STEPHEN F. WILLIAMS, Circuit Judge:

This dispute arises out of the lurching energy prices of the past two decades. An energy purchase contract that looked good for both parties on the date signed (November 4, 1985) soon afterwards looked dismal to the buyer. The Federal Energy Regulatory Commission declined to lower the price, but at the same time declined to relieve the seller from a delay of the effective date stemming from what may seem a technical default. We affirm the Commission against attack by both sides.

* * *

The contract obligates San Diego Gas Electric Company to buy 100 megawatts of generating capacity and associated energy from Public Service Company of New Mexico for a 13-year period starting May 1, 1988. At the time it was entered into, the seller was awash in excess generating capacity and the state Public Service Commission had ordered the unnecessary capacity excluded from its retail rate base. Worse still, more unwanted capacity loomed from New Mexico's interest in a nuclear generating station whose three reactors would start up over the next two years. San Diego needed more capacity and wanted to diversify its mix of power sources by securing entitlements to power from coal and nuclear plants. The price offered by New Mexico was the most favorable available to San Diego and was in line with then prevailing market prices.

November 1985 proved, however, to be something of a crest in the wave of energy prices. Oil prices fell from nearly $30 a barrel in November 1985 to $12 in March 1986. J.A. 208. The plunge triggered a drop in the bulk electric power market as well. As oil is used to produce electricity, a fall in its price (and that of close substitutes such as natural gas, whose price is directly affected by that of oil) drives power production costs down; variable costs fall and electricity producers are able to shift away from relatively capital-intensive nuclear and coal sources. J.A. 208-09, 210 n. 8. (These effects represent a downward shift of the supply curve. If the drop in oil and gas prices led bulk power users to substitute oil or gas for electricity, that effect would represent a shift along the supply curve, but the record does not appear to discuss the matter.) New contracts in 1988 evidently called for much lower prices — with demand charges (the charge for the purchasing utility's entitlement to take electricity) about half or one-third that specified in the contract.

Under § 205 of the Federal Power Act, 16 U.S.C. § 824d (1988), New Mexico could not collect rates under the contract until it had been accepted for filing with the Commission. When New Mexico filed, San Diego tried to persuade FERC to reject the contract as unjust and unreasonable, but the Commission refused. On the other hand, it also refused New Mexico a waiver that proved necessary if New Mexico was to be able to collect the demand charge starting on the date provided by contract, May 1, 1988. As a result of the refusal New Mexico was unable to collect the $3.3 million contract demand charge accrued from May 1 to June 13. Both parties appeal.

I.

San Diego attacks both FERC's way of considering New Mexico's costs in its assessment of the contract and its refusal to set it aside or modify the price on the basis of widespread changes in the energy market. In approaching New Mexico's costs, FERC initially noted that the San Diego-New Mexico deal was an "opportunity transaction," undertaken so that New Mexico could use its surplus capacity. FERC judges such sales by measuring the contract price (in excess of variable costs) against a benchmark of the contribution to fixed costs that the utility's ordinary customers would have made for the same power. Public Service Company of New Mexico, 43 FERC ¶ 61,469 at 62,150-51 (June 13, 1988) ( "Order"), citing Florida Power Light Company, 33 FERC ¶ 61,116 at 61,248 (1985). In making this calculation, FERC included the fixed costs of all New Mexico's generating plants (a capacity of 1805 MW), evidently reasoning that because the contract energy charge was to be computed by reference to all plants, the parties were assuming that all of them would provide power for San Diego. See Order, 43 FERC at 62,151. Dividing the aggregate fixed costs by New Mexico's peak demand (a mere 1050 MW), FERC calculated a benchmark of approximately $36.66 per kW/month. The contract demand charge was between 62% and 72% of this, depending on various calculations not relevant here. San Diego argued that FERC should have excluded New Mexico's surplus generating capacity from its computation of fixed costs (especially the expensive recent nuclear and coal plants), as had the New Mexico Public Service Commission in calculating the rate base for the utility's retail customers. The Commission's rejection of this claim was a policy choice that the Commission was surely free to make, not a decision unsupported by "substantial evidence," as San Diego would have it. FERC's approach accorded with its consistent precedent. Moreover, exclusion of the surplus capacity would have been "entirely inappropriate," as the Commission reasoned, because "it is the very existence of this surplus capacity which permits [New Mexico] to make this sale." Id.

In its petition for rehearing San Diego advanced an alternative method for calculating the benchmark charge. This method was similar to FERC's in that it included the fixed costs of all New Mexico's generating plants, but it spread those costs over a larger load than New Mexico actually enjoyed. San Diego suggested an imputed load of 1504 MW, which New Mexico's 1805 MW generating capacity could have sustained with a 20% reserve margin. Spreading the fixed costs of all generating units over this load, San Diego would have reduced the benchmark charge by about $11, thus making the contract price more suspect.

In its order on rehearing the Commission made no mention of this alternative method, but we do not find the omission fatal. San Diego's proposal was a clever way of juggling the figures to reduce the benchmark. No genius is needed to understand that San Diego was attempting to do indirectly what FERC said it could not do directly — penalize New Mexico for having too much surplus capacity, even though it was that surplus which made the sale possible. The alternative suggested on rehearing presented no compelling policy reason for FERC to vary its rule; having been denied the loaf, San Diego asked for half. Under these circumstances, there was no need for the Commission to have responded to the intricacies of San Diego's alternative. This is a case where silence qualifies as "tolerably terse" rather than "intolerably mute." Greater Boston Television Corp. v. FCC, 444 F.2d 841, 852 (D.C. Cir. 1970).

San Diego also claims that the Commission did not give enough weight to market changes between the execution of the agreement and the time for its performance. These changes were so great, it argues, as to require that the Commission set the contracts aside under its statutory obligation to ensure that interstate wholesale power rates are just and reasonable. See Federal Power Act § 205, 16 U.S.C. § 824d.

In fact the Commission was far from blind to the prevailing market conditions:

[G]iven our statutory obligation to judge the justness and reasonableness of the rate in the Agreement, we are required to explore [San Diego's arguments]. The freedom of the parties to contract in this instance must yield to the Commission's duty to ascertain the justness and reasonableness of the rates.

Order, 43 FERC at 62,153. Against this it placed great weight on the policy considerations behind contract stability:

The certainty and stability which stems from contract performance and enforcement is essential to an orderly bulk power market. If the integrity of contracts is undermined, business would be transacted without legally enforceable assurances and we believe that the market, the industry and ultimately the consumer would suffer.

Id.

This has long been the Commission's position. See Gulf States Utilities Company v. Southern Company Services, Inc., 43 FERC ¶ 61,003 at 61,015-16, 61,020 (1988), aff'd mem., sub nom. Gulf States Utilities Co. v. FERC, 886 F.2d 442 (D.C. Cir. 1989), cert. den. ___ U.S. ___, 110 S.Ct. 2206, 109 L.Ed.2d 532 (1990); Pennsylvania Power and Light Company, 23 FERC ¶ 61,325 at 61,716 (1983). Indeed, the Supreme Court has insisted on great Commission deference to freely arrived at contract prices. See United Gas Pipe Line Co. v. Mobile Gas Service Corp., 350 U.S. 332, 76 S.Ct. 373, 100 L.Ed. 373 (1956); FPC v. Sierra Pacific Power Co., 350 U.S. 348, 76 S.Ct. 368, 100 L.Ed. 388 (1956).

As the Commission recognized, the purpose of this contract was to allocate the risk of market price changes between the parties:

The volatility of oil and gas prices, often reflecting large and dramatic swings, is old news. Indeed, this is the very reason that San Diego pursued a long-term purchase from [New Mexico] at rates reflecting [New Mexico's] nuclear and coal units.

Order, 43 FERC at 62,153.

San Diego wanted to diversify its mix of power sources, so to reduce the risk from further surges in the cost of oil and gas. J.A. 256. But each source carries its own set of risks. That is an inevitable part of energy production in a world where sources are varied and their relative prices subject to constant change. Scanning all the sources to which it had access, San Diego selected what it viewed as the optimal mix. On the other side, the contract gave New Mexico partial protection from the risk it had taken in investing heavily in precisely the sources that San Diego (in 1985) felt it needed. San Diego's current conclusion that it got the mix wrong is no reason to allow it to shift the risk (now the loss) back to New Mexico.

We note that New Mexico is no great winner either. The capital-intensive coal and nuclear plants to which it committed itself cannot generate the once-expected revenues. Nothing suggests that New Mexico is a more efficient bearer of the risks that the literal terms of the contract shifted to San Diego. Indeed, San Diego's participation in the 1985 contract as part of a systematic, deliberate program of diversifying risks would seem to mark it as an efficient risk bearer. See Richard A. Posner and Andrew M. Rosenfield, Impossibility and Related Doctrines in Contract Law: An Economic Analysis, 6 J. Legal Stud. 83, 91-92 (1977) (promisor's inferior ability to diversify risks may make it less efficient risk bearer and justify its discharge).

San Diego invokes our decision in Associated Gas Distributors v. FERC, 824 F.2d 981 (D.C. Cir. 1987), where this court required FERC to proffer adequately "reasoned decisionmaking" for its decision not to use § 5 of the Natural Gas Act, 15 U.S.C. § 717d (1988), to relieve pipelines of enormous take-or-pay liabilities under contracts that had become uneconomic. But the root of that decision was the Commission's failure to face directly the effects of its own decision, Order No. 436, 50 Fed.Reg. 42,408, FERC Stat. Reg. [Regulation Preambles 1982-85] ¶ 30,665 (1985), restructuring the gas pipeline industry and for the first time subjecting pipelines as gas-sellers to competition from gas sold by others but transported by the pipelines themselves. As we framed the issue, it was whether the Commission's inaction on the contracts was "permissible, particularly in light of Order No. 436's shift in the balance of forces." AGD, 824 F.2d at 1021. We assessed at length the likely effects of Order No. 436, id. at 1023-25, including the pipelines' contention that they were faced with a possible "death-spiral," id. at 1024. We criticized the Commission's "seeming blindness to the possible impact of Order No. 436." Id. at 1025. Our ultimate conclusion was that on remand the Commission "must more convincingly address the magnitude of the problem and the adverse consequences likely to result from the nondiscriminatory access and CD adjustment conditions [i.e., the Commission's own regulatory interventions]." Id. at 1044 (emphasis added). And, although asking for more reasoned decisionmaking, we acknowledged that even in the context of risks that the Commission itself had severely exacerbated, FERC's pro-contract policy arguments were "powerful and well grounded in the statutes it is authorized to enforce." Id. at 1027. San Diego here offers nothing remotely comparable to the AGD petitioners' plea for regulatory relief; the Commission's reasoning amply served the case at hand.

II.

Section 205(d) of the Federal Power Act, 16 U.S.C. § 824d(d), requires that a public utility must file a proper notice with the Commission 60 days before any change in rates or service. See also 18 CFR § 35.3(a). According to the parties' agreement, New Mexico's service to San Diego was to begin on May 1, 1988. New Mexico initially filed its notice on March 1, 1988, which if complete would have satisfied the requirement in time for the intended start. On April 4, however, the Commission rejected that filing because it lacked some cost data required under part 35 of the Commission's regulations, 18 CFR §§ 35.0 et seq. See Order, 43 FERC at 62,146. New Mexico refiled, this time properly, on April 14. But now it needed a waiver if its rates were to take effect before June 14, 1988. Order, 43 FERC at 62,154. The Commission denied the waiver except for one day, allowing the rates to take effect contemporaneously with its June 13 order.

Section 205(d) allows the Commission to grant a waiver of the 60-day notice requirement "for good cause shown." The party seeking the waiver has the burden to state a "specific" and "concrete" case for its petition. United Gas Pipe Line Co. v. FERC, 707 F.2d 1507, 1511 (D.C. Cir. 1983), quoting WAIT Radio v. FCC, 418 F.2d 1153, 1157 n. 9 (D.C. Cir. 1969). Moreover, the scope of our review of an agency's denial of a waiver is extremely limited; it is a discretionary decision, vested by Congress in the agency, not the courts. City of Girard v. FERC, 790 F.2d 919, 925 (D.C. Cir. 1986). We will reverse the Commission only if it failed to give "meaningful consideration" to the waiver application. United Gas Pipe Line Co., 707 F.2d at 1511.

New Mexico opens with an argument that the Commission misconstrued the contract as not providing for a starting date of May 1. Though recognizing that the Commission characterized the May 1 date as "[a]n integral part of the bargain," New Mexico seizes (for its misconstruction theory) on FERC's observation that "the parties contemplated that this Agreement could become effective on less than 60-days notice, if the Commission agreed." Order, 43 FERC at 62,154. Far from representing a misapprehension of the contract, however, this was merely the Commission's statement of the legal truth that the contract rates could not take effect unless the 60-day notice requirement was either satisfied or waived by the Commission. Cf. Union Electric Co. v. FERC, 890 F.2d 1193, 1195-96 (D.C. Cir. 1989).

New Mexico further argues that the Commission cannot alter the contract's May 1 starting date except upon a finding that the contract is "unjust, unreasonable, unduly discriminatory or preferential" under § 206(a) of the Federal Power Act. Brief for Petitioners at 23, citing FPC v. Sierra Pacific Power Co., 350 U.S. 348, 353, 76 S.Ct. 368, 371, 100 L.Ed. 388 (1956). This ignores the independent regulatory force of the statute's 60-day notice requirement, which the parties are not free to modify by contract. While under the Mobile-Sierra doctrine the Commission generally may not modify the terms of a contract unless they "adversely affect the public interest," id. at 355, 76 S.Ct. at 372, it is bound to enforce the 60-day notice requirement, subject to the allowance for waiver.

Under the Commission's precedent, "good cause" will be found "only when the parties to the rate have agreed at some point in their negotiations on an effective date and the waiver is in the public interest." City of Girard, 790 F.2d at 925. Thus the contractual agreement to a May 1 starting date is a necessary but not sufficient condition for the waiver. See Dayton Power and Light Co., 12 FERC ¶ 61,296 at 61,679 (1980) (denying waiver even though parties had agreed on a date).

The decisive factor against a waiver seems to have been the Commission's judgment that New Mexico was responsible for its late filing because its original filing was deficient. Order, 43 FERC at 61,154; Order on Rehearing, 45 FERC at 61,120 n. 2 and accompanying text. In previous cases, the Commission has granted waivers to parties it has deemed innocent of any delay. See Towns of Concord and Wellesley, Mass. v. FERC, 844 F.2d 891, 896-97 (1st Cir. 1988) (utility not at fault for late filing where customers provided insufficient notice to utility concerning changes in service); Dayton Power Light Co., 2 FERC ¶ 61,037 at 61,090 (1978), aff'd sub nom. City of Piqua, Ohio v. FERC, 610 F.2d 950 (D.C. Cir. 1979) (utility was not at fault for late filing where it could not submit filing until its customer had completed state statutory procedure). Here we have the converse. While the penalty may be high in proportion to New Mexico's fault, we cannot call it an abuse of the Commission's discretion.

* * *

The petitions for review are Denied.