. The AU examined PNM’s estimate and its supporting rationale and specifically found that “the 15% figure appears reasonable in light of the circumstances then existing.” Record, Case One, joint app. vol. 2 at 398. The AU noted that PNM’s actual incentive components in the three years prior to making the estimate in question had fallen precipitously from 61.2 percent in 1975 to 32 percent in 1976 to 22.1 percent in 1977. In view of this “sharply downward” trend, the AU found a fifteen percent estimate to be reasonable. Furthermore, it concluded that use of the estimate would not have unreasonable consequences because the actual revenues did not deviate dramatically from the estimated revenues. We conclude that PNM satisfied its burden of establishing the validity of its cost estimate regarding the anticipated incentive component of its 1978 economy energy sales.

In Case Two, PNM projected no economy energy sales for the 1979 test-year period because a boiler explosion in 1977, causing a year-long shutdown of one of its coal-fired generators, required PNM to purchase a large amount of energy. Any surplus capacity generated in 1979 was budgeted as “payback energy” to Colorado River Storage Project, the utility from which PNM purchased its power while its generator was under repair. Despite its payback obligation, PNM was able to make economy energy sales in 1979 to-talling $3,307,046, including an incentive component of $1,143,632,¹⁷ because its wholesale customers’ peak demands for the projected period were lower than estimated.

The AU rejected PNM’s failure to budget for economy energy sales, finding the zero estimate unreasonable when made because it was based on atypical circumstances — the boiler outage. He then substituted figures based on PNM’s actual experience and incorporated $1,143,632 as the incentive component of economy energy sales for Period II. The Commission affirmed without comment.

PNM argues that its zero estimate was reasonable when made in light of PNM’s payback obligation. Alternatively, it asserts that its underestimation was offset by its overestimation of revenues relating to wholesale peak demands, and thus no adjustment to the economy energy sales estimation was warranted.¹⁸

We are persuaded that the AU’s resolution of this issue was reasonable both on its face and in its consequences. Just as atypical actual data should not be substituted for sound estimates, see Villages of Chatham and Riverton, 662 F.2d at 29-30; Indiana Mun, Elec. Ass’n, 629 F.2d at 483, 484, aberrational estimates based on unique and extraordinary circumstances should not be preferred to more typical actual data. The test-period approach, unlike the historic-data approach, is based on the assumption of normal conditions that are reasonably expected to prevail in the future. The AU reasoned:

The Commission has explained that the purpose of a test period cost of service is to:

... develop adjusted test year data that reflects the relationship among costs and revenues which will prevail over a period of years.

PNM has a responsibility to demonstrate on the record that its estimates represent normal and typical sales that can be expected to occur in the future:

The company has a particular responsibility to ... substantiate the Period II figure in terms of its typicality not only for the test period but also for the projected effective term of the tendered rates.

The fact that it did not budget economy sales for 1979 because of an atypical outage is not a reason for failing to estimate economy energy sales for the test period for ratemaking purposes. Our task is to project the likelihood of PNM making economy energy sales in the future, not just in 1979. The pattern of PNM’s economy energy sales for 1975-78 indicates that the pattern will continue in the future. Indeed, the pattern continued in 1979, because PNM made substantial economy sales in that year, and PNM witness Brenner anticipated economy energy sales by PNM for 1980.

Record, Case Two, joint app. at 191-92 (citations omitted).

Nor do we find PNM’s offsetting-errors argument convincing. PNM asserts that its $1.1 million incentive component was more than offset by its $2.5 million revenue shortfall. However, that argument compares apples and oranges. The $1.1 million figure is analogous to a net-profit figure (based on economy energy sales revenue of $3.3 million), whereas the $2.5 million figure is analogous to a gross-receipts figure which does not reflect the incremental and 0 & M costs of producing the energy that would have been sold for the $2.5 million. It is not at all clear on the record that there would have been net profit in the $2.5 million revenue or that such profit would approach the $1.1 million incentive component of the economy energy sales.

Moreover, contrary to what PNM argues, the chief cases cited by PNM as prohibiting FERC’s action here do not require the Commission to offset alleged errors rather than substitute actual data for estimates. For example, in Villages of Chatham and Riverton, the D.C. Circuit stated that

in deciding whether to substitute actual for estimated numbers, the Commission retains some discretion to look beyond the challenged line item to related figures and to determine that, if actual numbers were substituted on a uniform basis, the result would not favor the party that had objected to the period II figures in the first instance. When, as here, the Commission explicitly makes the determination and can produce substantial evidence to support it, we fail to see the harm in upholding their decision to rely upon test-year estimates.

662 F.2d at 34 (emphasis added). The court’s language does not proscribe the use of test-year estimates; it merely requires a showing of substantial evidence supporting the offsets before permitting the practice. In Indiana Municipal Electric Association, 629 F.2d 480 (7th Cir.1980), the record showed that a $4 million overestimate in operating expenses was offset nearly dollar for dollar by a $4 million underestimate in revenues — a comparison of apples with apples based on substantial record evidence. “These were both wholesale revenue and expense figures, so that the Commission did not offset unrelated accounts.”¹⁹ Id. at 484. Therefore, the Seventh Circuit upheld FERC’s adherence to the estimates.

In neither case did the record show, as it does here, that the estimate was unreasonable when made because it was based on unusual circumstances. Even if the two miscalculations cancelled out each other in 1979, there is no reason to believe that the revenue underestimation would recur each year to offset the incentive components that are very likely to occur in the future (based on PNM’s history) but which are not factored into Gallup’s rates. As discussed, Period II rate making requires forward-looking estimates that take into account revenues and expenses over the long term. In addition, the record in this case simply fails to supply the requisite substantial evidence to justify the requested offset. We therefore affirm FERC’s adjustment to PNM’s Period II estimate of the incentive component of its economy energy sales.

IX. Cash Working Capital Allowance

Under Commission regulations, a utility is permitted to collect, as part of its rate base, an amount denominated “cash working capital.” This allows the utility to meet current obligations by accumulating revenue before payment for services is received. Commenting on the practice, the D.C. Circuit has explained:

“[T]he need for working capital arises largely from the time lag between payment by the company of its expenses and receipt by the company of payment for service in respect of which expenses were incurred.”

A utility's actual need for working capital can be most accurately determined by performing a “lead-lag” study of the average number of days that passes between payment of expenses and receipt of revenues for a given service. One part of this calculation is the “revenue lag,” which is in turn made up primarily of the “service lag” — the number of days between the time expenses are incurred for services and the date of billing for those services — and the “payment lag”— the number of days between billing and payment. A utility also experiences “lead time” when it receives payment for services before it pays the expenses associated with those services. The number of lag days minus the number of lead days yields a net lag which represents the utility’s actual need for working capital.

Boroughs of Ellwood City, Grove City, New Wilmington, Wampun, and Zelienople, 731 F.2d at 963 (footnote and citations omitted). Recognizing the burden of producing a fully developed and reliable lead-lag study, the Commission will allow a utility to employ instead the “forty-five day rule.” Id. at 963-64. Under this rule, a utility is entitled to include in its rate base an amount equal to one-eighth (45/365) of the company’s total annual operating and maintenance expenses exclusive of purchased power expense.

In Case One, the Commission adopted as fully developed and reliable a lead-lag study initially sponsored by Plains Electric Generation & Transmission Cooperative, Inc., a party which eventually settled and is not involved in this appeal. Plains’ study revealed only a 6.78-day lag between expenses incurred and receipt of revenues based on those expenses, yielding a cash working capital requirement of $2,298,329. Both PNM and Gallup, each on different grounds of course, challenge the use of the Plains’ study.

PNM supported use of the forty-five-day rule in this case, which would have resulted in a cash working capital requirement of $11,795,318. PNM argues that the Plains’ study is fatally flawed because it failed to reflect PNM’s actual cash needs during the time in question by assuming PNM’s proposed rate increase to be in effect during the test period when the increase did not actually become effective until several years after the test year. In PNM’s view: “the study did not fulfill the goal of a cash working capital allowance, i.e., to assume that a utility's investors are compensated for funds expended to provide service in advance of receiving revenue to compensate for the service.” Brief of Petitioner Public Service Company of New Mexico, Case One at 20.

PNM’s challenge to a study which “measured only timing factors,” id., reveals PNM’s basic misunderstanding of what a lead-lag study is designed to accomplish. By definition, a lead-lag study focuses “solely upon timing differences.” Its express purpose is to calculate the number of days between expenditures and payments for service provided — no matter what the service cost or how much PNM receives for providing it. As the AU observed:

[I]t is the timing of revenue receipt, and not the level of revenues, which is the critical determinant. The plains study would have produced the same cash working capital allowance had the level of revenues been higher (or lower) than that shown in the Exhibit, as long as the lag between service date and revenue receipts remained unchanged.

Record, Case One, joint app. vol. 2 at 363 (footnote omitted). We find FERC’s observation and analysis persuasive:

What PNM actually argues is that a utility should receive an increase in rate base to compensate for a different type of timing problem — the cash shortfalls that may result from the fact that a utility collects rates during the test period that are different from the increased rates sought in the ongoing rate proceeding, i.e., cash shortfalls resulting from regulatory lag. Such cash needs do not fall within the meaning of “a short-term loan from investors to provide for interim cash expenses.” [Anaheim, Riverside, Banning, Colton, and Azusa v. FERC, 669 F.2d 799, 806 (D.C.Cir.1981) ]. As recognized by the AU, there is no authority for increasing a utility’s rate base to compensate for cash shortfalls resulting from regulatory lag.

Brief for Respondent Federal Energy Regulatory Commission, Case One at 40-41. PNM has not convinced us that use of the Plains’ study leads to unreasonable results.

Gallup challenges the cash working capital allowance as too high. Gallup accepted the revenue and expense payment lags calculated in the Plains' study with the exception of the coal-cost component; it substituted a lower coal cost calculated under cost-of-service pricing. Gallup then purported to allocate working capital requirements among PNM's various wholesale customers and thereby derive a cash working capital requirement for Gallup individually. Under its calculations, PNM has a negative cash working capital requirement of $119,013 in serving Gallup.

The AU rejected both adjustments, and FERC affirmed. We agree with this disposition. The salient inquiry in this court is whether the record shows that the Plains' study is fully developed and reliable-a factual conclusion we may modify only if unsupported by substantial evidence-or, alternatively, if use of the study leads to unreasonable results. Once again, rather than focusing on whether the record supports the study on which the Commission eventually relied, Gallup argues that its calculations should have been chosen in the first instance. Even if we were to recast Gallup's arguments in a forum proper for this forum, we find no error. Incorporation of PNM's actual coal costs in the Plains' study is based on record evidence and does not lead to unreasonable results. Similarly, we do not find the ALT's rejection of Gallup's novel customer-allocation method unreasonable, particularly in light of our own rejection of Gallup's arguments on cost-of-service pricing in Section V above. We agree with the AU's conclusion that "[w]hile determination of working capital allowances on an allocated basis is not necessarily conceptually erroneous, its dependence on a particular allocation method makes the exercise more, and perhaps needlessly, complex, and prevents useful comparison of its results with presentations made on a systemwide basis." Record, Case One, joint app. vol. 2 at 369-70.²⁰

In Case Two, PNM again supported use of the forty-five-day rule, which would have resulted in a $4,190,202 cash working capital requirement. Gallup introduced a lead-lag study showing that PNM's cash working capital requirement for service to Gallup was negative in the amount of $283,925. Gallup did not, however, advocate a negative allowance (a subtraction from the rate base), but rather urged adoption of a zero cash working capital allowance. The AU found Gallup's lead-lag study to be fully developed and reliable and granted PNM a cash working capital allowance of zero. FERC affirmed without comment.

PNM puts forward the same argument here as in Case One, i.e., that a study considering only timing factors to the exclusion of actual cash received cannot be deemed fully developed and reliable. As in Case One, we reject its argument.

The AU permitted PNM to include $22,691 in the rate base to recoup prepayments it had made. The Commission routinely adds prepayments to the rate base when the forty-five-day rule is used because prepayments are not part of the operating and maintenance expenses used to compute leads and lags. The AU concluded that it should do likewise when a lead-lag study is substituted. Gallup argues that prepayments should not have been added to the rate base because some prepayments were already factored into the lead-lag study. In addition, it contends that the prepayments were more than offset by the lenient zero cash working capital allowance when the lead-lag study actually supported a deduction from the rate base in the amount of $283,925.

With respect to Gallup's first contention, the ALT found insufficient evidence that Gallup's study incorporated all of PNM's prepayments. Gallup acknowledged that only an unspecified portion of the prepayments was included. In view of this inconclusive evidence, the ALJ did not act unreasonably in making the usual addition of prepayments to the rate base. Furthermore, there is no possibility of “double recovery” here, for eliminating all prepayments from Gallup’s study would only have resulted in a figure that was less negative. The allowed cash working capital requirement would still have been zero. Yet, in that case, Gallup could not argue that prepayments should not then be added, as under the forty-five-day rule. Therefore, addition of PNM’s prepayments to the rate base when some portion of them may have been included in the lead-lag study used to calculate PNM’s cash working capital requirement does not lead to unreasonable results.

We are also unpersuaded by Gallup’s argument that the prepayments must be netted against the negative cash working capital requirement demonstrated by the lead-lag study. Cash working capital requirements and prepayments are separate, though somewhat similar, adjustments to the rate base. The former deals with operating and maintenance expenses while the latter pertains to monies spent by PNM for such items as insurance. Prepayments are included in the rate base so that the rate payers will reimburse the company over' the life of the service. Merely because PNM generally receives revenues 3.67 days before paying the 0 & M expenses associated with its service to Gallup does not lead inexorably to the conclusion that its prepayments for unrelated items should not be reflected in the rate base. While PNM may not need a temporary “loan from investors” to meet 0 & M expenses for service to Gallup, it does not necessarily follow that it has the cash on hand to meet its prepayment obligations or that the computation of the rate base should ignore such obligations. The two figures deal with theoretically distinct components of the rate base which do not mesh perfectly for netting purposes. While we do not conclude that FERC would be acting unreasonably if it allows a utility to net negative cash working capital requirements with prepayments, neither do we find it unreasonable or unprincipled if the Commission declines to do so.

In Case Three, the lead-lag study adopted also revealed a negative cash working capital allowance requirement.²¹ Unlike in Case Two, though, Gallup did not advocate a zero cash working capital requirement. Rather, it argued that the rate base should be reduced by $129,193, the amount of the negative allowance requirement. FERC declined Gallup’s request, adhering to its long-standing policy that only upward adjustments to rate base are made in recognition of the need to supply a return on the short-term “loan” advanced by investors to meet the utility’s bills.

We conclude, as did the Commission, that even if the Commission’s policy is in need of reevaluation, this case is not the proper vehicle to reach the issue. See record, Case Three, joint app. vol. 2 at 310. PNM’s total rate base was approximately $708 million, of which approximately $16.8 million was allocated to Gallup. Whether or not the $129,193 is included in Gallup’s rate base, the resulting rate is within the zone of reasonableness and thus is beyond our review. See Permian Basin, 390 U.S. at 790, 88 S.Ct. at 1372; see also Hope Natural Gas Co., 320 U.S. at 602, 64 S.Ct. at 288.

We also reject Gallup’s challenge to the addition of PNM’s prepayments to the rate base for the same reasons discussed with respect to Case Two and, therefore, affirm the Commission’s computations of PNM’s cash working capital requirements for purposes of establishing the rate base in all three cases.

X. Effective Dates of Rates

Section 206(a) of the Federal Power Act provides in relevant part: “[T]he Commission shall determine the just and reasonable rate ... to be thereafter observed and in force, and shall fix the same by order.” 16 U.S.C. § 824e(a) (1982) (emphasis added). In Electrical District No. 1 v. FERC, 774 F.2d 490 (D.C.Cir.1985), the D.C. Circuit held that rates are not “fixed” under the statute, and thus may not be effective, until the utility’s compliance filing²² detailing the rates has been accepted by the Commission.²³ In each of the cases before us, the Commission made PNM’s rates effective as of the date it issued its order establishing the utility’s revised rate schedule. The Commission’s order preceded PNM’s submission of a compliance filing and, therefore, the Commission’s accept-anee of the filing. Gallup challenges the. Commission’s order as unlawful under the statute. While FERC admitted at oral argument that it now conforms to the rule in Electrical District No. 1, the agency argues that Electrical District No. 1 should not control the disposition of this case for two reasons: (1) it was wrongly decided and (2) the facts of Electrical District No. 1 are sufficiently distinguishable from those of the present case to preclude the application of the D.C. Circuit’s rule.

Before we reach the merits of this issue, we must address a threshold issue raised by FERC. Section 313(b) of the Federal Power Act provides: “No objection to the order of the Commission shall be considered by the court unless such objection shall have been urged before the Commission in the application for rehearing unless there is reasonable ground for failure to do so.” 16 U.S.C. § 825i(b) (1982). In Cáse Two, Gallup squarely argued in its petition for rehearing before the Commission that a rate may not become effective until the Commission accepts the utility’s compliance filing. Thus there is no dispute about whether we may resolve this issue in that case. FERC argues, however, that Gallup failed to satisfy the statutory requirement in Case One and Case Three, and we are thus deprived of jurisdiction to hear the issue in those cases. See Federal Power Comm’n v. Colorado Interstate Gas Co., 348 U.S. 492, 497-99, 75 S.Ct. 467, 470-71, 99 L.Ed. 583 (1955); Kentucky Utils. Co. v. FERC, 789 F.2d 1210, 1214-15 (6th Cir.1986); Kansas Cities, 723 F.2d at 85.

In Case One, Gallup did specifically challenge FERC’s order allowing the rates to go into effect on the date the Commission issued its opinion and did argue that the rate was not fixed under the statute on that date. However, because Gallup’s argument with respect to why the rate was not fixed on that date differed in substance from the problem addressed in Electrical District No. 1, FERC asserts Gallup failed to preserve the issue for appeal.²⁴ We conclude that the question whether the rate was fixed under the statute on the date FERC issued its opinion in case one was sufficiently pressed before the Commission to allow our review of the issue as a matter of law.

Gallup failed to raise the Electrical District No. 1 issue in its petition for rehearing in Case Three but argues that it may nevertheless pursue the matter on this appeal because it had a “reasonable ground for [its] failure to do so” under the statute. In Arkansas Power & Light Co. v. Federal Power Commission, 517 F.2d 1223 (D.C.Cir.1975), cert. denied, 424 U.S. 933, 96 S.Ct. 1146, 47 L.Ed.2d 341 (1976), the utility failed to raise in its petition for rehearing under the Natural Gas Act a claim that the National Environmental Policy Act (NEPA) required the filing of an environmental impact statement with respect to proposed curtailment plans. The Natural Gas Act has the identical mandatory petition-for-rehearing requirement and identical reasonable-ground exception as the section of the Federal Power Act at issue here. The court, deciding that the reasonable-ground exception was satisfied, reasoned:

[A]t oral argument, counsel for the FPC advised us that the Commission now files impact statements in connection with all permanent curtailment plans.

[T]his is a case in which the agency has for all practical purposes had an opportunity to pass on the issue sought to be raised. The Commission expressed its opinion that NEPA did not apply to curtailment proceedings in El Paso Natural Gas Company, 48 FPC 371 (1972), and reiterated the same view in United Gas Pipe Line Company, 49 FPC 179, 192-93 (1973), vacated and remanded sub nom. Louisiana v. FPC, supra, seven days after the filing of Opinion No. 643 in this case. There is no reason to believe that its position would have been any different in the Arkla proceedings had the issue been raised. See Great Falls Community TV Cable Co. v. FCC, supra [416 F.2d 238 (9th Cir.1969)]. Were we to hold that section 19(b) forecloses us from considering the NEPA objection, we would be presented with the unseemly prospect of a violation of a statutory duty now recognized by the Commission to exist without the means to direct compliance.

Id. at 1237.²⁵

In Case Three, FERC also "for all practical purposes had an opportunity to pass on the issue sought to be raised [and there was] no reason to believe that its position would have been any different had the issue been raised." Arkansas Power & Light Co., 517 F.2d at 1237. In both Case Two and Case Three, FERC merely cited to the opinion in Case One for the proposition that the rates may be effective on the date it issued its opinion resolving the issues raised in the case. When it did this in Case Three, it had already explicitly rejected Gallup's petition for rehearing in Case Two, in which Gallup had argued that the statute prohibited FERC's action. Moreover, just as in Arkansas Power & Light Co., the Commission admitted at oral argument that it now was complying with Electrical District No. 1 in all cases and making rates effective on the date the compliance filing is accepted. We thus conclude that the reasonable-ground exception has been satisfied, and our resolution of the merits will therefore apply to all three cases before us.

Unlike the other issues presented in this appeal, this one involves interpretation of the Federal Power Act. The issue is not one of fact requiring deferential review but rather one of law. We thus review the problem de novo. See Chevron U.S.A., Inc. v. Natural Resources Defense Council, Inc., 467 U.S. 837, 843 n. 9, 104 S.Ct. 2778, 2781 n. 9, 81 L.Ed.2d 694 (1984); Phillips Petroleum Co. v. FERC, 786 F.2d 370, 374 (10th Cir.), cert. denied, - U.S. -, 107 S.Ct. 92, 93 L.Ed.2d 44 (1986).

We disagree that Electrical District No. 1 was simply wrongly decided and are persuaded by the reasoning stated therein. We see no need to repeat it here. PNM's attempt to distinguish Electrical District No. 1 from the present case on its facts is equally unpersuasive. PNM argues that "the filed rate considerations upon which the Electrical District No. 1 decision turned are not present." Brief of Intervenor Public Service Company of New Mexico in Support of Respondent, Case Two at 34. Once a statutory term such as "fix" is interpreted, that interpretation applies to that issue in other factual contexts as well. The facts in that case helped the court, as they should, decide the proper construction of the statutory term, but the interpretation was not thereby limited to cases factually identical to Electrical District No. 1. The legal issue there was phrased identically to the issue we must decide here: "[T]he question it presents is the lawfulness of FERC's decision to make a rate increase effective as of the date of its order directing a compliance filing, rather than upon the date of acceptance of the compliance filing." Electrical District No. 1, 774 F.2d at 491. The resolution of that issue of law must be consistently applied. In short, we do not subscribe to PNM's ad hoc factual approach to statutory interpretation.

We therefore conclude that the Commission erred in ordering the rates effective on the date it issued its opinions adjudicating the litigated issues in the various proceedings. Under the statute, the rates are fixed on the date FERC accepts PNM's compliance filings.

XL Conclusion

We affirm FERC's orders in all respects with one exception. In each case, FERC's order establishing PNM's new rates on the date FERC issued its order approving the revised rate schedule is set aside. The case is remanded to FERC for further proceedings consistent with this opinion.

Affirmed in part; reversed and remanded in part.

1 . Three cases dealing with rate-increase tilings initiated in 1978, 1979, and 1980, respectively, are before us. Gallup and PNM each appeals the issues decided adversely to it and intervenes on the side of FERC with respect to those issues decided in its favor. Because many of the same issues are raised in each, we dispose of all three appeals in this opinion. Nos. 82-1148, 84-1624, and 84-1677 will be referred to as "Case One”. Nos. 82-1882 and 84-1235 will be referred to as "Case Two." Nos. 83-1228 and 84-1236 will be referred to as "Case Three.”

2 . FERC determines an "overall” rate of return to be applied to the company’s rate base by taking the three elements of a company’s overall capital structure, i.e., the amount of bonds, preferred equity, and common equity capital outstanding, weighting each in proportion to its representation in total capital, and applying the appropriate cost rate to each element to produce an overall cost rate or overall rate of return.

Nepco Mun. Rate Comm’n v. FERC, 668 F.2d 1327, 1335 (D.C.Cir.1981), cert. denied, 457 U.S. 1117, 102 S.Ct. 2928, 73 L.Ed.2d 1329 (1982).

3 . Substantial evidence is "such relevant evidence as a reasonable mind might accept as adequate to support a conclusion." Consolidated Edison Co. v. NLRB, 305 U.S. 197, 229, 59 S.Ct. 206, 217, 83 L.Ed. 126 (1938). While "more than a mere scintilla," id., substantial evidence is "something less than the weight of the evidence, and the possibility of drawing two inconsistent conclusions from the evidence does not prevent an administrative agency’s finding from being supported by substantial evidence.” Consolo v. Fed'l Maritime Comm'n, 383 U.S. 607, 620, 86 S.Ct. 1018, 1026, 16 L.Ed.2d 131 (1966).

4 . Although that case interprets the Natural Gas Act and not the Federal Power Act, the Supreme Court has repeatedly noted that these two Acts "are in all material respects substantially identical,” Arkansas Louisiana Gas Co. v. Hall, 453 U.S. 571, 577 n. 7, 101 S.Ct. 2925, 2930 n. 7, 69 L.Ed.2d 856 (1981), and cases interpreting them may be cited interchangeably. Id.

5 . The records in these cases are exhaustive — the first case alone consisting of twenty-six volumes — and the AU’s initial decisions are detailed and lengthy. With respect to certain issues, the Commission explicitly adopted the ALJ’s conclusions and underlying reasoning without further refuting each and every Gallup claim of AU error, which consisted of preferred methodologies or facts found significant by Gallup that were previously presented to and rejected by the AU. Gallup now repeatedly claims on appeal that the truth of each of Gallup’s contentions and theories not specifically addressed in the Commission's opinion must be deemed “admitted" for purposes of appeal, implying that “FERC's lack of discussion of several grounds cited in support of [Gallup’s] position affirmatively demonstrates a failure to consider those grounds." Nepco Mutt. Rate Comm., 668 F.2d at 1347. This unusual assertion is not the law.

We agree that

the administrative adjudicator, by written opinion, [must] state findings of fact and reasons that support its decision. These findings and reasons must be sufficient to reflect a considered response to the evidence and contentions of the losing party and to allow for a thoughtful judicial review if one is sought.

Harborlite Corp. v. ICC, 613 F.2d 1088, 1092 (D.C.Cir.1979) (footnote omitted). However, this requirement is fully met, and we have an adequate basis for review, when the Commission adopts as its own an AU decision that is reasoned, not conclusory, and supported by substantial evidence. We can see no benefit in requiring the Commission to parrot the AU reasoning it found persuasive. ‘The Commission is not required to recapitulate the reasoning of the AU if it is satisfied that the initial decision and the reasoning underlying it are sound. We have not been left to guess at the Commission’s findings or reasons; they are to be found in the AU’s decisions.” Boroughs of Ellwood City, Grove City, New Wilmington, Wampum, and Zelienople v. FERC, 731 F.2d 959, 967 (D.C.Cir.1984) (citations omitted). “[Assertions of a lack of rational explanation and policy discussion merely express ... dissatisfaction with FERC’s determination. The opinions of the AU and FERC make plain that the record evidence was reviewed and the involved equities were weighed. No more is required.” Nepco Mun. Rate Comm., 668 F.2d at 1334.

6 . “Price squeeze” is industry vernacular for discriminatory pricing with anticompetitive consequences which may occur when a wholesale supplier also serves retail customers. It is a price differential between retail and wholesale rates that is not justified by differing costs and which places the wholesale customer at a disadvantage in competing with its own supplier for retail customers. Although retail rates are not within FERC’s jurisdictional purview, the Commission may consider charges of price squeeze in determining a just and reasonable wholesale rate and reduce the wholesale rate to the lower end of the zone of reasonableness in order to ameliorate price squeeze effects. See Federal Power Comm’n v. Conway Corp., 426 U.S. 271, 276-82, 96 S.Ct. 1999, 2003-05, 48 L.Ed.2d 626 (1976); Public Serv. Co. v. FERC, 628 F.2d 1267, 1270-71 (10th Cir.1980), cert. denied, 451 U.S. 907, 101 S.Ct. 1974, 68 L.Ed.2d 295 (1981).

7 . PNM was the eventual victor on this issue. Even if we were to rule in PNM’s favor, no order could alleviate the putative damage done. We have considered the several exceptions to the mootness doctrine, including whether the issue is capable of repetition but evading review, see Nebraska Press Ass'n v. Stuart, 427 U.S. 539, 546-47, 96 S.Ct. 2791, 2797, 49 L.Ed.2d 683 (1976), but find none applicable.

8 . PNM erroneously argues that the Commission’s decision is inconsistent with City of Cleveland v. Federal Power Commission, 525 F.2d 845 (D.C.Cir.1976). Unlike in the present case, the Commission specifically noted there that the city "has for some time refused to pay, or has been quite late in paying, certain bills...." Id. at 850 n. 38. The District of Columbia Circuit therefore found the Commission to be fully justified in requiring a late charge ”[i]n view of the City’s history of delayed payment.” Id.

9 . Gallup has admitted that the factual, testimonial, and methodological arguments it brought before us had been presented to both the AU and the Commission. The City also has conceded that large parts of its brief were taken verbatim from its petition for rehearing before the Commission. This strategy on appeal fails to recognize this court’s limited role in reviewing administrative proceedings. Unlike the Commission, we are not empowered to grant a plenary rehearing on issues of fact. In such matters, we may only determine whether the evidence the ALJ relied upon was substantial and whether, considering the entire record, the rates approved by the Commission were just and reasonable.

10 . PNM alleges that none of the challenged construction costs were included in the rate base. See Brief of Intervenors PNM at 9 n. 16.

11 . Dilution occurs when a company sells stock at a price less than book value. In effect, a portion of the capital of existing stockholders is transferred to new stockholders.

12 . Due to geographical and other barriers, Western's coal could not be transported out of the mine area. PNM therefore built a mine-mouth generating plant and transmission lines to its load centers to overcome the transportation problem.

13 . The remaining 50% of Western is owned by Tucson Gas & Electric Company. The coal is mined, not by Western, but by Utah International, Inc., a nonaffiliated operator, pursuant to a contract between Utah International and Western.

14 . Gallup makes one somewhat unfocused argument to the effect that, as we understand it, because PNM makes wholesale sales in interstate commerce and is thus subject to FERC’s jurisdiction, it need not have paid any taxes due on intrastate sales, such as its sales to Gallup, under the New Mexico statute. Consequently, because the taxes were imprudently paid, PNM may not recover them in rates charged to Gallup.

Gallup misreads the statute. The statute does not excuse interstate sellers from paying taxes on that portion of its sales that are within New Mexico, as Gallup implies, but rather specifically addresses the amount interstate sellers do owe. The precise words of the statute quoted by Gallup as supporting its thesis in fact undermine it: "In the case of utilities engaged in interstate business, the fee shall be measured by the gross receipts of such utilities from intrastate business only ... and not in any respect upon receipts derived wholly or in part from interstate business.” N.M.Stat.Ann. § 62-8-8 (1978). PNM’s receipts from Gallup are wholly from intrastate business and are not derived "in part from interstate business.” Gallup is simply wrong in asserting that "revenues derived from Gallup are revenues derived from interstate commerce," Initial Brief of the City of Gallup, Case One, at 57, and that "PNM's sales to Gallup were sales in interstate commerce.” Id. at 59. We do not dispute that PNM is engaged in interstate commerce and within FERC’s jurisdiction, see id., but those facts do not transform PNM’s sales to Gallup, located in New Mexico, into interstate sales. Such sales occurred entirely within the State of New Mexico and were thus intrastate. Under Gallup’s interpretation, no interstate seller could ever owe tax under the statute, but that construction ignores the express language defining the amount interstate sellers shall in fact pay.

If, instead, Gallup is arguing that the State of New Mexico is constitutionally prohibited from taxing the revenue PNM derives from Gallup because the utility also happens to engage in interstate business, we are equally unpersuaded.

15 . Gallup is correct that the principle of res judicata is inapplicable because Gallup was not a party to the prior proceeding. However, it is incorrect in implying that the AU relied on res judicata in disposing of this issue. The AU simply was persuaded by the reasoning in Public Service Co., supra, and applied it in this case.

16 . See record, Case One, joint app. vol. 2 at 352.

17 . As happens frequently, actual data for Period II became available prior to the close of the hearing in this case.

18 . A third argument advanced can be dealt with summarily. PNM asserts that because economy energy sales are unpredictable by nature, its zero estimation was not unreasonable. Were we to accept this argument as sufficient, no utility would need to budget for economy energy sales. This argument rings especially hollow in PNM’s case in view of its consistent history of being a net seller of surplus energy.

19 . While the purported apples and oranges in Indiana Municipal Electric Association were wholesale accounts and retail accounts, the principle is equally applicable when comparing the essentially net-profit figures and gross-revenue figures at issue here.

20 . Gallup also argues that because its methodology was accepted in Case Two, the Commission acted unreasonably in relying on Plains' study in this case. Gallup's argument presumes that there is only one reasonable way to establish wholesale rates. However, "FERC need follow no single method in arriving at just and reasonable rates." Nepco Mun. Rate Comm., 668 F.2d at 1340. As quoted above, the AU expressly noted that Gallup's approach was not conceptually unsound. We do not find FERC's practices inconsistent simply because the agency relied on Gallup's data in Case Two, when no traditional study was offered, and then relied on a traditional study when one was presented, as in Case One.

21 . PNM again advocates use of the 45-day rule, challenging the lead-lag study for the same reasons advanced in Case One and Case Two. Just as in those cases, we reject its argument here.

22 . A compliance filing is a "rate schedule [submitted by a utility to the Commission] to make effective the rates, charges, classifications or services, or any rule regulation, practice, or contract relating thereto, established in conformity with an order of the Commission....” See 18 CFR § 35.18 (1987).

23 . We disagree with FERC’s assertion that the court "imposed no such hard and fast rule.” Brief for Respondent Federal Energy Regulatory Commission, Case One at 62. FERC erroneously argues that the court held the rate can be fixed at a "variety of times ... such as when the Commission issues its opinion deciding the contested issues if then the Commission also identifies with certainty the rate that will be effective under it, when the utility makes a compliance filing, when the Commission subsequently approves that filing, or some other time.” Id. at 62. The opinion explicitly rejected this argument:

It is uncontested (and uncontestable) that under current FERC practice no numerical rate is specified until after the compliance filing is accepted. The assumption of the Commission’s argument, however, is that to "fix” a rate within the meaning of the statute it is enough to prescribe the legal and accounting principles which, properly applied, will yield one particular rate; whereas petitioners maintain that the statute means what it says, and requires the rate itself to be specified. We agree with petitioners, since we think the provision must be read in light of the Federal Power Act's primary purpose of protecting the utility’s customers.

774 F.2d at 492-93. The court reasoned that the goal of predictability would be frustrated under FERC’s approach advocated in these cases. It specifically recognized:

It is true that the degree of unforeseeability may in some cases be minimal or even nonexistent — for example, where the only change mandated by the Commission is elimination of a single fixed-cost element in the cost of service, whose impact on the rate structure can be readily calculated.

Id. at 493. The court nevertheless refused “to assess case-by-case whether the ratemaking principles set forth by the Commission in its initial order ‘amount to the same thing” of informing the ratepayers of their filed rates." Id. In short, the court did, indeed, adopt a hard and fast rule that rates are fixed when the compliance filing is accepted.

The leeway that FERC finds in the court’s language at 774 F.2d at 494 is illusory. That language deals solely with the possibility of establishing new procedures to expedite the actual filing of rates so that they can become effective as early as possible.

[I]t might not be beyond its authority to establish expedited procedures for initial approval of rates filed by the utility on the basis of principles outlined in an earlier order, and to make the availability of those procedures contingent upon the utility’s agreement to refund any excess payments if final review of the rates produces an adjustment. Cf. Kansas Cities v. FERC, 723 F.2d at 96. Whether or not such a solution, or any solution short of new legislation, is available, the Commission cannot fix a rate, as it purports to have done here, without ever seeing it.

Id. at 494-95.

Finally, the Commission’s statement in an unrelated opinion regarding the holding of Electrical District No. I belies its position here and is consistent with our reading of the case.

[0]n October 4, 1985, the United States Court of Appeals for the District of Columbia Circuit held that a rate increase authorized under section 206 should not be made effective on the date of the Commission’s opinion prescribing the rates, but rather, such rates could become effective only as of the date of the Commission’s acceptance of the utility’s compliance filing.

Kansas Gas & Electric Co., 34 F.E.R.C. (CCH) 1161,288 at 61,517 (1986).

24 . The Commission’s opinion in Case One addressed only a just-and-reasonable-but-for-price-squeeze rate and provided that the rate increase would be subject to refund if Gallup substantiated its price-squeeze allegations. See supra section II. Gallup contended that a rate that does not reflect unresolved allegations of price squeeze is not final. The D.C. Circuit thereafter accepted the practice of allowing a rate to become provisionally effective, subject to refund if price-squeeze allegations are subsequently proved. See Kansas Cities v. FERC, 723 F.2d 82 (D.C.Cir.1983).

25 . In Asarco, Inc. v. F.E.R.C., 777 F.2d 764 (D.C.Cir.1985), the Court of Appeals for the District of Columbia rejected the assumptions that underlay its analysis in Arkansas Power & Light. Yet the court did not overrule its holding in the earlier case, "which rested not upon one of the ordinary exceptions to exhaustion (such as the futility of seeking relief from the agency) but upon what the court regarded as acknowledgment by the agency, through subsequent revision of its practice, that its action under challenge had been unlawful." 777 F.2d at 774.