BYRD v. U.S. E.P.A.

Opinion for the court filed by Circuit Judge HENDERSON. Separate opinion concurring in part and dissenting in part filed by Circuit Judge WILLIAMS.

KAREN LeCRAFT HENDERSON, Circuit Judge:

Appellant Daniel M. Byrd seeks reversal of the district court's grant of summary judgment to the Environmental Protection Agency (EPA) on his claim that EPA violated the Federal Advisory Committee Act (FACA), 5 U.S.C.App. II §§ 1-15. Specifically, Byrd contends that a peer review panel convened by an EPA contractor, the Eastern Research Group (ERG), to update EPA's interim benzene report constituted a federal "advisory committee" and therefore its proceedings were governed by FACA, with which it admittedly did not comply. Byrd seeks either reversal and a declaration that the panel's proceedings violated FACA or, alternatively, remand for discovery pursuant to Fed.R.Civ.P. 56(f). EPA counters that Byrd lacks standing, his claim is now moot and he is wrong on the merits. We affirm for the reasons set forth below.

I. BACKGROUND

In 1985, EPA issued an interim report discussing the carcinogenic effects of benzene. By 1996, EPA had prepared a draft update of its interim benzene report (Benzene Update). See Sonawane Decl. ¶¶ 2-5, Joint Appendix (JA) 173-75. Before finalizing the Benzene Update, EPA decided to subject it to external peer review.

Under a contractual arrangement with EPA, ERG, a private environmental consulting firm, convened and conducted the peer review. See id. ¶ 5, JA 175. The contract required ERG to select a panel of qualified experts, organize a public meeting of the panel to discuss the proposed Benzene Update and compile and submit a report to EPA summarizing the panel's assessment. See Statement of Work at 1-7, JA 184-90; Work Plan for Work Assignment No. 0-5 Contract No. 68-C6-0041, Expert Panel Peer Review of Benzene Risk Assessment Update (May 14, 1997) [hereinafter Work Plan], JA 199-204. In addition, the contract specified that EPA was to pay ERG a fixed sum and that ERG was to compensate the panel members. See Work Plan, JA 201. The contract also allowed EPA to determine the issues for the panel to evaluate and to comment in writing on ERG's draft final report. See Statement of Work at 5, JA 188.

Pursuant to the contract, EPA submitted to ERG for its consideration a list of twenty-four scientists who, in EPA's view, possessed the professional credentials necessary to serve on the peer review panel. See JA 192-93 (list of potential panelists). From the list, ERG selected four individuals to be panelists. ERG also selected two panelists from its own database of consultants. See EPA Mem. from Barbara Cook to Billy Oden, Re: Work Plan/Cost Estimate Approval, ERG Contract No. 68-C6-0041, WA 0-5 (June 9, 1997) [hereinafter 6/9/97 Mem.], JA 220; 6/13/97 Letter, JA 221. EPA suggested no modifications to the list of panel members selected by ERG. See 6/9/97 Mem., JA 220; 6/13/97 Letter, JA 221; see also Statement of Work at 2, JA 185 (stating that "final approval of selected experts will be made by EPA").

On June 27, 1997 EPA held a teleconference with ERG and the selected panelists, during which the panelists were instructed to prepare pre-meeting comments on the draft Benzene Update "specifically addressing a series of questions that [EPA] had provided" to ERG. Sonawane Decl. ¶ 7, JA 176. The panelists circulated their pre-meeting notes among themselves and provided a copy to EPA. See id. ¶ 8, JA 176. On June 30, 1997 EPA gave public notice in the Federal Register of the panel's scheduled meeting. See Draft Carcinogenic Effects of Benzene: An Update, 62 Fed.Reg. 35,172, 35,172-73 (1997), JA 213-14. The Federal Register notice explained the purpose of the meeting and noted that the draft was publicly available on the Internet or in writing from EPA. The notice also stated that ERG was to provide "logistical support for the workshop" and that interested persons could attend and participate in the meeting and advised that written comments could be submitted to EPA during a 60-day period ending August 29, 1997. 62 Fed.Reg. at 35,173, JA 214.

The panel meeting took place as scheduled on July 16, 1997. "The meeting was managed by ERG. Although several EPA employees who had been involved in developing the draft benzene update attended the meeting and effectively participated ..., no EPA employee or officer supervised the conduct of the meeting."1 Byrd Decl. ¶ 8, JA 345. Byrd, a self-employed "consulting toxicologist and risk assessor," id. ¶ 2, JA 342, also attended after "learn[ing] about the [July 16, 1997] meeting through EPA's [public notice] in the Federal Register."2 Id. ¶ 4, JA 344. Byrd participated in the meeting, twice expressing his views to the panel and others present. In addition, because of his concerns regarding the assumptions underlying the Benzene Update and his desire to be more informed, Byrd had earlier sought a copy of the panel members' pre-meeting notes but had been rebuffed three times. See id. ¶¶ 11, 13-15, JA 345-47; Sonawane Decl. ¶¶ 12-13, JA 177-78. Byrd made no additional attempt at the meeting to secure the notes. After the meeting, Byrd timely submitted written comments to EPA on the draft Benzene Update. See Sonawane Decl. ¶ 15, JA 178.

On August 22, 1997, Byrd filed this action alleging that the expert panel assembled by ERG was an "advisory committee" within the meaning of FACA3. Byrd sought both declaratory relief and a use injunction barring EPA from using the panel's work product. See Compl. ¶ 16. One month later, ERG submitted to EPA its final report, including its analysis of the draft Benzene Update. See Sonawane Decl. ¶ 14, JA 178; Schalk Decl. ¶ 8, JA 219; Panel Report, JA 228-329. EPA "did not participate in ERG's preparation of the final report." Sonawane Decl. ¶ 14, JA 178.

On October 10, 1997, almost three months after the meeting, Byrd's counsel wrote a letter to EPA's FOIA officer requesting a copy of the panel's pre-meeting notes. See Letter from Thomas R. Bartman to Jeralene Green, EPA, Re: Written Comments Prepared for or by Members of the Advisory Committee Convened July 16, 1997 (Oct. 10, 1997), JA 216. EPA provided all of the requested notes and invited Byrd to submit additional comments. See Letter from William H. Farland, Director, Office of Research and Development, to Thomas R. Bartman, Re: FOIA Request HQ-Rin-00186-98 (Nov. 14, 1997), JA 215. Byrd, however, declined to do so. EPA then moved to dismiss Byrd's complaint or, alternatively, for summary judgment. EPA challenged Byrd's standing and, on the merits, argued that the peer review panel assembled by ERG was not an "advisory committee" under FACA. The district court ruled in favor of EPA. Byrd v. EPA, C.A. No. 97-1923 (D.D.C. May 1, 1998) (Mem. and Order) [hereinafter Mem. & Order], JA 5-9. Although it "assum[ed] without deciding" that Byrd had standing, Mem. & Order at 2-3 n.1, JA 6-7, the district court held that a panel convened by a private contractor is not a FACA "advisory committee" as that term has been construed by the Supreme Court and by this Court. See id. at 2-5, JA 6-9 (citing Public Citizen v. United States Dep't of Justice, 491 U.S. 440, 109 S.Ct. 2558, 105 L.Ed.2d 377 (1989), and Food Chem. News v. Young, 900 F.2d 328 (D.C.Cir.), cert. denied, 498 U.S. 846, 111 S.Ct. 132, 112 L.Ed.2d 99 (1990)). Byrd timely filed his appeal.

II. DISCUSSION

A. Standing

EPA first attacks Byrd's standing to bring this action. Although the district court "assum[ed] without deciding" Byrd's standing, Mem. & Order at 2-3 n.1, JA 6-7, its approach is incorrect in light of the Supreme Court's recent holding in Steel Company v. Citizens for a Better Environment, 523 U.S. 83, 118 S.Ct. 1003, 140 L.Ed.2d 210 (1998), that standing is a "threshold jurisdictional question" that cannot be assumed in resolving litigation. Id. at 1016. "Moreover, because Article III standing is always an indispensable element of the plaintiff's case, neither we nor the Congress can dispense with the requirement—even if its application renders a FACA violation irremediable in a particular case." Natural Resources Defense Council v. Pena, 147 F.3d 1012, 1020 (D.C.Cir.1998) (NRDC); see also Federal Express Corp. v. Air Line Pilots Ass'n, 67 F.3d 961, 963 (D.C.Cir.1995) ("The requirement of a case or controversy is no less strict when a party is seeking a declaratory judgment than for any other relief."). Therefore, we must decide EPA's challenge to Byrd's standing.

The Steel Company holding requires us to focus on three elements:

118 S.Ct. at 1016-17 (quotations and citations omitted).

According to the Supreme Court, a refusal to provide information to which one is entitled under FACA constitutes a cognizable injury sufficient to establish Article III standing. See Public Citizen, 491 U.S. at 449, 109 S.Ct. 2558 ("refusal to permit appellants to scrutinize [committee's] activities to the extent FACA allows constitutes a sufficiently distinct injury to provide standing to sue"). By denying Byrd timely access to the panel's written comments and pre-meeting notes, EPA directly caused his informational injury. See Byrd Decl. at 6 ¶ 15; Sonawane Decl. at 5-6 ¶ 13, JA 177-78; Panel Report at 30, JA 260. EPA therefore can make no serious challenge to the injury and causation elements of Byrd's standing. See Food Chem. News v. Department of Health & Human Servs., 980 F.2d 1468, 1469 (D.C.Cir.1992) ("[W]henever practicable, all [Federal Advisory Committee] materials must be available for public inspection and copying before or on the date of the advisory committee meeting to which they apply.") (emphasis added).

EPA does question whether Byrd can meet the redressability prong. It first contends that declaratory relief will no longer redress Byrd's inability to obtain timely access to the panel's documents because they have since been made available and the panel has completed its work and been disbanded. See Appellee's Br. at 13, 21-27; Sonawane Decl. at 6 ¶ 14, JA 178; Schalk Decl. ¶ 8, JA 219. EPA also stresses that declaratory relief will not prevent additional informational injuries resulting from any future noncompliance with FACA. [See Appellee's Br. at 21-27.] If Byrd had simply complained that EPA failed to release the documents he requested, his alleged injury could not be redressed by any action of this Court because he ultimately received the materials. Byrd's injury, however, resulted from EPA's failure to furnish him with the documents until long after they would have been of any use to him. Thus, contrary to EPA's contentions, declaratory relief will redress Byrd's injury because it will provide him with this Court's declaration that the agency failed to comply with FACA; and such a declaration will give Byrd "ammunition for [his] attack on the Committee's findings" in subsequent agency proceedings that make use of the Benzene Update. NRDC, 147 F.3d at 1026 n. 6. Such an attack might also prompt, in view of the importance placed on the Benzene Update by EPA, see Statement of Work at 1 (contracting with ERG to conduct "category 1 peer review of the draft benzene document"), JA 184; Sonawane Decl. at 3 ¶ 4 ("`Category 1' peer review is used when major scientific or technical work products are being generated. . . ."), JA 175, additional, FACA-compliant peer review on the issue. Moreover, declaratory relief might well cause EPA to reevaluate and change peer review practices not in conformity with FACA. Accordingly, we conclude Byrd has standing to maintain his action.

B. Mootness

EPA also contends that Byrd's request for declaratory relief is moot because it has already given him the panel's pre-meeting notes and it is not engaged in any ongoing violation of FACA. Nevertheless, "even the availability of a `partial remedy' is `sufficient to prevent [a] case from being moot'." Calderon v. Moore, 518 U.S. 149, 150, 116 S.Ct. 2066, 135 L.Ed.2d 453 (1996) (quoting Church of Scientology v. United States, 506 U.S. 9, 13, 113 S.Ct. 447, 121 L.Ed.2d 313 (1992)). Because Byrd's injury resulted not only from EPA's failure to provide him materials but also from the tardiness of their eventual release, his injury would be mooted if EPA convened another panel to review the Benzene Update in compliance with FACA and provided him with all panel documents either before or at the meeting. Because EPA has not taken such action, declaratory relief would afford Byrd some relief and prevent his action from becoming moot.

Byrd also argues that EPA has a policy of hiring contractors to conduct peer reviews without following FACA requirements. See Payne Enters., Inc. v. United States, 837 F.2d 486, 491 (D.C.Cir.1988) ("So long as an agency's refusal to supply information evidences a policy or practice of delayed disclosure or some other failure to abide by the terms of the FOIA, and not merely isolated mistakes by agency officials, a party's challenge to the policy or practice cannot be mooted by the release of the specific documents that prompted the suit.").4 Thus, the tardy release of the documents does not render the case moot because Byrd's challenge to the policy remains. See United States v. W.T. Grant Co., 345 U.S. 629, 632, 73 S.Ct. 894, 97 L.Ed. 1303 (1953) ("[V]oluntary cessation of allegedly illegal conduct does not deprive the tribunal of power to hear and determine the case, i.e., does not make the case moot. A controversy may remain to be settled in such circumstances, e.g., a dispute over the legality of the challenged practices. The defendant is free to return to his old ways. This, together with a public interest in having the legality of the practices settled, militates against a mootness conclusion. For to say that the case has become moot means that the defendant is entitled to a dismissal as a matter of right. The courts have rightly refused to grant defendants such a powerful weapon against public law enforcement.") (citations omitted).

C. The Merits

FACA defines an "advisory committee" as

5 U.S.C.App. II, § 3(2) (emphasis added). Because EPA did not "establish" nor did it "utilize" the panel within the meaning of section 3(2) of FACA, we affirm the district court's grant of summary judgment to EPA.5 The district court treated EPA's motion for dismissal and summary judgment as a motion for summary judgment and on that basis granted the motion. See Mem. & Order at 2, JA 6.

Relying on legislative history, Byrd suggests that "established" and "utilized" should be construed "in their most liberal sense, so that when an officer brings together a group by formal or informal means, by contract or other arrangement . . . to obtain advice and information, such group is covered by [FACA]." Appellant's Br. at 11 (quoting S.Rep. No. 92-1098, reprinted in V. McMurty, Fed. Advisory Comm. Act (Pub.L. 92-463), Source Book: Legislative History, Texts, and Other Documents at 158 (Cong. Res. Serv.1978)). The Supreme Court, however, in Public Citizen v. United States Dep't of Justice, 491 U.S. 440, 109 S.Ct. 2558, 105 L.Ed.2d 377 (1989), squarely rejected an expansive interpretation of the words, reading "established" and "utilized" narrowly to prevent FACA from sweeping more broadly than the Congress intended. See 491 U.S. at 452, 461, 109 S.Ct. 2558 (finding "utilized" a "wooly verb" and declining to adopt dictionary meanings of "established" and "utilized" in FACA); see also Animal Legal Defense Fund v. Shalala, 104 F.3d 424, 427 (D.C.Cir.) (noting "the term `utilized' was given a very narrow interpretation by the Supreme Court") (ALDF), cert. denied sub nom., National Academy of Sciences v. Animal Legal Defense, ___ U.S. ___, 118 S.Ct. 367, 139 L.Ed.2d 285 (1997). In addition, the Court indicated that an advisory panel is "established" by an agency only if it is actually formed by the agency, see id. at 452, 456-57, 109 S.Ct. 2558, and "utilized" by an agency only if it is "amenable to . . . strict management by agency officials," id. at 457-58, 109 S.Ct. 2558. The Court, therefore, held FACA inapplicable to the American Bar Association Standing Committee on the Federal Judiciary, rejecting the argument that that committee had to comply with FACA simply because the Department of Justice regularly sought its input regarding judicial nominees. See id. at 452-67, 109 S.Ct. 2558.

We have similarly interpreted "established" and "utilized." For example, in Food Chemical News v. Young, 900 F.2d 328 (D.C.Cir.) cert. denied, 498 U.S. 846, 111 S.Ct. 132, 112 L.Ed.2d 99 (1990), we held that a panel assembled by the Federation of American Societies for Experimental Biologies (FASEB) pursuant to a formal contract to advise the Food and Drug Administration (FDA) on food safety was not an advisory committee subject to FACA. In so holding, we explained that "`established' indicates `a Government-formed advisory committee,' while `utilized' encompasses a group organized by a nongovernmental entity but nonetheless so `closely tied' to an agency as to be amenable to `strict management by agency officials'." Id. at 332-33 (quoting Public Citizen, 109 S.Ct. at 2568, 2570) (footnote omitted). We have interpreted "utilized" to encompass "management ... `by [any] semiprivate entity the Federal Government helped bring into being.'" Id. at 333 (quoting Public Citizen, 109 S.Ct. at 2571) (alteration original); see also ALDF, 104 F.3d at 427 (noting Supreme Court and this Circuit have adopted "`management and control' test to determine whether a committee not established by a government agency is nevertheless `utilized'"). This "second prong" of Food Chemical News "utilized" standard is inapplicable here because EPA is a governmental agency and ERG is not an entity the government had a role in creating. Thus, contrary to the broad standard suggested by Byrd, "the utilized test is a stringent standard, denoting `something along the lines of actual management or control of the advisory committee.'" ALDF, 104 F.3d at 430 (quoting Washington Legal Found. v. Sentencing Comm'n, 17 F.3d 1446, 1450 (D.C.Cir.1994)) (emphasis original). Indeed, this Court has held that participation by an agency or even an agency's "significant influence" over a committee's deliberations does not qualify as management and control such that the committee is utilized by the agency under FACA. See Washington Legal Found., 17 F.3d at 1451.

Although this Court has held that an agency "establishes" a committee only if the agency forms the committee, see Food Chem. News, 900 F.2d at 332, Byrd contends that EPA "effectively created" the panel by "conceiving of the need for" it and implementing it by hiring ERG to handle the logistics. Appellant's Br. at 16-17 (noting EPA's presentation of panel as its own in Federal Register notice and at public meeting). According to Byrd, EPA's actions are unlike those of the FDA in Food Chemical News in that, there, the contractor (not the agency) "proposed using ad hoc groups of knowledgeable experts as a means of carrying out the contract." Id. at 13 (quoting Br. of Resp't in Opp'n, Food Chem. News, No. 90-23 (in Supreme Ct. on pet. for writ of cert.)). But our analysis of whether an advisory committee has been "established" does not turn on a determination of who determines the methodology or operation of the peer review. Notably, the contractors in both Food Chemical News and here received a "task order" or a "work assignment" from the relevant agency defining the objective, the method and the scope of the studies to be performed. See Food Chem. News, 900 F.2d at 330; Statement of Work at 1-7, JA 184-90; Work Plan, JA 199-204. Moreover, because ERG selected the membership of the benzene panel, see 6/9/97 Mem., JA 220; 6/13/97 Letter, JA 221, Byrd cannot show that it was "`a Government-formed advisory committee'" as required by our narrow interpretation of "established." Food Chem. News, 900 F.2d at 332 (quoting Public Citizen, 109 S.Ct. at 2570). Byrd nevertheless argues that EPA established the panel because it retained the power to approve ERG's panel member selections. Although EPA provided a list of suggested panel members to ERG, ERG was not required to select its members from that list and two of the panel members were not on the EPA list.6 See JA 192-93 (list of potential panelists); Work Plan, JA 201; 6/9/97 Mem., JA 220; 6/13/97 Letter, JA 221; Panel Report App. A, JA 275-76. Moreover, EPA approved ERG's panel member selections without changes. See Sonawane Decl. ¶ 6, JA 176; 6/9/97 Mem., JA 220; 6/13/97 Letter, JA 221. Finally, ERG, not EPA, paid the panelists from its own funds. See Schalk Decl. ¶ 4, JA 218. Although the contract between ERG and EPA afforded EPA significant potential authority in the panel selection process, EPA never fully exercised it. And there is no reason to assume that the threat of an EPA veto affected ERG's panel selections. The result in this case might have been different if EPA had exercised its authority. The record, however, belies any claim that EPA in fact "established" the panel as required by FACA. The statute describes a panel that "is . . . established," 5 U.S.C.App. II, § 3(2), not one that could have been established by a government agency. Accordingly, EPA did not establish the benzene panel within the meaning of FACA.

Byrd also contends that EPA "utilized" the benzene panel because it exercised much more control over it than the agencies in Food Chemical News and Washington Legal Foundation exercised over the committees at issue in those cases.7 See Appellant's Br. at 14-15 (asserting EPA provided list of experts from which ERG was to select panel, reserved final authority to approve composition of panel, consulted with ERG on choice of chairman and agenda, presented charge to panel in pre-meeting conference call and reserved right to make written comments on ERG draft report). But even assuming EPA exercised more influence here than did the FDA or the DOJ in relation to their committees, EPA did not manage and control the benzene panel within FACA's scope, keeping in mind that "the utilized test is a stringent standard, denoting `something along the lines of actual management or control of the advisory committee.'" ALDF, 104 F.3d at 430 (quoting Washington Legal Found., 17 F.3d at 1450) (emphasis original). As we held in Washington Legal Foundation, even "significant" influence does not represent the level of control necessary to establish that a government agency "utilized" an advisory panel. 17 F.3d at 1451 ("But influence is not control.").

Contrary to Byrd's contention, the record shows that ERG in fact actually managed and controlled the selection of the panel's membership. See Mem. & Order at 4 n.2, 5; Sonawane Decl. at 4, JA 176; 6/9/97 Mem., JA 220; 6/13/97 Letter, JA 221. Moreover, as even Byrd admits,

Byrd Decl. ¶ 8, JA 345; see Schalk Decl. ¶ 6, JA 219. Finally, ERG, rather than EPA, prepared the report of the panel's proceedings. See Statement of Work at 5, JA 188; Work Plan, JA 204. Although the contract authorized EPA to receive and comment on the draft report before it was finalized, the district court found "no evidence that EPA's input, if any, resulted in changes being made to the final Expert Panel Report." Mem. & Order at 4 n.2, JA 8; see Sonawane Decl. ¶ 14, JA 178 (EPA "did not participate in ERG's preparation of the final report."). Because our decision is based on what EPA in fact did, rather than on what it could have done under its contract with ERG, we conclude that EPA's actions regarding the benzene panel do not constitute "actual management and control." ALDF, 104 F.3d at 430; Washington Legal Found., 17 F.3d at 1450. Accordingly, the district court correctly determined that the benzene panel was not subject to the constraints of FACA because EPA neither "utilized" nor "established" it.

For the foregoing reasons, the district court's grant of summary judgment to the Environmental Protection Agency is

Affirmed.8

STEPHEN F. WILLIAMS, Circuit Judge, concurring in part and dissenting in part:

I agree with the majority that we have jurisdiction, albeit on a different theory. On the merits, however, though the case is close, I would reverse.

Jurisdiction rests, I think, entirely on EPA's policy of using contractors to do peer reviews of risk assessments under arrangements like those involved in the Benzene Update that triggered this suit. Because Byrd is a regular participant in risk assessment panels, the threat of future injury from the policy is likely and imminent enough to justify standing. Jurisdiction based on the policy rather than the benzene episode suffers no mootness problem: EPA never claimed it would back away from the alleged policy; indeed, counsel at oral argument more or less admitted that the procedures used for benzene represented EPA's ongoing policy.

Unlike the future informational injuries that will flow from EPA's refusal to apply FACA to its contractors' consultative process, Byrd's injury from EPA's applying that view to the Benzene Update appears irredressable. His claim to the documents, of course, is mooted by EPA's FOIA officer's releasing them to him. And I do not see how a mere declaration that he should have had them at the time of the meeting constitutes redress for that loss. The majority suggests that a declaration would help Byrd attack this committee's findings on benzene if EPA wishes to use them in some future proceeding. Perhaps this provides standing for one claiming threatened injury-in-fact from the outcome of this future proceeding, but Byrd made no such claim. Further, such a declaration would seem a telling weapon for Byrd in a hypothetical future proceeding only if he asserted that the documents belatedly turned over enabled him to poke a hole in the substance of the peer review, a hole that he was unable to perceive on a timely basis because of EPA's original refusal to deliver them. But he has identified no such gap.

Nor do I think NRDC v. Pena, 147 F.3d 1012, 1026 n. 6 (D.C.Cir.1998), see Maj. Op. at 244, extended "informational injury" so far. That footnote merely observed that denying an injunction against future use of findings from a FACA-defective proceeding would not render FACA entirely toothless. One such tooth may be declaratory relief, and its utility in some cases may depend on the winner's being able to use it to delegitimate such findings. But nothing in Pena suggested that the prospect of securing such a benefit from the court could alone support standing as a general matter. The majority's language extending the "informational injury" redressable under FACA appears to assume that a highly theoretical injury is adequate for standing; the language is unnecessary to jurisdiction here.

On the merits, I believe that FACA governs panels established under the challenged policy. Our precedent on this language is rather thin, but appears to say that an agency "establish[es]" a panel if it has real control over its personnel and subject matter at its inception. Thus in Food Chemical News v. Young, 900 F.2d 328, 333 (D.C.Cir.1990), we said that the agency had not "established" the panel because the contractor "proposed" it, "alone selected its members," "set the panel's agenda," "scheduled its meetings," and "would have reviewed the panel's work." Here EPA proposes the use of a panel, submits an initial list of suggested members to the contractor, retains veto power over the final membership, and sets the panel's agenda. (The procedure used for the Benzene Update is evidently representative of EPA's practice.) The veto power is key. That it was not used in the benzene episode does not much help EPA: not only may EPA exercise it in future applications of the policy, but the contractor was and is quite likely to take the fact of veto power into account in its selection decisions. Assuming that contractors will ignore this fact—as the majority appears to do, see Maj. Op. at 247—seems akin to believing that the President takes no account of senators' opinions when he nominates federal judges.

Although the issue of whether EPA "established" the panel is certainly a close one, it seems to me inconsistent with the statute's language and intent to exempt from FACA a panel controlled so closely in membership and purpose.