with whom WATT, C.J. and WINCHESTER, V.C.J., join, concur in part and dissent in part:

¶ 1 I dissent to the imposition of a twenty year contract term. I understand that the Corporation Commission has discretion on this issue. However, this contract is forced upon PSO due to the requirements of the 1978 federal law. It is not an arms-length contract. In view of all the law, evidence and circumstances of this case, I would hold that setting this contract term at twenty years is an abuse of discretion. A proper exercise of discretion would result in a contract being set at a much shorter term, subject to later review of the Commission.

II1 Two dispositive questions are presented for review: (1) Is the Corporation Commission’s order sustained by substantial evidence and (2) Does it conform to the requirements of federal and state law? These questions must be answered for each of the following Commission decisions: (a) determining that applicant incurred a legally enforceable obligation to deliver power to respondent, thereby obligating the Corporation Commission to set purchased power rates for the term of the proposed power sales agreement; (b) rejecting a market-based approach to the calculation of avoided costs; (c) setting an avoided capacity rate; (d) setting an avoided energy rate; (e) omitting from the power sales agreement a provision corresponding to the terms of 18 C.F.R. § 292.304(f); (f) setting a twenty-year term for the power sales agreement; (g) setting terms and conditions for the power sales agreement; and (h) altering the power sales agreement after the record was closed. We affirm the Commission’s decisions as to the issues raised in parts (a), (b), (c), (e), (f), and (h), but because the Commission failed to treat adequately issues material to the decision in parts (d) and (g), we vacate the order with regard to those parts. The Commission is directed to conduct further inquiry and make additional findings with respect to the issues raised in part (d) of this pronouncement and to adjust resolutions affirmed herein only if and to the eictent necessary to accommodate the Commission’s post-remand findings and conclusions with respect to the issues discussed in part (d).

I

ANATOMY OF THE PROCEEDINGS

¶2 Lawton Cogeneration, L.L.C. (“Law-ton” or “cogenerator”) is a limited liability company organized for the purpose of developing a cogeneration facility to be located in the Lawton Industrial Park in Lawton, Oklahoma. A cogeneration facility is a plant that produces two or more usable forms of energy, one of which is electricity.2 Lawton has been certified pursuant to the provisions of the Public Utility Regulatory Policies Act of 1978 (“PURPA”)3 as a qualifying cogeneration facility (“QF”), i.e. a cogeneration facility which meets certain standards for size, fuel use, and fuel efficiency and which is owned by a person not primarily engaged in the generation or sale of electric power.4 Lawton proposes to produce electricity and steam. It intends to sell the steam to two companies in Lawton that use steam in their operations.

¶ 3 On 23 January 2002 Lawton filed an application with the Oklahoma Corporation *868Commission (“Commission”) pursuant to the provisions of PURPA, for an order directing AEP-Public Service Company of Oklahoma (“AEP”, “AEP-PSO”, “PSO” or “the utility”)5 to purchase electric power generated by Lawton, setting purchased power rates, and approving the terms of a contract between the cogenerator and the utility. PSO moved on 29 March 2002 to dismiss the application, arguing that Lawton had not incurred a legally enforceable obligation to deliver power to PSO, a pre-condition under PURPA for obtaining a state regulatory agency’s order setting purchased power rates for the term of a power sales agreement. An administrative law judge (“ALJ”) heard oral argument on the motion and, upon finding that Lawton had not created a legally enforceable obligation to deliver power to PSO, recommended that the application be dismissed without prejudice. Lawton appealed. The Commission era banc ruled on 26 July 2002 that oral argument provided an inadequate evidentiary basis for deciding the issue raised in PSO’s motion to dismiss and remanded the cause to the ALJ to conduct a full evidentiary hearing.

¶4 After an additional legal challenge to the application’s sufficiency failed to secure its dismissal and after much wrangling over the procedural schedule and discovery issues, the cause was finally scheduled to be heard by the ALJ on 21 January 2003. Shortly before the hearing date arrived, the parties agreed to forego a hearing, submit their evidence to the ALJ by filing written testimony accompanied by exhibits, and waive cross examination of witnesses regarding their filed testimony.6 As ordered by the Commission, the record was opened on 21 January 2003 for public comment. It was again opened on 30 January 2003 for the identification of exhibits, for their placement into the record, and for the oral announcement of the ALJ’s recommendation. The ALJ issued a written Report and Recommendation on 14 February 2003, concluding that Lawton had created a legally enforceable obligation as of the date it had tendered to PSO a proposed power sales agreement and recommending a method for determining the rates PSO should pay Lawton for the purchase of the latter’s electrical output.

¶ 5 Lawton, PSO, and the Oklahoma Industrial Energy Consumers (“OIEC”), a trade organization that had earlier intervened in the cause,7 appealed to the Commission. The Commission era banc heard oral arguments on 11 March 2003. It then issued an order dated 28 April 2003, in which it reopened sua sponte the record, set a hearing date before the Commission era banc, identified witnesses whose testimony the commissioners wanted to hear, and allowed the parties to request additional witnesses be permitted to testify.

¶ 6 The Commission heard testimony on 20, 21 and 22 May 2003. On 26 November 2003, the Commission, two commissioners concurring, issued the order that is the subject of this appeal, finding that Lawton established a legally enforceable obligation no later than 26 September 2002 8 and was hence entitled to a determination — as of that date — of the rates PSO is to pay for Law-ton’s electrical power.9 The Commission then proceeded to make that determination and to order the parties, including both AEP *869and PSO, to execute the power sales agreement tendered by Lawton, as revised by the Commission,

¶ 7 PSO appealed and OIEC filed a “cross-appeal” under the same docket number.10 The Attorney General brought a separate appeal. The court consolidated the separate appeals under surviving Cause No. 100,123 and then granted a motion to retain the consolidated appeal. The City of Owasso, Oklahoma, and the City of Tulsa, Oklahoma, requested leave to file briefs amicus curiae in support of PSO and were granted leave to file a combined brief. The Lawton Fort Sill Chamber of Commerce and Industry requested and was granted leave to file a brief amicus curiae in support of Lawton. The City of Broken Arrow requested and was granted leave to join in the amicus brief to be filed by the cities of Tulsa and Owasso.

II

STANDARD OF REVIEW

¶ 8 The power to review Commission decisions is vested in this court by the Oklahoma Constitution, Art. 9, § 20.11 That provision fashions two standards of review— a de novo standard for appeals based on alleged violations of constitutional rights and a more deferential standard for all other appeals.12 Today’s pronouncement employs the more circumscribed standard, in which review extends no further than determining whether the Commission adequately performed its duty under federal and state law and whether the Commission’s findings are supported by substantial evidence.13 The term “substantial evidence” means “more than a mere scintilla”14 but may be something less than the weight of the evidence.15 It is proof that possesses something of real and relevant consequence and that carries with it a fitness to induce conviction.16 In *870testing evidence for substantiality, a reviewing court must consider not only the evidence supporting the decision, but also the evidence which detracts from it.17 In eases before the Commission involving the testimony of expert witnesses, a factual finding is supported by substantial evidence when the evidence is offered by a qualified expert who has a rational basis for his views, even if other experts disagree.18 It is for the Commission to weigh conflicting expert testimony. Because Commission decisions often involve complex issues of economics, accounting, engineering, and other special knowledge, a presumption of correctness accompanies the Commission’s findings in matters it frequently adjudicates and in which it possesses expertise.19

Ill

PURPA

¶ 9 The United States Congress enacted PURPA in 1978 in response to the nationwide energy crisis of the 1970’s. Its goal was to reduce the country’s dependence on imported fuels by encouraging the addition of cogeneration and small power production facilities to the nation’s electrical generating system. Cogeneration facilities are desirable because they are able to produce more than one form of energy at the same time with less fuel than it would take to produce them separately.20 PURPA requires electric utilities to purchase all electric energy made available by cogenerators at rates that (a) are just and reasonable to electric consumers, (b) do not discriminate against QFs, and (c) do not exceed “the incremental cost to the electric utility of alternative electric energy.”21 The incremental cost to the utility means the amount it would cost the utility to generate or purchase the electric energy but for the purchase from the eogenerator.22 The incremental cost standard is intended to leave *871ratepayers economically indifferent to the source of a utility’s energy by ensuring that the cost to the utility of purchasing power from a QF does not exceed the cost the utility would incur in the absence of the QF purchase.23 The Federal Energy Regulatory Commission (“FERC”) in 1980 issued rules implementing PURPA,24 in which it adopted what it called a utility’s “avoided costs” as the standard for implementation of the incremental cost requirement.25

¶ 10 While the applicable statutes and rules are matters of federal law, PURPA gives to state regulatory authorities the responsibility of determining a utility’s avoided costs.26 Accordingly, in 1981 the Oklahoma Legislature enacted 17 O.S. 34.1, giving the Commission the power to implement and administer PURPA.27 The Commission in turn promulgated its own rule, which imposes avoided cost informational filing requirements on utilities and provides cogenerators with rights commensurate with those granted by PURPA.28

*872IV

THE COMMISSION’S DETERMINATION THAT LAWTON INCURRED A LEGALLY ENFORCEABLE OBLIGATION TO DELIVER POWER TO PSO IS CONSISTENT WITH FEDERAL AND STATE LAW AND IS SUPPORTED BY SUBSTANTIAL EVIDENCE

¶ 11 Lawton’s application with the Commission seeks to compel PSO to purchase power from Lawton and to determine the “avoided cost” rate PSO must pay for that power. The FERC rules give QFs two options for the calculation of avoided costs. Under the first option, a QF can provide energy as the QF determines such energy to be available for purchase and have the purchasing utility’s avoided cost rates calculated at the time the power is delivered. Under the second option, a QF can provide power pursuant to a “legally enforceable obligation” over a specified term. If a qualifying facility chooses to provide power pursuant to a legally enforceable obligation, it may choose to have the purchase price based on the utility’s avoided costs calculated at the time the power is delivered or based on cost projections for the life of the obligation as calculated at the time the obligation is incurred.29 Lawton asked the Commission to find that it had incurred a legally enforceable obligation and calculate PSO’s avoided costs for the life of the obligation at the time the obligation was incurred. The Commission so found and fixed PSO’s avoided cost rates for the duration of the power sales agreement.30

¶ 12 The FERC has expressly delegated to the states the responsibility to determine whether a QF has incurred a legally enforceable obligation to deliver power and, if so, when the obligation arose.31 PSO and OIEC argue on appeal that the Commission erred in finding under state law that Lawton incurred a legally enforceable obligation entitling it to lock in avoided cost rates. PSO contends that a legally enforceable obligation to deliver power exists only if a utility is able to compel through legal process the cogener-ator’s performance of the obligation or to recover damages for the cogenerator’s failure to perform. PSO argues that it can do neither in this case because Lawton is not a viable project and because the power sales agreement Lawton tendered contains only illusory penalties for a failure to perform.

¶ 13 At the outset, we note that the creation of a legally enforceable obligation is not governed by the common law of contracts. It is a concept created by federal and state statutes, regulations and administrative rules. It is clear from these sources that electric utilities need not be willing partici*873pants in the creation of a legally enforceable obligation. Rather, a utility’s obligation to purchase power is imposed by law.32 As one court aptly described the context in which courts must assess whether a QF has incurred a legally enforceable obligation, “We are not, after all, dealing with completely free enterprise. We are, rather, dealing with the twilight world of regulated monopolies.” 33

¶ 14 We implicitly recognized in Smith Cogeneration Management v. Oklahoma Coloration Commission34 that only a viable project can incur a legally enforceable obligation. In concluding that the cogenerator in Smith had not created a legally enforceable obligation, the court noted that “Smith did not ... attempt to obtain a contract for construction, operation and maintenance of the proposed project or a contract for the purchase of natural gas.”35 Other jurisdictions have also required a degree of project development before finding that a QF is capable of incurring a legally enforceable obligation.36

¶ 15 The record in this case provides substantial evidence that significant progress has been made toward bringing the Lawton generating facility into existence. Lawton’s principals have invested significant amounts of time, effort, and money in the project. Although contracts for each and every element of the project may not have been finalized, meaningful progress has been made toward the project’s completion. The Commission’s decision that Lawton is viable is supported by substantial evidence.37

¶ 16 We also disagree with PSO’s contention that the power sales agreement approved by the Commission does not create a legally enforceable obligation because it is an “output contract.” An output contract is one in which a buyer agrees to buy a seller’s entire output of production.38 According to PSO, the power sales agreement is an output contract because “there is not a known specific amount of electric power contained in the PSA which PSO could legally require Lawton to either provide or pay damages for [its failure to provide].” PSO argues that under both the common law and the Uniform Commercial Code,39 an output contract does not create a legally enforceable obligation because it provides a remedy for non-performance only if the promisor fails to act in *874good faith. PSO insists that such a remedy is a non-remedy — it does not really obligate Lawton to do anything. Lawton responds that the PSA is not an output contract because it specifies a readily ascertainable numeric quantity of power to be delivered to PSO.40 The Commission found that the PSA specifies a numeric quantity of power to be delivered to PSO and is hence not an output contract.41

¶ 17 We need not decide whether the PSA is or is not an output contract. Assuming arguendo that it is, we hold that in the contemplation of PURPA, an output contract may give rise to a legally enforceable obligation.42 Contracts lacking a definite and certain quantity term are valid and enforceable in Oklahoma under both the common law43 and under the Uniform Commercial Code.44 Under the Uniform Commercial Code, a seller cannot tender an amount that is unreasonably disproportionate to the seller’s anticipated output as measured by any stated approximation. A stated minimum or maximum further establishes the boundaries of the agreement’s elasticity.45 The PSA provides an approximation of Lawton’s electrical output along with a minimum and maximum range of permissible deviation from that approximation. While the PSA does not provide a specific, unalterable numeric quantity, it clearly affords the necessary specificity under Oklahoma law for the agreement’s enforceability.46

¶ 18 PSO next argues that the PSA contains exclusive remedies for Lawton’s breach and that the remedies provided are illusory. Lawton responds that the remedies contained in the PSA are genuine and that in any event they are not the only remedies available to PSO. Lawton insists that nothing-in the PSA limits PSO’s resort to other remedies provided by law. The Commission agreed with Lawton. Its order states:

*875“Further, the Commission finds that the PSA tendered by Lawton does not contain any language limiting the remedies or sanctions available to AEP should Lawton fail to perform. See generally, Power Sales Agreement, § IX. Therefore, both the UCC and Oklahoma statutory and common law (regardless of which one an Oklahoma Court would apply) provide AEP adequate remedies for any breach by Lawton. These remedies are part of the total legal obligation of Lawton under the PSA. Practical Prod. Corp. v. Brightmere (sic), 1992 OK 158, 864 P.2d 330, 332 (citing 12 O.S. § l-201(ll))(sic).”47

¶ 19 We agree with the Commission’s construction of the PSA. It is a basic principle of contract law that the parties to a contract may agree to an exclusive remedy for breach, which if reasonable will be enforced and will exclude other remedies.48 At the same time, Oklahoma law recognizes a distinction between language in a contract that restricts remedies and language that actually expands rights under the contract. Provisions that expand rights give the non-defaulting party a course of action in addition to recourse through other legally available remedies.49 The Commission read the PSA as expanding the rights of PSO rather than as providing exclusive remedies. We find the Commission’s construction of the relevant provisions of the PSA to be reasonable.

V

AVOIDED COSTS

¶ 20 A critical matter in controversy in this case is the Commission’s calculation of PSO’s avoided costs. Some background information is necessary for an understanding of the issues raised. The FERC rules recognize two categories of costs avoided by a utility when it purchases power from a QF; “capacity costs” and “energy costs.”50 Capacity costs generally represent the fixed capital costs of a generating facility. These include the costs of constructing a plant, installing generating equipment, and the financial carrying costs of the utility’s investment in the plant. These costs do not vary with changes in the plant’s actual production.51 Energy costs are the variable costs of operating, maintaining and providing fuel to the plant. Energy costs vary depending on actual generation, i.e. they increase or decrease according to the amount of fuel consumed and the cost of operating and maintaining the plant.52

¶ 21 Capacity and energy can be self-generated or purchased. They are self-generated when a utility builds a new generating unit and places its production on its own system. Capacity purchases occur when a utility buys the right to call on the resources of another generating entity if the purchasing utility’s own ability to generate electrical energy is insufficient to satisfy its obligations. Energy purchases occur when a utility buys electricity from another generator. Energy purchases may be made to satisfy the purchasing utility’s obligations or they may be made solely because it is more economical at a certain point in time for the utility to buy energy than to generate it itself.

¶ 22 A QF is only entitled to an avoided capacity payment from a utility if the purchase of the QF’s capacity permits the utility to avoid building additional capacity of its own or purchasing it from another1 source.53 The state regulatory authority *876must hence consider whether a utility needs additional capacity and, if so, what type of capacity is needed. In contrast, a QF is always entitled to a payment reflecting avoided energy costs. This is so because a utility can always avoid costs associated with the production of energy by decreasing the operation of one or more of its own units or by foregoing an energy purchase and replacing that energy with energy from the QF.

¶23 Implementation of the avoided cost standard has proved quite problematic. States use a variety of methodologies for calculating avoided costs and their results have often either overestimated or underestimated the costs utilities actually avoid by purchasing energy from a QF. In 1988 the FERC issued a Notice of Proposed Rulemak-ing entitled “Administrative Determination of Full Avoided Costs, Sales of Power to Qualifying Facilities, and Interconnection Facilities ”54 (the “NOPR”), in which it addressed some of the problems encountered in determining avoided costs. One of its proposals was to have wholesale purchases play a greater role in the avoided cost determination. The FERC terminated the NOPR in 1998 without adopting new rules based on the ideas it expressed. Nevertheless, in 1995 in Southern California Edison Co.55 the FERC issued a ruling disapproving of an avoided cost determination that did not take into account “all sources of generation capacity” available to the purchasing utility.56

¶ 24 It is important to note that the FERC has never limited states to a single methodology for determining avoided costs. Each state regulatory authority continues to have its own rules and regulations and its own methodology for implementing PURPA. As long as the method employed both reasonably accounts for a utility’s avoided costs and encourages cogeneration57 it will be deemed in compliance with PURPA even if the avoided cost estimate differs from actual avoided costs at the time the energy is delivered.58

VI

THE COMMISSION’S REFUSAL TO BASE AVOIDED COSTS ON WHOLESALE MARKET PRICES IS SUPPORTED BY SUBSTANTIAL EVIDENCE AND IS CONSISTENT WITH PURPA

¶ 25 Appellants contend that if PSO is required to purchase Lawton’s power at the avoided cost rates ordered by the Commission, ratepayers will be subjected to un*877necessary and costly rate increases. Appellants argue that the Commission made a critical error by refusing to calculate avoided costs in accordance with the evidence PSO tendered establishing the availability of low cost electrical power for purchase in the wholesale electricity market and by opting instead to base avoided costs on the addition to PSO’s system of a hypothetical generating unit.

¶ 26 PSO witnesses testified that for the foreseeable future its resource plan is to purchase any needed capacity in the market rather than to build new generating plants. PSO witnesses also testified that the utility intends to supply its energy needs with purchases whenever that option is more economical than operating its own generating units. In support of this plan, appellants offered evidence that there is a glut of energy in the market resulting in market prices that are lower than the costs PSO would incur if it were to build and operate a plant of its own. Several witnesses testified that the chance of prices rising significantly in the next five to ten years is remote because it will take that long to work off the market surplus, making PSO’s planned reliance on market purchases both sensible and prudent.

¶27 Accordingly, PSO proposed that its avoided capacity costs should be calculated based on market prices, which would be determined through a competitive bidding process overseen by the Commission or from “offers” PSO had already received from market participants. PSO also proposed that its avoided energy costs be tied to market prices obtained from industry publications that provide forecasts of energy prices several years into the future. PSO advocated a contract term of no more than five years, corresponding to its prediction that market prices will remain the least cost alternative for at least that period of time.

¶ 28 The Commission’s Staff witness testified that he could neither support nor refute the claim that market prices would remain low for three to five years, but testified that markets are volatile in both pnce and power availability. He further testified that it is inappropriate and risky for a utility to rely on purchases at projected wholesale market prices for a significant portion of its resource needs without locking in a price and ensuring availability through a contract or other binding agreement. Although the Staff witness recommended using market prices to establish avoided costs for the first three to five years of the power sales agreement, he told the Commission that relying on the market for more than six to twelve months into the future is dangerous. A PSO witness agreed that energy markets are subject to “significant uncertainty and variability.”

¶ 29 The Commission found PSO’s reliance on market purchases for its resource plan to be an unacceptable means of providing for its future generation needs. The Commission criticized PSO for failing to offer any “specific long-range strategy in the way of planned generation assets ... other than simply further rebanee upon purchased power.” The Commission found that the market is volatile and that PSO’s reliance on market purchases is short-sighted and risky. With regard to the “offers” PSO received, the Commission found them to be unreliable evidence of future market prices and insufficiently detailed to be useful.59 The Commission also rejected competitive bidding as a method for determining market prices, concluding that there are numerous unresolved problems with instituting such a process.60 While rejecting *878the recommendation of its own staff witness to base avoided costs on the market for the first few years of the power sales agreement, the Commission was clearly persuaded by that witness’s testimony regarding the inadequacy of PSO’s resource plan and the volatility of the market.

¶ 30 We are presented here with highly conflicting evidence in a matter within the Commission’s expertise. Much of the evidence consists of conflicting opinions by experts as to the volatility of the electricity market and what effect market volatility should have on the Commission’s evaluation of PSO’s resource plan and its view of the market’s usefulness in determining avoided costs. While we might also have viewed a different outcome as supported by substantial evidence, we cannot say that the Commission’s decision to reject the market as the basis for PSO’s avoided costs does not rest on a substantial evidentiary basis. Qualified experts having a rational basis for their view offered evidence that supports the Commission’s decision. It is not for this court to reweigh the proof and substitute its judgment for that of the Commission as to where the weightier evidence lies. The Commission’s view of the electricity market and of its own ability to oversee a market-based procedure for determining avoided costs is within the Commission’s special knowledge as regulators in this field and we will not reverse its decision in the absence of a compelling reason to do so.

¶31 Finally, we do not agree with PSO that PURPA requires the Commission to use wholesale market prices to determine avoided costs. FERC has stated that:

“there is no requirement in our regulations that avoided costs be established through competitive bidding or other competitive procurement mechanisms. Our existing regulations permit avoided cost to be established administratively, so long as all alternative sources of electric energy, i.e., all resource technologies and all types of sellers (QF and non-QF), are taken into account.”61

¶ 32 The FERC’s proviso that all alternative sources should be taken into account does not mean that every alternative source proposed by one party or another must be utilized in determining avoided costs. A source may be “taken into account” by dismissing it as inappropriate. In other words, state regulatory authorities retain discretion to determine what sources of capacity and energy should be considered in determining avoided costs.62

VII

THE COMMISSION’S DETERMINATION OF AVOIDED CAPACITY COSTS IS SUPPORTED BY SUBSTANTIAL EVIDENCE

¶ 33 The determination of a utility’s avoided capacity costs begins with an analysis of the utility’s capability to meet the demand made on it for electricity. Demand for electricity varies hourly and seasonally. A utility must have sufficient generating capability to meet the maximum demand, whenever that occurs.63 Because electricity cannot be produced in advance and be then stored, a utility must have generation resources available to meet periods of peak demand.64 Consequently, a portion of a utility’s generating units will stand idle most of the day or year, *879coming online only during periods of peak demand.65

¶ 34 To economically deal with the situation of variable demand, utilities use three types of generating units that optimally match costs with usage. Baseload plants are designed to operate continuously to meet a system’s minimum load. They are expensive to build, but because they use low cost fuels, they are relatively inexpensive to operate.66 In contrast, peaking units are designed to run for only short periods of time when demand for electricity is at its highest.67 They are less expensive to build than base-load plants, but because their fuel costs are higher, they are more expensive to operate.68 The third category of generating units are intermediate plants, which as the name suggests operate more than a baseload unit and less than a peaking unit with commensurate costs.69

¶ 35 Based on evidence of PSO’s capability to meet projected future demand, the Commission found that PSO will need to secure additional peaking capacity beginning in 2005 and that its need for capacity to meet peak demand and reserves will increase each year thereafter. Accordingly, the Commission found that PSO needs the capacity Law-ton will provide. Having rejected competitive bidding and the “offers” received by PSO as means of determining avoided costs, the Commission selected a newly constructed combustion turbine peaking plant as the most efficient alternative source of supply of PSO’s next unit of required capacity. These findings and conclusions are supported by substantial evidence.

¶ 36 The Commission found that a levelized capacity payment of $77.01/kW/year was a reasonable estimate of the capacity costs of a peaking plant. This amount was consistent with the testimony of expert witnesses regarding the capacity costs of a peaking plant. We hold that the Commission’s findings and conclusion as to PSO’s avoided capacity costs are supported by substantial evidence.

¶ 37 Although PSO does not dispute that its future need is for peaking capacity, it suggests that it might be more appropriate in this case to use the costs avoided by the addition of a hypothetical baseload unit on its system. This argument is based on the fact that the power sales agreement approved by the Commission requires PSO to accept power from Lawton not just at hours of peak demand when PSO needs Lawton’s power to meet demand, but all day every day of the year.70 In other words, PSO’s need for peaking capacity is not matched by the amount of power it will have to accept from Lawton. PSO argues that the amount of power it must take from Lawton is characteristic of a baseload unit, not a peaking unit, so that regardless of what its future needs may be, its avoided costs for taking Lawton’s power would be better reflected by using a baseload unit as the proxy. As PSO points out, energy costs make up a large percentage of total avoided costs. Hence, while the fixed costs of a baseload unit would be considerably greater than those of a peaking unit, over the course of the contract’s term the higher ca-paeity(fixed) costs of a baseload unit would be more than offset by that unit’s lower operating costs. The Commission’s staff witness in fact recommended that a baseload unit be used as a proxy for avoided costs during a portion of the contract’s term.

¶ 38 In the next part of this pronouncement, we hold that on remand the Commission must revisit its determination of PSO’s avoided energy costs. While the evidence of PSO’s future capacity needs clearly supports the Commission’s decision to base avoided *880capacity costs on the capital costs of a peaking unit, reconsideration of PSO’s avoided energy costs may create a need to reconsider the use of a peaking unit as the proxy for avoided capacity costs during the entire contract term. We are not directing the Commission to make any particular decision in this regard, but are merely giving the commissioners permission to revisit this issue if necessary. Regardless of how energy payments are determined, it may well be that the lower costs of building a peaking unit make that unit a more accurate measure of capacity costs in current dollars than generating units that take longer to build, are more complex, and have greater exposure to regulatory and environmental measures.71 We simply want to make clear that while we affirm the Commission’s use of a proxy peaking unit for capacity costs, we are not forbidding the Commission to revisit the proxy unit chosen for capacity payment purposes to the extent that the Commission determines such revisiting of the subject is necessitated by its reconsideration of PSO’s avoided energy costs.

VIII

THE COMMISSION’S APPROACH TO PSO’S AVOIDED ENERGY COSTS IS NOT SUPPORTED BY SUBSTANTIAL EVIDENCE

¶ 39 Energy costs are the variable costs of operating, maintaining and providing fuel to an electrical generating unit. They are variable because they change with production. Energy costs have two principal components: a heat rate and a fuel cost. Heat rate measures the number of British Thermal Units (Btu)72 necessary to generate one kilowatt-hour 73 of electricity. Heat rate is represented as x Btu/KWh. The lower a generating unit’s heat rate, the less fuel it burns to generate electricity and the less it costs to operate. For example, a generating unit that is able to produce one kilowatt-hour of electricity with 7,000 Btu of fuel is 30% less costly to operate than a generating unit that requires 10,000 Btu of fuel to produce one kilowatt-hour. Fuels used for the production of electricity include natural gas, coal, hydro power, and nuclear power. Fuels vary widely in cost. Energy cost is a function of the price of the fuel used by a particular generating unit and that unit’s heat rate.

¶ 40 As stated earlier, the quantity of power a utility produces varies hourly and seasonally depending on demand. In order to produce the quantities of electricity necessary to meet variations in demand, a utility uses different combinations of fuel and equipment. When demand for electricity is low a utility can “dispatch”, or operate, its lowest operating-cost units and back down, or turn off, its higher cost units. As demand increases, a utility dispatches its more costly units, beginning with the least costly and moving to the most costly. When load is at its highest, a utility must use units with the highest operating costs. Each increase in the use of a higher cost unit increases the marginal or incremental cost of producing electricity. Proper dispatch decisions result in lower incremental energy costs. Dispatch decisions are made on an hourly and daily basis.

¶ 41 In making dispatch decisions, a utility considers not only the economics of using its own generating units, but also whether to purchase electricity produced by other producers who may be willing to sell electricity for less than what it would cost the utility to operate its less efficient, more expensive gen*881erating units. These purchases, known as “economy energy purchases,” are another way a utility can lower its incremental cost of producing electricity.

¶ 42 The Commission’s order establishes two heat rates for the determination of PSO’s avoided energy costs: 10,800 Btu/kWh for Summer On-Peak Hours and 10,460 Btu/ kWh for all other hours of the year.74 In support of its decision, the Commission states:

“These avoided heat rates are lower than information from the Oak Ridge National Laboratory and the Department of Energy’s Energy Information Administration regarding expected CT [combustion turbine] peaking heat rates, and therefore, is a conservative calculation. These heat rates are also close to AEP’s current year round system average gas heat rate and will make Lawton the third most efficient unit out of all of AEP’s total gas fired generation in terms of efficiency, which will benefit AEP’s customers significantly.”

¶43 The Commission notes that none of the parties in this case gave it much help in determining the avoided heat rate.

“The real challenge in determining the proper heat rate stems from the parties’ reluctance to submit evidence relating to how Lawton’s facility would be dispatched in the absence of uncommitted short-term purchases. The ‘avoided heat rate’ depends on the point in time at which Law-ton’s unit would be dispatched, and the heat rate of the units that Lawton’s facility would be displacing. Such information is necessary when evaluating PSO’s resource options and the type of unit being avoided in order to establish avoided costs. PSO so focused on using market purchases to meet its future capacity and energy needs that it offered little evidence in the way of avoided proxy unit costs in the absence of short term market purchases.”

¶44 The Commission nevertheless contends that its dual heat ratfes apply principles of economic dispatch by ordering different heat rates for summer on-peak hours and all other hours. The Commission explains that by differentiating between summer on-peak hours and all other hours, its decision reflects “the variability of the cost of producing electricity during different hours of the day and seasons of the year.”

¶ 45 Appellants argue that the Commission-ordered heat rates do not apply economic dispatch principles in any meaningful way. PSO agrees that it cannot physically dispatch Lawton to run only the number of hours it would run under economic dispatch principles,75 but it argues that Lawton’s energy should be priced as if the QF were dis-patchable. The utility contends that if it economically dispatched Lawton based on the Commission-ordered heat rate, Lawton would run no more than 15% of the time.76

The Commission counters that the heat rates it ordered are in fact the heat rates of a dispatchable peaking unit.

¶ 46 The Commission’s order moves seamlessly from finding that a peaking unit would be the most appropriate capacity-addition to PSO’s system to using the heat rates of a peaking unit as the basis for calculating ener*882gy payments. There seems to be an underlying assumption in the Commission’s order, the basis of which is neither expressed nor evident to the court, that if a peaking unit is used for capacity payments, then a peaking unit’s heat rate must serve as the basis for enei'gy payments. Utilizing a single proxy unit for both calculations makes sense if the utility’s energy needs match the cogenerator’s energy output, but where as here the cogenerator will supply a great deal more than the utility’s capacity requirements, the rationale for using a unit specific approach to both capacity and energy payments is not clear. Hence, while the record does indeed contain evidence that the non-dispatchable heat rate of a peaking unit is 10,456 Btu/kWh at non-summer conditions and 10,800 Btu/ kWh at summer conditions, we fail to see how these ratings relate to the actual generation mix and the resulting incremental costs that would arise from the addition of Law-ton’s power to PSO’s system.

¶ 47 The Commission also says that the heat rates it ordered are “close to AEP’s current year round system average gas heat rate and will make Lawton the third most efficient unit out of all of AEP’s total gas fired generation in terms of efficiency, ...” This argument at least has the benefit of looking at the operation of PSO’s system, but it too falls short of producing conviction that the ordered heat rates are sustainable. First, the Commission neither provides the number representing PSO’s actual historical system average gas heat rate for the court to use as a comparison with the ordered heat rates nor does the Commission identify where in the record that information can be found. Second, even if the Commission determined what PSO’s historical system average annual gas heat rate was, it failed to explain how that historical figure relates to the avoided incremental heat rate that PSO will experience by Lawton’s displacement of PSO generation.

¶ 48 Lawton argues that the Commission-ordered heat rates should be affirmed because they are lower than the average avoided incremental heat rate produced by the eogenerator’s computer simulation of PSO’s system. Lawton’s computer modeling was rejected by the Commission along with the computer modeling done by PSO. The Commission found that both parties had skewed the inputs into the computer modeling in ways that made them unreliable guides to avoided costs. We have no reason to second-guess the Commission’s assessment of the parties’ computer models. Hence the fact that the Commission-ordered heat rates are lower than the heat rate Lawton produced in its computer modeling does not convince us that the Commission’s heat rates are appropriate.

¶ 49 Finally, although the record does show that both a Commission staff witness and a PSO witness proposed using a heat rate of 10,500 Btu/kWh, we do not believe that their testimony supports the conclusion that the Commission-ordered heat rates comply with the PURPA standard. This is so because the former used that heat rate only in conjunction with a 10% capacity factor and the latter for just a 13-day peaking capacity purchase. Their testimony does not compel the conclusion that a heat rate in the neighborhood of 10,500 Btu/kWh reflects PSO’s incremental avoided heat rate for the Lawton purchase.

¶ 50 In reviewing the Commission-ordered heat rates, we are confronted with the task of assessing a Commission decision on a technical matter involving terminology, concepts, and data commonly used by electrical engineers and economists, but not easily understood by those outside those professions. Yet in the final analysis, it is a judicial decision that we are called upon to make, not a decision as engineers or economists. The engineers and economists in this case had widely diverging views. Our task is to determine whether the Commission’s decision is supported by substantial record evidence. It may be that the Commission-ordered heat rates would result in avoided energy payments in line with the requirements of PURPA, but from the record provided we are unable to ascertain that this is so. On remand of this cause, it is imperative that the Commission identify in its order the evidence that serves as the factual basis for the heat rate or heat rates chosen and specify how its *883decision reflects the incremental energy costs PSO will avoid.

¶ 51 With respect to the avoided fuel, the Commission found that gas is the only fuel that would be avoided by the addition of a peaking unit to PSO’s system and ordered the cost of gas to be derived from PSO’s weighted average cost of gas (“WACOG”) as shown on its monthly Fuel Adjustment Clause filings with the Commission.77 In support of this decision, the Commission states:

“The Commission finds that the use of WACOG to set the avoided price for fuel is a very conservative and customer protective approach that will result in energy payments below AEP’s full avoided cost, i.e. below AEP’s incremental highest priced gas avoided by AEP for its gas purchases.”

¶ 52 PSO asserts that natural gas is not the only fuel avoided by a round-the-clock purchase from Lawton and that the weighted average cost of that fuel is not an accurate measurement of PSO’s incremental cost. Post-remand reconsideration of PSO’s avoided energy costs will necessitate a reconsideration of the avoided fuel. Accordingly, we decline to review this matter until the Commission has completed the further inquiry required by today’s pronouncement.

IX

THE ABSENCE OF A PROVISION IN THE PSA CORRESPONDING TO 18 C.F.R. § 292.304(f) DOES NOT VIOLATE PURPA OR THE FERC REGULATIONS IMPLEMENTING IT

¶ 53 PSO argues that the Commission’s order violates PURPA by approving a power sales agreement that does not include a clause corresponding to the provisions of 18 C.F.R. § 292.304(f). The provisions of § 292.304(f)(1) permit an electric utility to suspend purchases of' energy or capacity from a qualifying facility when “operational circumstances” make such purchases more expensive than generating the electricity itself.78 The provisions of § 292.304(f)(2) require a utility exercising its rights under § 292.304(f)(1) to notify the affected QF of its decision in time for the QF to cease delivering power.79 This provision would be of no use unless utilities could actually cease purchasing QF power during “operational circumstance” periods. The regulation is hence an exception to the PURPA requirement that electric utilities purchase the entire output of qualifying facilities.80

*884¶ 54 We disagree with PSO that a provision incorporating the terms of § 292.304(f) must be included in a QF contract. Extant applicable law is a part of every contract in this state as if it were expressly cited or its terms incorporated in the contract.81 An intent to modify applicable law by contract is not effective unless the power is expressly exercised.82 A contractual adjustment of rights contrary to law must be clearly expressed in the agreement if applicable law is not to be applied.83 Hence, the provisions of § 292.304(f) remain available to PSO regardless of whether they are explicitly included in the contract.

¶55 The Commission advocates an alternate basis for upholding its decision. We ordinarily would not consider this argument, having just affirmed the Commission’s decision on other grounds. Yet in this ease we choose to address the Commission’s argument because the post-remand inquiry we order today will necessarily entail the Commission’s reconsideration of how best to account for the operational circumstance periods covered by § 292.304(f). We deem it prudent to offer some guidance to the Commission on this point.

¶ 56 The Commission argues that the power sales agreement under review does not have to contain a provision corresponding to § 292.304(f) because the Commission made that regulation moot by taking operational circumstances into account through the prescribed heat rates.84 PSO counters that the record contains no evidence to support the Commission’s contention that the prescribed heat rates reflect an adjustment to account for the periods of operational circumstances contemplated by 292.304(f). While we agree with the Commission that purchase rates may take periods of operational circumstances into account, thereby rendering moot the provisions of § 292.304(f),85 we agree with PSO that the record in this case does not provide substantial evidentiary support for the Commission’s contention.

¶ 57 To summarize, the provisions of § 292.304(f) remain available to a utility even if its terms are not expressly included in the power sales agreement, hut its provisions may not he utilized by the utility if operational circumstances have already been taken into account in calculating the utility’s avoided costs. Thus, should PSO ever invoke the provisions of § 292.304(f), its availability will depend on whether the purchase rates ordered by the Commission already take into account periods of operational circumstances. To avoid confusion and prevent future litigation over this issue, we urge that, on remand, if it takes operational circumstances into account through the avoided cost rate structure, the Commission specify on remand what adjustment was made and where in the record evidentiary support for the decision can be found.

*885x

THE CONTRACT TERM ORDERED BY THE COMMISSION IS SUPPORTED BY SUBSTANTIAL EVIDENCE AND IS CONSISTENT WITH FEDERAL LAW

¶ 58 Lawton asked the Commission to set the term of the PSA at twenty-five years. The Commission rejected this request and set the term at twenty years. PSO argues that the Commission-ordered contract term merely exacerbates the contract’s substantive inequities and inefficiencies in violation of PURPA and the FERC’s regulations. The Attorney General argues that ratepayers will be harmed by a contract of twenty years duration that reduces PSO’s flexibility to rely on market purchases because market purchases will provide electricity to consumers at a lower cost than the PSA. OIEC argues that in today’s market a long-term contract is a contract of three to five years duration. Lawton responds that the twenty year term ordered by the Commission is supported by substantial evidence and is consistent with PURPA and the FERC rules implementing it. Lawton points to testimony from Commission Staff that a short-term contract would discourage cogen-eration in Oklahoma and that a long-term contract is desirable to protect ratepayers from the volatility of the electricity market.

¶ 59 In Smith,86 we held that a qualifying facility is entitled to full avoided costs set for the duration of a long-term contract,87 but did not address what constitutes a long term contract. Neither PURPA nor the FERC rules require any particular contract length, leaving the decision on this issue to the discretion of the state regulatory authority to be resolved on a case-by-case basis. Evidence was introduced in this case that in light of market forecasts, five years constitutes the long term, but the Commission, relying on other evidence, was unwilling to rely on predictions of future market conditions. Instead, the Commission concluded that the ability of PSO to purchase power at a known, set price for twenty years would provide greater protection to ratepayers while at the same time promoting the goal of PURPA to encourage cogeneration. The Commission’s choice of a twenty-year term does not violate PURPA or the FERC rules implementing it and is supported by the record.

XI

THE COMMISSION HAS EXTENSIVE AUTHORITY UNDER PURPA TO DETERMINE THE TERMS AND CONDITIONS OF QF CONTRACTS, BUT CERTAIN PARTS OF ITS ORDER EXCEED THAT AUTHORITY

¶ 60 PSO argues that certain provisions of the PSA approved by the Commission imper-missibly interfere with the utility’s internal business decisions. Such interference has on more than one occasion met with this court’s disapproval. As early as 1934, in Lone Star Gas Company v. Corporation Commission,88 we said of the powers of the Commission:

The powers of the Commission are to regulate, supervise, and control the public service companies in their services and rates, but these powers do not extend to an invasion of the discretion vested in the corporate management. It does not include the power to approve or disapprove contracts about to be entered into, nor to the approval or veto of expenditures proposed.89

In Oklahoma Gas & Electric Company v. Corporation Commission>90 we held that the Constitution does not clothe the Commission with the general power of internal management and control of the utilities it regulates.91 In Public Service Company v. State,92 the Commission sought to prevent a utility’s construction of a plant by refusing to permit the issuance of securities. We held *886that this was an impermissible interference with the utility’s internal business decisions.93

¶ 61 In each of these decisions' the power asserted by the Commission and rejected by the court originated in state constitutional or statutory law. These cases are inapplicable to the Commission’s exercise of authority under PURPA to order a utility to enter into a contract with a qualifying facility. We recognized this distinction in Smith.94 The utility in that case argued that the amount of its generating capacity and associated costs is an internal management decision beyond the jurisdictional reach of the Commission. We disagreed, holding that the Commission had subject matter jurisdiction to set reserve margins and establish capacity needs insofar as doing so was for the purpose of furthering the federal policy objectives of PURPA.

¶ 62 By filing an application with the Commission pursuant to PURPA and submitting a proposed contract, Lawton invoked the full power and duty of the Commission to examine all of the contract’s terms and conditions for compliance with PURPA and the FERC regulations implementing it. FERC has recognized that the States have this authority. In rejecting a utility’s challenge to a state regulatory authority’s determination that a legally enforceable obligation had been incurred, the FERC noted that the specific provisions of QF contracts are up to the States to determine:

“It is up to the States, not this Commission, to determine the specific parameters of individual QF power purchase agreements, including the date at which a legally enforceable obligation is incurred under State law. Similarly, whether the particular facts applicable to an individual QF necessitate modifications of other terms and conditions of the QF’s contract with the purchasing utility is a matter for the States to determine. This Commission does not intend to adjudicate the specific provisions of individual QF contracts.”95 (emphasis added)

Cogenerators have the right to have disputes settled by the Commission.96 The Commission did not exceed its authority under PURPA or under state law and regulation in reviewing the entire proposed contract and determining the provisions that would be included in the final version.

¶ 63 PSO also contends that the Commission’s order that PSO treat Lawton as a “network resource” not only interferes with management discretion, but also exceeds the Commission’s jurisdiction because network service is part of PSO’s FERC-approved Open Access Transmission Tariff.97 Lawton *887says that the Commission’s directive merely ensures “that AEP-PSO will not be able to impair the operation of the Lawton Facility (including the right to produce and deliver its energy) and that the Facility will be treated fairly by AEP-PSO.” The Commission’s appeal briefs fail to specifically address this issue, instead including it within its general defense of its decisions as not encroaching upon management discretion.

¶ 64 Network service is a type of relationship a transmission provider may have with a customer for either interconnection services or transmission services or both. Network Resource Interconnection Service requires the transmission provider to build the network upgrades which will allow an interconnection customer to designate a generating facility as a network resource and obtain Network Integration Transmission Service.98 When an electric utility purchases a QF’s total output, it is obligated to interconnect under the provisions of 18 C.F.R. § 292.303 and the relevant state agency exercises authority over the interconnection and the allocation of interconnection costs.99 Network Integration Transmission Service is a transmission or delivery service that places network customers on a footing comparable to that of the transmission provider on the transmission provider’s transmission system.100 Network resource status for interconnection service does not convey transmission service.101 The interstate transmission of electricity within the jurisdictional boundaries of the FERC.102

¶ 65 The Commission’s directive that PSO treat Lawton as a network resource occurs within a discussion of matters relating to interconnection but under a general heading referring to both transmission and interconnection. It is unclear to the court precisely what the Commission is ordering PSO to do. The Commission’s order does not delineate the nature and scope of the Commission’s directive or point to anything in the record supporting its decision. Its appeal briefs, too, shed no light on the issue. The Commission must convey in the order enough information that the court can determine whether the findings are supported by the law and substantial evidence.103 Findings made in general terms are insufficient.104 The Commission’s finding that PSO must treat Lawton as a network resource does not meet this standard.

¶ 66 Finally, PSO argues that the Commission exceeded its authority by ordering PSO’s parent company, AEP, to execute the power sales agreement along with PSO. PSO points out that both PURPA and the FERC’s implementing regulations limit the QF purchase obligation to electric utilities.105 *888PSO argues that while AEP may be its parent company, that does not make it an electric utility.

¶ 67 The Commission argues in its brief that the power sales agreement can only be fully implemented if another AEP subsidiary, American Electric Power Service Corporation, carries out some of its provisions. The Commission contends that for the latter’s cooperation to be assured it is necessary to require AEP, the parent of both PSO and American Electric Power Service Corporation, to join in the agreement. The order makes no findings of fact and points out nothing in the record that supports this contention. Further, PURPA and the FERC implementing regulations clearly assign the QF purchase obligation to electric utilities. The order contains no finding that AEP is an electric utility rather than a company that owns electric utilities, that AEP and PSO are not separate entities, or that they should not be treated as separate entities. On this record, the Commission’s order to AEP to execute the agreement cannot be sustained.

XII

THE COMMISSION DID NOT UNILATERALLY CHANGE THE POWER SALES AGREEMENT IN VIOLATION OF THE LAW

¶ 68 PSO argues that the Commission added a new definition to the final version of the PSA and altered another provision without giving the constitutionally required notice to PSO106 and without substantial support in the record. In the version of the PSA originally tendered by Lawton, PSO was allowed to reduce capacity payments if Lawton did not perform at a 92.3% capacity factor during the summer on-peak hours. PSO argues that the Commission altered this provision in the final version of the PSA to give Lawton the entire year in which to meet this capacity factor, permitting Lawton to receive full capacity payments even if capacity is unavailable when PSO most needs it — during the summer when it experiences peak load.107

¶ 69 We disagree that the PSA was changed without notice to PSO. Inherent in the regulatory authority the Commission has over avoided cost determinations is the authority to determine related terms and conditions to be included in power sales agreements. PSO had notice of the nature of the proceeding before the Commission and fully participated in it. Its contention that it did not have notice that this particular provision would be subject to the Commission’s authority to approve QF contract terms is without merit.

¶ 70 The change in the testing temperature ordered by the Commission is matched in the final version of the PSA by a corresponding increase in the QF’s net electrical generating capacity. That capacity must be made available to PSO year round. The Commission hence required capacity testing to be based on total annual hours rather than on summer hours. The Commission has authority under PURPA to make decisions of this nature and so long as they are supported by substantial evidence, they will not be disturbed on appeal.

*889XIII

SUMMARY

¶ 71 The Commission’s order stands on firm legal support as to issues resolved in Parts IV, VI, VII, IX, X, and XII and is affirmed insofar as it determines the issues discussed in these parts. Because the Commission failed to provide substantial record support of its calculation of avoided energy-costs, and because certain of its directives are not sufficiently detailed to provide a basis for determining their conformity to PURPA and FERC regulations, the order is vacated as to the issues addressed in parts VIII and XI. The proceeding is remanded to the Commission for further inquiry to be conducted, and findings to be made, in a manner conformable to directions given in this pronouncement.

¶ 72 ORDER AFFIRMED IN PART AND VACATED IN PART; PROCEEDING REMANDED WITH DIRECTIONS TO CONDUCT FURTHER INQUIRY AND MAKE ADDITIONAL FINDINGS.