A cogeneration plant produces not only electric power but also steam or other thermal energy that can be used for various industrial or commercial purposes. 16 U.S.C. § 796. To encourage this source of energy, the Public Utilities Regulatory Policies Act of 1978, 16 U.S.C. § 824a-3 ("PURPA"), imposed a requirement on electric power utilities to purchase power from any qualifying facility ("QF")--a cogeneration plant that meets PURPA's QF standards. See 18 CFR §§ 292.101(b)(1), 292.203(b). The utility must pay for the power at a rate no greater than its "avoided cost"--the cost it would incur to generate an equivalent amount of power itself. See 16 U.S.C.s 824a-3(b); 18 CFR §§ 292.304(a), 292.101(b)(6). Although the phrase "avoided cost" has the ring of an economically sound price, it was suggested without contradiction at oral argument that it has not so proved.

Niagara, a producer and seller of electricity in upstate New York and subject to the PURPA mandate, suspected that some of the facilities from which it buys may not actually qualify for QF status, and to pursue the matter filed a request with the Department of Energy ("DOE") in 1995 under the Freedom of Information Act ("FOIA"), seeking information collected by DOE on forms that these facilities are required to file with DOE's Energy Information Administration--Forms EIA-867. DOE disclosed some but withheld other information, invoking FOIA's Exemption 4, 5 U.S.C. § 552(b)(4), which covers "trade secrets and commercial or financial information obtained from a person and privileged or confidential." The information withheld relates particularly to quantities of fuel consumed and power generated, from which, given market prices for fuel, outsiders could go far to calculating a facility's unit cost of power.

Niagara sued in district court to compel release of the withheld data. Independent Power Producers of New York, Inc., a trade group, and Sithe Energy Inc. ("QF Intervenors" collectively) intervened in support of DOE. Both DOE and the QF Intervenors moved for summary judgment, supporting the motion with affidavits describing the QF industry. Niagara moved for discovery, and on the district court's denial of its motion filed an opposition to the motion for summary judgment, attaching an affidavit depicting the industry in rather different terms. The district court granted summary judgment. It held that Niagara had failed to raise an issue of material fact against DOE's position that the information was exempt because its release (1) would cause substantial competitive harm to the entities submitting the information (the QFs), and (2) would impair the agency's ability to collect this information in the future. The court also rejected Niagara's claim that no FOIA exemption could apply because the information was already publicly available.

* * *

The language of Exemption 4 protects from disclosure "commercial information" obtained from a non-government source, so long as it is "privileged or confidential." The only dispute here is over the last phrase. In National Parks & Conservation Ass'n v. Morton, 498 F.2d 765, 770 (D.C.Cir.1974) ("National Parks I"), this court adopted a narrow reading of the word "confidential," saying that information was confidential within the meaning of Exemption 4 only if its disclosure was likely to (1) impair the government's ability to obtain necessary information in the future, or (2) cause substantial harm to the competitive position of the person from whom the information was obtained. The district court found here that DOE had satisfied both alternatives.

In support of its claim that release would impair government interests, DOE offered two conclusory affidavits, claiming that disclosure would impair the EIA's ability to collect such information in the future. See Walton Decl. 57 39-43; Grutsch Decl., p 10. The claim is inherently weak where, as here, the agency has secured the information under compulsion. Critical Mass Energy Project v. NRC, 975 F.2d 871, 878 (D.C.Cir.1992) (en banc). Yet DOE and the QF Intervenors offer nothing but Walton's speculative opinion that QFs may not be forthcoming in the data they submit if DOE allows disclosure, see Walton Decl., p 41, and Grutsch's terse and self-serving statement that as an executive of various QFs he would "attempt to minimize the scope and specificity of the information provided." See Grutsch Decl., p 10. But the agency has the burden of showing that requested information comes within a FOIA exemption, National Parks & Conservation Ass'n v. Kleppe, 547 F.2d 673, 679 (D.C.Cir.1976) ("National Parks II"), and we have more than once held that such conclusory and generalized assertions are not enough to establish the requisite risk of impairment. Id. at 680; Washington Post Co. v. U.S. Dep't of Health and Human Serv., 690 F.2d 252, 269 (D.C.Cir.1982).

DOE insists that summary judgment is proper because Niagara did not controvert the assertions of impairment. But on a summary judgment motion, "[f]acts not conclusively demonstrated, but essential to the movant's claim, are not established merely by his opponent's silence; rather, the movant must shoulder the burden of showing affirmatively the absence of any meaningful factual issue." See National Assoc. Of Gov't Employees v. Campbell, 593 F.2d 1023, 1027 (D.C.Cir.1978). A paper asserting the affiant's intention to sail as close to the wind as possible is hardly enough for this case--especially as the data sought appears to take the form of hard, cold numbers on energy use and production, the fudging of which may strain all but the deliberately mendacious. As the DOE and QF Intervenor affidavits are in these circumstances too vague, the grant of summary judgment on this issue was unjustified.

DOE fares no better in its effort to show that there is no genuine issue of material fact on the likelihood of substantial competitive harm. In National Parks II, we held that for the government to preclude disclosure based on a competitive injury claim, it must prove that the submitters "(1) actually face competition, and (2) substantial competitive injury would likely result from disclosure." 547 F.2d at 679. Here, DOE's assertions of the existence of competition are somewhat conclusory. See Walton Decl. 57 22-31. But assuming arguendo that DOE met its initial burden of proving that QFs were engaged in competition, Niagara's response was adequate to raise genuine issues of material fact. The affidavit submitted by Niagara, from James Cifaratta, its Director of Unregulated Generation, flatly disputes the assertions of competition. For example, so far as competition by QFs in the sale of power is concerned, Cifaratta asserts that arrangements under which QFs sell electricity in an unregulated market (i.e., outside the shelter of the PURPA mandate) are uncommon and "truly exceptional." See Cifaratta Decl. 57 11-12. He further says that the long term contracts that QFs have with their steam hosts--buyers of the thermal energy produced by cogeneration--preclude competition among the QFs for such hosts. Id. p 13. This theoretically leaves competition among the QFs as contracts expire. But our decision in National Parks II, that the district court was clearly erroneous in finding that certain concessionaires faced substantial competitive harms in contract renewal when the contracts were for long periods and thus renewal competition would only occur infrequently, 547 F.2d at 681-82, suggests that a competitive injury is too remote for purposes of Exemption 4 if it can occur only in the occasional renegotiation of long-term contracts. Niagara's response thus puts in dispute whether there is a likelihood of substantial competition among QFs in contract renegotiation with their steam hosts.

Further, though implicitly accepting DOE's and the QF Intervenors' assumption that QFs' competition with their steam hosts for the division of rents (quite apart from the long terms of the contracts) qualifies as "competition" for purposes of National Parks, Niagara contests the claim that they actually do so. See Cifaratta Decl. p 14. ("[T]he arrangements between QFs and their steam hosts typically are not determined by ordinary market-based concerns ... QFs often provide steam to their hosts at very low rates, sometimes for free. Some QFs, in fact, subsidize their steam hosts in order to secure PURPA and PSL § 66-c benefits for themselves."). While this struggle-free relation may seem unlikely, the contention directly contradicts the assertions on which the district court relied.

DOE's other arguments relating to competitive injury are legally inadequate under the National Parks standard. For example, DOE argues that the QFs may face future or potential competition. But the test explicitly requires proof that the submitters face actual competition. National Parks II, 547 F.2d at 679. DOE also insinuates that Niagara's interest in challenging the regulatory entitlement of QFs shows a competitive relationship between the QFs and Niagara. The argument would make sense only if a producer's regulatory entitlement to governmentally administered prices could be said to put the producer in "competition" with the involuntary purchaser. But if National Parks I embraced any such expansive idea of competition, the case would have come out differently. There the court recognized that the private sources of the disputed data, park concessionaires, enjoyed monopoly contracts with the Park Service, contracts that by statute were to be renewed so long as the concessionaires performed satisfactorily. 498 F.2d at 770 n. 20. The data sought to be collected bore on that performance. If the court thought that a firm's interest in protecting such an entitlement from the outsider scrutiny qualified as a competitive interest, it would have affirmed the district court's application of Exemption 4.

As each legally sound theory offered by DOE is plagued by factual disputes, summary judgment was improper. On remand the district court will want to consider the Supreme Court's observation that "categorical decisions may be appropriate and individual circumstances disregarded when a case fits into a genus in which the balance characteristically tips in one direction," U.S. Dep't of Justice v. Reporters Committee for Freedom of the Press, 489 U.S. 749, 776, 109 S. Ct. 1468, 103 L. Ed. 2d 774 (1989); see also Critical Mass, 975 F.2d at 879. Thus, a finding that QFs as a class generally do or do not have competitive interests that would be injured by release of the information on Form EIA-867 Form may be suitable.

Niagara also claimed that Form EIA-867 information is already in the public domain--a proposition that if true would give victory to Niagara independent of the matters discussed above. Niagara's position here is a little odd: if the information is publicly available, one wonders, why is it burning up counsel fees to obtain it under FOIA? But the logic of FOIA compels the result: if identical information is truly public, then enforcement of an exemption cannot fulfill its purposes. See Davis v. U.S. Dep't of Justice, 968 F.2d 1276, 1279 (D.C.Cir.1992) (Exemptions 7(C) & 7(D)); CNA Financial Corp. v. Donovan, 830 F.2d 1132, 1154 (D.C.Cir.1987) (Exemption 4); Afshar v. Dep't of State, 702 F.2d 1125 (D.C.Cir.1983) (Exemptions 1 & 3). On this issue the party favoring disclosure has the burden of production, for otherwise the party opposing disclosure would theoretically have to identify all public sources not reproducing the information. Id. at 1130. Niagara has sought to meet the burden with the argument that the data on Form EIA-867 is substantially equivalent to the data the QFs are required to file on Form 556 in their applications to the Federal Energy Regulatory Commission for QF certification, which is publicly available.

Before this court, both DOE and the QF Intervenors claim that the information on Form EIA-867 is narrower in scope than that submitted on Form 556. But that claim is at odds with their position before the district court. At argument on the summary judgment motion, counsel for the QF Intervenors conceded that the information provided by QFs upon initial certification was "substantially identical to that found in form 867 and that information is public." And DOE counsel supported this observation with equally emphatic language: "[The information] may be absolutely totally identical, but it's projected. It's not actual, and that's the big difference." Although obviously there is a world of difference between projected and actual data, these positions either assert or assume that the information on the two forms is identical in scope.

The district court accepted DOE's argument that while the certification information on Form 556 was only projected, the operational and performance information on Form EIA-867 was drawn from actual experience. Niagara did not dispute this, but countered by arguing that FERC regulations require QFs to make corrective filings once there are material changes in a QF's operations. See 18 U.S.C. § 292.207(d)(1). But as the district court observed, such new information is required only when there are material changes in facts and representations included in the initial self-certification filing. Since these corrective filings are not necessarily made on an ongoing and continuous basis, Niagara seems to have initially failed to carry its burden.

But that is not the end of the story. After the district court's holding in this case, the New York Public Service Commission rendered a decision authorizing electric utilities in New York to monitor the compliance standards set out in PURPA. See Re Motion to Establish Programs for Monitoring Qualifying Facility Status, Nos. 96-E-0775, 95-E-0264, 1997 WL 114364 at * 1 (N.Y.P.S.C. Jan.9, 1997). It is undisputed that this decision requires New York QFs to provide actual current performance data of the sort required for Form 556. Niagara claims that this requirement puts the requested information squarely in the public domain.

In response to the specific question why in light of this availability Niagara is still trying to obtain the Form EIA-867 information, Niagara responds that the Public Service Commission decision required information only from 1994 onwards, and that it wants to relieve itself (and in the end presumably its customers) from the costs of erroneous PURPA applications from earlier years. In fact, the time disparity is worse than that argument suggests, because the Federal Energy Regulatory Commission, though rejecting a claim by New York independent power producers that the New York decision was preempted by PURPA, denied it any effect as to data before the New York independents were on notice of the requirement,1 on the ground that such a mandate would impose an undue burden on the producers. See Independent Power Producers of New York, Inc., 80 FERC p 61,125 at 61,399 (1997).

Before us DOE and the QF Intervenors try to undercut the relevance of the New York decision by arguing that the scope of information on Form EIA-867 is materially broader than that on Form 556. But that was the scope claim that they effectively disavowed in district court; to allow them to raise it now for the first time on appeal would be grossly unfair to Niagara.

But Niagara is still by no means home free on this issue. For the period of most concern to Niagara, i.e., before the effective date of the Public Service Commission decision as modified by FERC, the New York mandate obviously fails to put the data into the public domain. But even as to those earlier data Niagara may have an argument on remand.2 Even if the district court finds that the QFs are in competition that could be adversely affected by disclosure of the earlier data if it alone were disclosed, those data may turn out to add so little to what is covered by the New York decision that its public disclosure will cause no additional competitive harm.

* * *

The district court's order granting DOE's motion for summary judgment is vacated and remanded.

So ordered.