Petitioner Brazos Electric Power Cooperative, Inc. (“Brazos”) seeks review of an order of the Federal Energy Regulatory Commission (“FERC,” or “the Commission”) denying Brazos’ motion and petition to revoke the certification of Tenaska IV Texas Partners, Ltd. (“Tenaska”) as a “qualifying cogeneration facility” under the Public Utilities Regulatory Policies Act of 1978. We deny the petition for review.
I.
Tenaska is a privately-held partnership engaged in the production of wholesale *237electric power. Tenaska developed and owns a cogeneration plant in Cleburne, Texas. A cogeneration plant is a facility which produces electric energy and either steam or some other form of useful energy which is used for commercial, industrial, heating, or cooling purposes. See 16 U.S.C. § 796(18)(A). Brazos is an electric utility cooperative engaged in the generation and transmission of electric power. The utility is comprised of individual electric cooperatives in Texas and provides power to those cooperatives. Currently, Brazos is purchasing electricity from Te-naska pursuant to the facilities’ Power Purchase Agreement. The Power Purchase Agreement was certified under a Texas statute that granted certification of such contracts only if the cogeneration facility met the requirements of the Public Utilities Regulatory Policies Act of 1978 (“PURPA”), 16 U.S.C. § 823a et seq. Brazos seeks to undo the contract, arguing that Tenaska no longer meets PURPA’s requirements.
A further understanding of the facts of this case requires some explanation of the statutes and regulations that control the relationship between a private producer such as Tenaska and public utility corporation such as Brazos. PURPA was enacted in response to the nation’s fuel shortage, and its primary aim was to promote conservation of oil and natural gas in electricity generation. See FERC v. Mississippi, 456 U.S. 742, 745-46, 102 S.Ct. 2126, 72 L.Ed.2d 532 (1982). To those ends, PURPA required FERC to promulgate rules encouraging the development of alternative generators of electricity, such as cogeneration facilities. See 16 U.S.C. § 824a-3(a). The rationale behind encouraging cogeneration is that the production of electricity frequently results in the production of thermal energy as a byproduct; by using small amounts of additional fuel, cogenerators can produce large amounts of thermal energy to be used in other processes. Congress created regulatory benefits to provide economic encouragement to such nontraditional power producers. For example, qualifying co-generators are exempt from wholesale rate regulation under all federal and state public utility statutes, see 18 C.F.R. §§ 292.601, 292.602, and utilities can be compelled to interconnect with them, paying rates no greater than the utility’s full avoided costs, see 18 C.F.R. §§ 292.303, 292.308, 292.101(b). In this way, PURPA ensures the cogenerator a market for its electricity production and allows it to make a profit when it can produce power at an average cost lower than the utility’s avoided cost.
Of relevance to the instant appeal are PURPA’s guidelines for the certification of facilities as “qualifying cogeneration facilities,” and FERC’s rules prescribing the standards for that certification. The statute defines “cogeneration facility” as one that produces “(i) electric energy, and (ii) steam or forms of useful energy (such as heat) which are used for industrial, commercial, heating, or cooling purposes.” 16 U.S.C. § 796(18)(A). To determine which nontraditional power producers could receive benefits, PURPA created a category of “qualifying cogeneration facilities,” or QFs, which includes any facility FERC determines has met the regulatory requirements. See 16 U.S.C. § 796(18)(B)(i).
FERC’s regulations prescribe operating, efficiency, and ownership standards for facilities seeking QF status. See 18 C.F.R. § 292.205 (operating and efficiency standards); 18 C.F.R. § 292.206 (ownership criteria). Relevant here is the requirement that electric utilities hold less than 50% of the equity interest in the cogeneration facility. See 18 C.F.R. § 292.206. In addition, the cogeneration facility must “produce electric energy and forms of useful thermal energy (such as heat or steam), used for industrial, commercial, heating, or cooling purposes, through the sequential use of energy.” 18 C.F.R. § 292.202(c) (emphasis added).
FERC has explained that “the ultimate determination of usefulness will be made *238in the marketplace.” See Electrodyne Research Cow, 32 FERC ¶ 61,102, ¶ 61,278 (1985). It therefore applies one of three economic tests in determining whether a thermal output is useful for purposes of QF certification. First, if a cogenerator proposes to use its thermal energy in a common industrial or commercial process, that energy is considered presumptively useful. See id. at ¶ 61,279. A process, or thermal application, will be deemed “common” after the Commission has received a satisfactory number of QF applications proposing the same use for the thermal output. See Kamine/Besicorp Allegany, L.P., 63 FERC ¶ 61,320, ¶ 63,158 (1993). The Commission reasons that, if a thermal application is a common one, the technology involved must be established and there must be a market for the application’s end-product. See Arroyo Energy, L.P. (Arroyo II), 63 FERC ¶ 61,198, ¶ 62,545 (1993); Polk Power Partners, L.P., et al, 61 FERC ¶ 61,300, ¶ 62,128 (1992). As such, when a facility’s proposed use of thermal energy is common in the industry, FERC presumes the energy used in that application is useful and performs no further analysis regarding the economics of the thermal application. See Bayside Cogeneration, L.P. (Bayside II), 67 FERC ¶ 61,290, ¶ 62,006 (1994).
When the facility proposes an uncommon application, i.e., one that involves a new technology or creates an end-product without an established market, FERC’s analysis is different. See Elec-trodyne, 32 FERC at ¶ 61,278. It employs separate analyses depending on whether the purchaser of the thermal energy — the “thermal host” — is an entity unaffiliated or affiliated with the cogenerator. If an independent entity, unaffiliated with the cogenerator, purchases the thermal energy, FERC considers the energy useful because it assumes no entity would purchase the thermal output, or the end-product produced with the aid of the thermal output, unless it served some legitimate purpose. See Liquid Carbonic Industries Corp. v. FERC, 29 F.3d 697, 700 (D.C.Cir.1994). In other words, purchase by the thermal host establishes that there is an arm’s-length market for the output. See Kamine, 63 FERC at ¶ 63,158; Electrodyne, 32 FERC at ¶ 61,279. FERC, therefore, deems the thermal energy useful and performs no further analysis regarding the economics of the thermal application. See LaJet Energy Co., 44 FERC ¶ 61,288, ¶ 61,194 (1988); Electrodyne, 32 FERC at ¶ 61,279.
If the use of thermal energy is uncommon and the thermal host is the cogenerator itself, or its affiliate, only then will FERC inquire into the economic viability of the thermal use. See Electrodyne, 32 FERC at ¶ 61,279. Specifically, the cogen-erator is required to provide evidence that “the output would be economically justified in an independent business setting.” Id. at ¶ 61,278. That is, the cogenerator must show that the thermal use is itself profitable without subsidy from the sale of electricity. FERC imposed this requirement on affiliated thermal hosts to prevent co-generators whose thermal outputs have no established market from pawning off their thermal energy for an impractical purpose, while retaining their QF status and concomitant right to sell power at avoided cost rates. See id.
FERC’s certification process occurs pri- or to the construction of the facility, and QF status is granted or denied based on the representations in a facility’s application. The regulations provide, however, that FERC may revoke the QF status of a previously-certified facility if the facility, when operational, fails to comply with any of the statements in its application. See 18 C.F.R. § 292.207(d)(1).
Brazos challenges Tenaska’s certification as a QF. Tenaska and Brazos entered their Power Purchase Agreement (“PPA”) in 1993, which obligated Brazos to purchase electric power from Tenaska for twenty-three years, with a seventeen-year rollover, at prices fixed in the PPA. This was *239not a situation where Brazos was compelled under PURPA to purchase electricity from a QF. Rather, both parties were equally interested in Tenaska’s becoming certified as a QF. Tenaska wanted to qualify for PURPA benefits. Brazos wanted the best rates. According to Philip Seg-rest, Brazos’ attorney at the time, the only power sources in Texas were public utilities and QFs. As a QF, Tenaska would only be able to charge rates up to Brazos’ avoided costs. The public utilities, however, were currently charging rates above Brazos’ avoided costs. Thus, because Te-naska’s rates were favorable and because the option of building its own plant was impractical, Brazos “insistfed] that the PPA be certified by the [Public Utility Commission of Texas] pursuant to PURA [the Public Utility Regulatory Act].... ” Affidavit of Philip Segrest. Under PURA, certification of power purchase agreements was permitted only if the power was being purchased from a QF, as that term was defined in PURPA. See Tex.Rev.Civ. Stat. Ann. art. 1446c (West Supp.1994) (repealed 1995).
Therefore, on October 20, 1994, Tenaska applied to FERC for QF certification, obviously with Brazos’ blessing. According to its application, Tenaska intended to sell its electrical output to Brazos, while its thermal output, steam, was to be converted into distilled water for sale to “a third party.” FERC published notice of Tenas-ka’s application in the Federal Register but received no protests or requests for interventions to Tenaska’s certification. Therefore, on January 13, 1995, the Commission granted Tenaska QF status. In doing so, FERC determined, in relevant part, that Tenaska fulfilled its ownership requirement that utilities own less than 50% equitable interest in the facility. More importantly, FERC also concluded that the conversion of steam to distilled water was a common industrial process and application of thermal energy for that use was, therefore, presumptively useful.
Tenaska entered into an arrangement with the City of Cleburne (the “City”) in which Tenaska would (1) purchase the City’s potable water for use in its steam generator, (2) recover the reject water stream from the steam generator’s boiler makeup water treatment system and use it to supply the distilled water system, (3) sell the distilled water to the City, and (4) purchase effluent water from the City’s wastewater treatment facility for use in its cooling tower. Under this arrangement, the City gave Tenaska a ten-dollar credit on its water bill for its production of the distilled water, and the City was obligated to construct the facilities necessary for transporting both the effluent water to Tenaska and the distilled water from Te-naska. The City was responsible for the initial financing of the construction, including the issuance of tax-exempt municipal bonds, for which Tenaska would reimburse the City in monthly payments when the debt service was owed. The facility became operational in January 1997.
The City had originally agreed to purchase Tenaska’s distilled water in order to attract industries to an industrial park near Tenaska’s facility, by offering the ready supply of distilled water for sale as process water. Water which was not resold was to be used to augment the flow of Buffalo Creek, a stream running near the City’s business district whose stagnant waters were encouraging nuisance conditions and an increased mosquito population. While negotiations continued with potential occupants of the industrial park, the City ran into difficulty garnering permits from the Environmental Protection Agency to increase Buffalo Creek’s flow with distilled water. Initially, therefore, the City had no specific use for the distilled water it was purchasing from Tenaska, and it released its purchase into the City’s sewer system. An occupant of the industrial park began purchasing the distilled water in September 1997.
On August 22, 1997, Brazos filed with FERC a motion and petition for revocation *240of Tenaska’s QF status. According to Brazos, the use of Tenaska’s thermal output for the production of distilled water had not proven to be “useful.” The presumption of usefulness on which Tenaska’s QF status was certified, Brazos argued, was rebutted by actual operation of the facility — notably, by the fact that the City paid only ten dollars a month for thousands of gallons of water and then dumped the water in the sewer. Meanwhile, Brazos was being forced under the PPA to pay fixed rates which, five years into the deal, were no longer below the market price. In addition, Brazos contended that Tenas-ka did not satisfy the ownership requirements for QF status. Although utilities owned less than 50% of Tenaska, Brazos alleged that the utilities’ 45% interest gave them effective control in a voting procedure requiring a 70% vote to take action.
FERC denied Brazos’ motion. FERC stated that once it determines the proposed use of thermal energy is common, it presumes the thermal energy is useful. FERC would not inquire thereafter into how the thermal host used its purchase, nor would it question whether the cogener-ation facility was actually making money from its sale. FERC also found that Te-naska satisfied the ownership requirements for QF status, noting that the utilities’ 45% interest was insufficient to effect day-to-day action without the votes of 25% of the non-utility owners. Additionally, the Commission noted that Tenaska’s ownership structure had not changed since its certification, and because Brazos failed to object then, its complaint now was untimely-
Subsequently, Brazos filed a request for rehearing and its request was denied. Brazos now petitions this Court for review of FERC’s order denying the revocation of Tenaska’s QF status.
II.
We must affirm FERC’s order unless it is “arbitrary, capricious, an abuse of discretion, or otherwise not in accordance with law.” 5 U.S.C. § 706(2)(A). The scope of review under this standard is narrow; it does not authorize a reviewing court to substitute its judgment for that of the agency. See Motor Vehicle Mfrs. Ass’n v. State Farm Mutual Ins. Co., 463 U.S. 29, 43, 103 S.Ct. 2856, 77 L.Ed.2d 443 (1983). Rather, we must examine “ ‘whether the decision was based on a consideration of the relevant factors and whether there has been a clear error of judgment.’ ” Id. (quoting Bowman Transp., Inc. v. Arkansas-Best Freight Sys., Inc., 419 U.S. 281, 285, 95 S.Ct. 438, 42 L.Ed.2d 447 (1974)). Where an agency has considered the relevant factors and provided a satisfactory explanation for its actions, its decision will be upheld.
III.
Noting that FERC regulations allow for post-operational challenges to a QF’s certification, Brazos maintains that Tenaska is not entitled to benefits under PURPA because, after its facility was certified and became operational, Tenaska faded to uphold the regulatory requirements for QF status in accordance with projections contained in its application for QF certification. Brazos advances the same arguments it did below: the Commission should revoke Tenaska’s QF status because the production of distilled water has not proven to be useful, and because Te-naska does not meet PURPA’s ownership requirements. We address each contention in turn.
A. Useful Thermal Output
The thrust of Brazos’ argument is that the Commission’s precedent has not established an irrebuttable presumption of usefulness. Brazos does not take issue with the use of the presumption during certification, before the facility is even built; then, Brazos reasons, the Commission is justified in relying on hypothetical facts and an applicant’s claim that, because the proposed thermal use is common, the *241thermal energy will be useful. Rather, Brazos asserts that the Commission was obligated to consider post-operational facts that could rebut the presumption of a thermal output’s usefulness. Specifically, Brazos avers first that, although Tenaska represented in its application for QF status that distilled water would be produced for sale to a third party, the sale of water to the City is a “sham” sale designed only to retain PURPA benefits; it is not a sale serving an independent business purpose that could be economically justified. Second, Brazos contends that the water was not useful because before the City found a purchaser, it was pouring the water in the sewer.
In response, FERC asserts that Tenas-ka continues to satisfy PURPA’s regulatory requirements, as represented in its application for QF status. First, Tenaska’s sale of water was one piece of a legitimate, integrated financing package. Second, Te-naska’s thermal energy has been useful since the day it was certified, for Tenaska has used “an established technology to produce a common product with an existing market.” Arroyo, 63 FERC at ¶ 62,545 n. 4. According to the Commission, there is no statutory or precedential requirement that, upon the facility’s operation, it examine how each individual thermal host is using its purchases or how economically sound every transaction turned out to be. In fact, FERC argues, doing so would undermine its directives under PURPA, for conditioning the maintenance of QF status on investigations into the economics of a thermal host’s purchase would impede the development of cogener-ation. We agree that Brazos has misconstrued the Commission’s prior holdings. Further, to the extent that the present factual scenario differs from that of the Commission’s precedents, FERC’s use of its presumption here furthers its congressional mandate under PURPA consistently with its regulations promulgated thereto.
The Commission has consistently refused to inquire into the economics of common thermal applications to rebut the presumption of a thermal output’s usefulness. See Polk Power Partners, L.P., et al., 61 FERC ¶ 61,300, ¶ 62,128 (1992) (“We think it better, once a thermal process has been found to be common, to refrain from second-guessing the decision to use cogener-ated thermal energy in a particular industrial process and in a particular manner.”); Arroyo II, 63 FERC at ¶ 62,546 (stating that, when a cogenerator proposes a common application for its thermal energy, “the Commission does not perform an economic analysis” because “[a] contrary approach would act to discourage the development of the cogeneration industry”). So too has the Commission refused to examine the economics of a particular thermal host’s ultimate use of its purchase. See Brooklyn Navy Yard Cogeneration Partners, L.P., 74 FERC at ¶ 61,015, ¶ 61,046 (1996) (stating that, if the thermal application is common, FERC “will not inquire further into how the [end] product is being used [by the thermal host]”); Arroyo Energy, L.P. (Arroyo I), 62 FERC ¶ 61,257, ¶ 62,722-23, reh’g denied 63 FERC ¶ 61,-198 (1993) (concluding that it is the common nature of the industrial or commercial process, not the use to which the product is put, that determines that the thermal energy is useful).
In Bayside II, for example, a utility challenged the cogenerator’s proposed application of its thermal output to create distilled water, arguing that the proposed thermal host’s intended use of the water was not economically viable and that the cogenerator’s thermal output was therefore required to serve “an independent business purpose with some economic justification.” Bayside II, 67 FERC at ¶ 62,006 (internal citation omitted). Noting that the distillation of water was a common application, and thus the thermal output presumptively useful, the Commission rejected this argument:
[The utility] would have the Commission view this “presumption” as an evidentia-ry presumption that is rebuttable upon *242the submission of economic evidence by a party opposing certification. But this has never been the Commission’s intention or practice. Rather, the Commission, upon a finding that the usefulness of a thermal application has been established by common practice, is making a finding that practice has established that a particular use of cogenerated thermal energy is economic. It thus “presumes” that the thermal output is useful; there is no need to engage in a further inquiry into the usefulness of the particular output. In other words, when the Commission has found a use to be common, there is no need to determine whether a particular use of a particular applicant’s output would be economically justified.
Id.
Given the consistency with which FERC has denied further inquiry into common thermal applications, nothing in these precedents persuades us that FERC’s presumption of usefulness is rebuttable, even post-operation, by the two sets of facts Brazos presents. Brazos asserts first that the Commission should have examined the economics of Tenaska’s Distilled Water Supply Agreement with the City and applied the independent business purpose test to that transaction. For this proposition, Brazos relies on LaJet Energy Co., 43 FERC ¶ 61,288 (1988), which was decided before the Commission deemed the distillation process common. In LaJet, the Commission stated that “the thermal application, here the distilled water process, must be economically viable on its own to be considered ‘useful,’ economic viability meaning that anticipated revenues of the distilled water process should be higher than anticipated expenses.” 43 FERC at ¶ 61,790. As noted above, however, the Commission has steadfastly refused to examine the economic viability of transactions in common applications, noting that it “looks at the economic viability of the use of thermal output to assess whether the energy is ‘useful’ only in very limited circumstances,” Bayside II, 67 FERC at ¶ 62,006: only when the cogenerator’s thermal host is an affiliate (or the cogener-ator itself), and then only when the cogen-erator has proposed an uncommon application. See LaJet, 43 FERC at ¶ 61,790; Electrodyne, 32 FERC at ¶ 61,278. Such was the case in LaJet, where FERC examined an affiliate’s use of what was, in 1988 anyway, a novel thermal application. It was for that reason that the Commission applied the independent business purpose test for affiliate use of novel applications, rather than the presumption of usefulness for non-affiliate use of common applications.
Brazos is correct, however, that the precedents FERC relies upon in this case focus on FERC’s initial certification of co-generators, not the revocation of QF status after the facility is operational. Kamine/Besicorp Allegany, L.P., 63 FERC ¶ 61,320 (1993), is instructive to Brazos’ complaint. There, the cogenerator applying for QF status entered a contract to sell its thermal energy, steam, to a non-affiliate distillation plant. The distillation of water was considered an uncommon application at the time, but because the cogenerator had already entered an arm’s-length contract with its thermal host, it satisfied the test for non-affiliate use of uncommon applications. Relying on La-Jet, the public utility requested that the Commission review the cogenerator’s contract to determine if the sale was in fact “useful,” because the cogenerator had not included the contract as part of its QF application. The Commission refused to review the contract, but not because the cogenerator had provided specific evidence of an arm’s-length market. Instead, the Commission-held that,
[GJiven that we now have seen at least four applications, other than Kamine’s, proposing to distill water with the assistance of the thermal output of a cogen-eration facility ..., we believe that the distillation of water is a common industrial process, and that ultra-pure or distilled water is a common product; thus, *243the distillation of water is common for purposes of determining usefulness....
Because the thermal energy output of the Kamine facility is presumptively useful, the Commission has no need to review the contracts between Kamine and the non-affiliated purchaser of the facility’s thermal output.
Id. at ¶ 63,158 (footnote omitted). Similarly, the City in the instant case is a non-affiliated purchaser of a product produced by a common application. Further, as in Kamine, a questionable thermal energy contract was available for the Commission to review to establish whether the distillation of water met LaJet’s, stringent independent business purpose test. As Brazos points out, Tenaska’s’ contract with the City was in place when it applied for certification as a QF. Because Tenaska was proposing a common application, however, it was not required to submit economic evidence. See Bayside Cogeneration, L.P. (Bayside I), 66 FERC ¶ 61,259, reh’g denied, 67 FERC ¶ 61,290, ¶ 61,631 (1994) (stating that “applications involving water distillation ... need not be accompanied by evidence of arm’s-length contracts ... or economic viability”). Further, if Brazos had requested such evidence, as it does now, the Commission would have declined to review it. PURPA and its implementing regulations require only that the thermal energy be useful; they do not demand that the sale of every end-product be profitable. See Bayside II, 67 FERC at ¶ 62,006 (“There is no statutory requirement that the Commission find that the thermal output is being used in an economic manner.”). Said differently, the issue is not whether the eogenerator makes money from its common application, but that, because there is a market for the application, it is capable of doing so. See id.
The Commission has previously opined that treating its presumption as rebuttable would be inconsistent with PURPA’s goals, in that “[providing an opportunity for evi-dentiary hearings before the Commission ... would seriously impede the very development of cogeneration ... that Congress sought to facilitate.” Id. at ¶ 62,006 n. 5 (quoting American Paper Institute, Inc. v. American Electric Power Service Corp., 461 U.S. 402, 420, 103 S.Ct. 1921, 76 L.Ed.2d 22 (1983)) (internal quotations omitted). Nonetheless, even if we were to look behind the presumption here, we, like the Commission, are not persuaded that the sale to the City is a sham. Tenaska correctly points out that it received much more than ten dollars in its transactions with the City, for the Distilled Water Supply Agreement was only one piece of their arrangement’s puzzle: Tenaska purchased potable water from the City for use in its steam generator, the reject stream from the generator was turned into distilled water, the City purchased the distilled water to attract industrial customers to an adjacent industrial park, the plant’s blowdown water was transported to the City’s sewage treatment center, and Tenaska purchased the treated sewage effluent from the City for use in the cogenerator’s cooling tower. In addition, Tenaska received tax abatements, as well as access to the City’s debt for construction of its water facilities so all of this could take place. Seen in the context of a complex project financing, Tenaska’s arrangement with the City garnered it more than ten dollars a month. Thus, as proposed in its QF application, Tenaska has sold its product to a third party.
In addition to the economics of Tenas-ka’s transaction with the City, Brazos urges the Commission to examine the City’s use of the distilled water during its first year of operation because, according to Brazos, the City’s actual use would rebut the Commission’s initial presumption of the thermal energy’s usefulness. Relying heavily on Arroyo II, 63 FERC ¶ 61,198 (1993), Brazos contends that, before determining that a thermal output is useful, the Commission must be satisfied that the thermal host’s use of its purchase is a bona fide industrial or commercial use. When Tenaska first became operational, it *244was pouring its distilled water into a sewer, which Brazos maintains was not a bona fide use and renders the thermal energy used to create the water non-useful. Brazos contends that the Commission’s failure to take into account the City’s actual use of the water was an unexplained departure from its precedents.
In Arroyo II, the cogenerator’s thermal output, steam, was to be used in absorption refrigeration (AR) equipment to provide ice to an adjacent ice rink. The utility complained that the use of thermal output to help create and maintain an ice rink was a novel use requiring application of the independent business purpose test. The Commission declined to apply the test because it found that AR technology was a common use for steam, and the steam was therefore useful. Brazos, however, relies on several passages in the opinion as evidence that the Commission has tempered its presumption of usefulness by also examining the proposed end-use to which the thermal product would be put:
Our review of the evidence compiled in this proceeding confirms that the proposed use of fhe Arroyo [cogeneration] facility’s thermal output for refrigeration purposes is indeed bona fide. SDG&E [the utility] presents us with no new reason to upset our earlier determination that the technology to be applied by Arroyo, as well as the end product, are established and, accordingly, that the thermal output of the facility is presumptively useful.
Arroyo II, 63 FERC at ¶ 62,545 (citation omitted) (Brazos’ emphasis added). This passage does not support Brazos’ contention that FERC examines the thermal host’s end-use of its purchase in determining the usefulness of a cogenerator’s thermal output. It is the cogenerator’s use of thermal energy that must be bona fide, not the thermal host’s end-use of the end-product, and the cogenerator’s use is bona fide when it is common in the industry. Thus, the passage states that the thermal energy, was useful because the thermal output (steam) was put to a bona fide use in a common application (AR technology) and created a common product (ice). This passage does not say that the cogenerator’s thermal energy was useful because the thermal host’s use of ice to build an ice rink was bona fide. The passage does not refer at all to the thermal host’s end-use of the thermal product. Nor should it. As the Commission has noted, “if a cogenerator produces a product that has already met the Commission’s usefulness requirement, there is no further inquiry to determine if the product is being used by the recipient for a common purpose.” Brooklyn Navy Yard, 74 FERC at ¶ 61,046.
In this way, Brazos’ complaint that the distilled water was not “useful” misses the point. The distillation of water is common, so the steam used to create it is useful. The use an unaffiliated thermal host makes of its arm’s-length purchase is irrelevant. See Arroyo I, 62 FERC at ¶ 62,723 (“The fact that this is the first instance before the Commission in which this common refrigeration technology is associated with a common refrigeration product for end-use in an ice rink is irrelevant.”). This is because the Commission, as the arbiter of “usefulness,” has defined the concept in terms of economics. If an application is common, the technology is established and there is a market for the product. If the technology is established and there is a market for the product, that which is .used in the application to create the product is “useful.” Once the energy used in the established technology or the product with the established market is purchased by a thermal host, FERC has no further involvement. The purchaser bears the market risk of its purchase, not the seller, and whatever use or profit the purchaser makes of its purchase, whether by pouring it in a sewer or reselling it, is of no moment to the seller. See Bayside II, 67 FERC at ¶ 62,006 n. 7 (stating that “PURPA does not require that the Commission ensure that a thermal host make *245as much money as possible, or make any money at all; all PURPA requires is that the Commission ensure that the thermal host takes useful thermal energy that is used for industrial, commercial, heating, or cooling purposes.” (internal quotations omitted) (emphasis added)). The point is that the seller has successfully sold its output in an arm’s-length market and the purchaser has access to the same market for resale. There is, of course, proof of this point in the instant case — two weeks after Brazos filed its motion and petition for revocation, the City found a purchaser for its distilled water from among those industries it was trying to attract with the supply of that water.
Brazos also points us to FERC’s response to the utility’s suggestion in Arroyo II that the Commission’s presumption of usefulness would allow certification of a QF proposing to throw away the product of a common application “in an underhanded effort” to meet the regulatory requirements. The Commission stated:
[T]he Commission does not apply the presumption of usefulness so cavalierly and, as explained above, must be assured that the thermal energy output is being used in a bona fide manner for a legitimate industrial, commercial, heating, or cooling purpose.
Arroyo II, 63 FERC at ¶ 62,545 n. 4. Again, the Commission’s response to the utility’s hypothetical provides no support for Brazos’ assertion. First, a cogenerator seeking QF status by proposing that it throw away its own common end-product is markedly different from an unaffiliated, third-party thermal host having to waste its arm’s-length purchase from a QF because of protracted negotiations with buyers and permit problems. That is, a cogenerator’s use of thermal energy is obviously not “bona fide” if its intention from the outset is to dispose of it itself. However, a cogenerator’s use of thermal energy is not “mala fide” if a third-party purchaser of the output, through no fault of its own, cannot resell its purchase.
Second, Brazos neglected to include the quoted footnote’s final sentence: “The flaw in SDG&E’s argument is that Arroyo proposes to apply the thermal energy output of its facility in a useful manner using an established technology to produce a common product with an existing market.” Id. Similarly, Brazos again fails to recognize that Tenaska was using its thermal output (steam) in an established technology (distillation) to produce a common product (distilled water) with an existing market (the City).
Although this case presents us with what, at first blush, appears to be the proverbial “peppercorn” scenario — a party paying ten dollars for thousands of gallons of distilled water that for nine months it poured into the sewer — we are, for a number of reasons, hesitant to look beyond FERC’s presumption of usefulness to release Brazos from its contractual obligations. First, Tenaska and the City entered an arm’s-length contract, one amongst many contracts in which the risks and benefits of the typical project finance arrangement were traded. That Brazos now finds itself paying above-market prices for electricity because it entered a “front-loaded” contract fails to undermine the utility of the Commission’s presumption of the usefulness of thermal energy in common applications. A front-loaded contract means that the utility’s, payment rates are determined at the time the obligation to buy from the cogenerator is incurred, rather than at the time of delivery. Such contracts are often used because they allow the QF to finance the construction and operation of the facility in the early years of the contract. As the Ninth Circuit observed in Independent Energy Producers Ass’n, Inc. v. California Pub. Util. Comm’n, 36 F.3d 848, 858 (9th Cir.1994), such contracts have been upheld notwithstanding the recognized risk that the prices set by the contract might at times exceed the utility’s actual avoided costs, because “certainty as to rate was important.” By ensuring a predictable flow of *246income, such contracts encourage the development of alternative facilities that might never be built. In FERC’s words:
The Commission recognizes this possibility [that current avoided costs might be lower than the rates provided in the contracts] but is cognizant that in other cases, the required rate will turn out to be lower than the avoided cost at the time of purchase.... Many commentators have stressed the need for certainty with regard to return on investment in new technologies. The Commission agrees with these ... arguments, and believes that, in the long run, “over estimations” and “under estimations” of avoided costs will balance out.
Small Power Production and Cogeneration Facilities; Regulations Implementing Section 210 of PURPA, 45 Fed.Reg. 12224 (1980) (quoted in Independent Energy Producers Ass’n, 36 F.3d at 858).
Second, allowing post-operation rebuttal of FERC’s presumption of a thermal output’s usefulness on these grounds would impede the development of cogeneration facilities, a development PURPA was enacted to encourage. See Arroyo II, 63 FERC at ¶ 62,546 (explaining that performing economic analyses on common applications would discourage cogeneration). By sanctioning such rebuttal, we would ensure that every time the market for electricity fluctuated and the utility that was the QF’s only market recipient was suddenly displeased with its rates, the utility could back out because it thought either the QF or the thermal host was not being as economically wise or efficient as it should be. This scenario poses several problems.
Owners of QFs would have little incentive to sell electric energy if they had to go through an evidentiary hearing before FERC in Washington, D.C., every time a utility claimed someone else was behaving inefficiently with a common application. See Polk Power Partners, 61 FERC at ¶ 62,128 (refusing to review evidence of economic inefficiency because to do so would allow third parties to “compel a hearing simply by the submittal of evidence purporting to show that a thermal process is not the most economic, no matter how common the process”). Presumptions are over-inclusive by definition. FERC’s decision to apply one strictly in this case neither contravenes PURPA’s mandates nor supercedes the discretion afforded agencies in interpreting their own regulations. See Chevron U.S.A., Inc. v. Natural Resources Defense Council, Inc., 467 U.S. 837, 844, 104 S.Ct. 2778, 81 L.Ed.2d 694 (1984). PURPA, and FERC’s regulations promulgated thereto, require only that a cogenerator produce useful thermal energy for legitimate processes. See 16 U.S.C. § 796(18)(A); 18 C.F.R. § 292.202. Neither the statute nor the regulations insist or presuppose that FERC should engage in such heavy-handed oversight as to keep tabs on QFs’ arm’s-length purchasers. If we were to hold otherwise, we would embrace the very form of micromanagement that the Commission has determined QFs are supposed to be freed from, and we would “impute to Congress a purpose to paralyze with one hand what it sought to promote with the other.” Clark v. Uebersee Finanz-Korporation, A.G., 332 U.S. 480, 489, 68 S.Ct. 174, 92 L.Ed. 88 (1947).
More importantly, FERC’s presumption of usefulness is meant to enable cogeneration facilities to obtain financing. The presumption provides certainty for investors that their investment is duly certified under PURPA and entitled to the benefits of PURPA’s statutory imperatives, including the presence of a utility to purchase the facility’s output. Because the value of a facility’s hard assets is usually less than the project debt, debt repayment and anticipated equity returns depend on performance under project contracts. The contracts constitute the framework for project viability because the ability of the project sponsor to produce revenue from project operation is the foundation of a project financing. The PPA is the principal *247source of project revenue. Therefore, banks lend money for construction and permanent financing on the strength of the utility’s obligation to purchase power from a QF. Revocation of a facility’s QF status releases the utility from its obligation under the PPA. It leaves the facility without a market recipient and thus without a revenue source for debt repayment. We would be hai’d pressed to imagine the investor who would contribute to a project so susceptible to such a scenario. In sum, both FERC’s precedents and PURPA’s mandates persuade us that Tenaska continues to produce useful thermal energy in accordance with the representations in its application for QF status.
B. Ownership Criteria
In order to obtain QF status under PURPA, a cogeneration facility must not be owned by persons primarily engaged in the generation or sale of electric power. See 16 U.S.C. § 796(18)(B). The Commission clarified this requirement, determining that electric utilities may own no more than 50% of the equity interest in a QF. See 18 C.F.R. § 292.206(b). FERC’s regulations thus equate “ownership interest” with “equity interest,” but they do not define the term “equity interest.” See Ultrapower 3, 27 FERC ¶ 61,094, ¶ 61,183 (1984). Cases discussing the Commission’s ownership criteria emphasize the stream of benefits accruing to each partner, but the voting interests of each partner have also been examined to avoid a utility partner’s manipulation of those benefits. See id. at ¶ 61,184. Accordingly, “a utility partner may not have more than 50% control of a qualifying facility.” Brooklyn Navy Yard, 74 FERC at ¶ 61,048.
According to Brazos, Tenaska has not satisfied the ownership criteria for QF status because utilities have effective control over the facility’s operation. Affiliates of three utilities own a 45% interest in Tenas-ka and have a 38.9% voting interest in the facility, in conformity with the regulatory limits. Before FERC, however, Brazos contended that the utilities’ 45% interest in the facility gives them effective control over the facility’s operation, because a 70% vote of Tenaska’s Executive Review Committee is needed to take significant actions.1 The Commission disagreed, pointing out that the utility affiliates would still need the approval of the non-utility owners to take significant action. The Commission also pointed out that Tenaska’s ownership structure had not changed since it was certified as a QF. Because Brazos was listed as the utility-purchaser in the QF application and failed to object when the application was noticed for comment, the Commission determined that Brazos’ challenge was untimely.
In response, Brazos argues that the utility affiliates’ 45% interest is enough to block significant action, thereby giving them effective control. Further, Brazos contends that it should not be faulted for failing to object because it was relying on Tenaska’s contractual obligation to meet QF standards.
Brazos’ arguments are unpersuasive. All certification orders granting QF status state the following: “To the extent that facts or representations which form the basis for this order change, this order cannot be relied upon.” Tenaska IV Texas Partners, Ltd., 70 FERC ¶ 62,026, ¶ 64,081 (1995). Tenaska’s ownership structure has not changed since it applied for and was granted QF status. It follows that the order may still be relied upon. Brazos’ contention that it was relying on Tenaska to meet QF standards does not explain why it waited two and a half years after Tenaska’s certification to object. We agree with the Commission that “Allowing such belated challenges to QF certifications despite unchanged facts would undermine the contractual reliance QFs need in order to finance and build their pro*248jects.” Brazos Electric Power Cooperative v. Tenaska IV Texas Partners, Ltd., 85 FERC ¶ 61,097, ¶ 61,348 (1998).
Briefly, even if we were to give Brazos the benefit of the doubt, its challenge is still without merit. As noted, FERC’s ownership criteria are intended to prevent utilities from diverting to themselves the stream of benefits flowing from a QF, such that the utilities would gain some undue advantage vis-a-vis non-utility partners. See Dominion Resources, Inc., 43 FERC ¶ 61,079, ¶ 61,251 (1988). In order to accrue such benefits, “control” requires action, not inaction. That is, a minority interest’s ability to block significant actions does not garner the benefits the controlling interest can manipulate by taking significant actions. Furthermore, if the ability to block significant action constituted “control,” then Brazos is actually contending that utilities may not have more than 30% control — a proposition that finds no support in the Commission’s precedents. See id. at ¶ 61,251 (stating that “a facility will meet the ownership requirements of PURPA ... so long as the interest in the stream of benefits and control by a utility or utilities, by whatever mechanism used, does not exceed 50%”). The utility affiliates’ equity and voting interests in Tenaska satisfy the Commission’s ownership requirements for QF status.
IV.
For the foregoing reasons, we DENY Brazos’ petition for review in No. 98-60684; No. 98-60568 is DISMISSED.
. It is unclear why Brazos' "effective control” argument focuses on the utility-affiliates' ownership interest rather than their voting interest.