Peer Review and Regulatory Reform

Lars Noah

The author is a Professor of Law at the University of Florida.

In recent years, lawmakers of all sorts have become interested in scientific peer review, and have the hope that scrutiny by independent experts can improve the quality of their own decisionmaking. As the phrase implies, peer review refers to the process of having work scrutinized by fellow experts, and it has long served as a quality control mechanism for the scientific community. Traditionally relegated to research funding and publication decisions, peer review recently has become of interest to regulatory agencies making decisions in the face of scientific uncertainty as well as to judges struggling to make sense of conflicting claims by expert witnesses. Although noteworthy, these applications remain fairly ad hoc to date.

More significantly, Congress has come to view expert peer review as a method for improving agency decisionmaking.¹ In the last several years, for example, regulatory reform proposals have included requirements that agencies call on independent scientists to check their work. The latest incarnation, the Regulatory Improvement Act of 1999 (S. 746), was reported out of the Senate Committee on Governmental Affairs last summer.² Like predecessor bills, S. 746 would mandate that agencies prepare risk assessments in many cases, and it would require some form of independent expert scrutiny of these and similar analyses.³

Several features of S. 746's peer review provision are notable. First, the bill does not specify a single peer review method. Instead, it may "be conducted through panels, expert bodies, or other formal or informal devices,"⁴ and the degree of formality "shall be commensurate with the significance and complexity of the subject matter."⁵ Second, the bill calls for "independent" peer review,⁶ which means scrutiny by experts "who are independent of the agency."⁷ Third, the bill would require "broadly representative" peer review groups,⁸ though the emphasis here appears to be on the resentation of a wide range of scientific disciplines rather than the inclusion of interested groups.⁹ Fourth, agencies would have to provide written responses to "significant" peer review comments,¹⁰ which would become part of rulemaking record,¹¹ presumably as fodder for judicial view. Finally, the bill recognizes and attempts to soften the added procedural burdens associated with routine external peer review of risk assessments and other analyses, in in by requiring only one round of scrutiny.¹²

Although controversy continues to surround the application of risk assessment to agency decisionmaking, it seem that essentially everyone applauds the idea of using independent peer review in the regulatory process. In a 1983 report, the National Resource Council concluded that "at agency's risk assessment should be reviewed by an independent scientific advisory panel before any major regulatory action."¹³ A few years later, the Administrative Conference of the United States (ACUS) endorsed the use of advisory committees more generally because "the quality and legitimacy of . . . agency decisions can be improved by counseling with eminent scientific experts outside the [30 ELR 10607] agency."¹⁴ In 1997, a report issued by the President's Commission on Risk Assessment and Risk Management echoed this conclusion.¹⁵ Along with imposing risk assessment requirements, S. 746 would implement such peer review recommendations.

Whether or not Congress ultimately enacts this measure, the bill provides a valuable template for evaluating regulatory peer review. The call for external peer review of scientific judgments by agency officials represents something of a throwback to the New Deal's enthusiasm for decision-making by expert regulators and a repudiation of the more recent conception of agencies as forums open to wide interest group representation. Even so, by harnessing nongovernmental expertise and retaining current mechanisms for review by nonexperts, regulatory peer review may help administrative agencies aspire to the deliberative ideal recently espoused by civic republican scholars. Although experts from outside the government may help improve the quality of risk assessments, this Article ultimately strikes a more ambivalent note about the desirability of routinely asking independent scientists to referee the work of federal regulatory agencies.¹⁶

Peer review, in its broadest sense, represents the scientific community's effort to police itself and to assure a certain minimum level of quality so that scientists and others can rely on the results of reported scientific research. Peer review contributes to the advancement of science not merely through the screening of scientific work (its quality control function), but also by helping proponents of new hypotheses to improve their research and interpretations and by engaging others in a dialogue about important new discoveries. Most importantly for present purposes, peer review as a mechanism differs from the broader notion of an unhurried dialogue among experts searching (usually in vain) for some ultimate truth. Far too frequently, it seems, these mechanisms are expected to provide a scientific stamp of approval for a particular claim.¹⁷ Instead, peer review helps allocate scarce resources—whether research grants, faculty posts, journal pages, or telescope time—to the most promising scientific work. As legal institutions often must do, peer review operates under constraints of limited time and information. Over the long haul, science may approach a gradual maturation of consensus, but the scientific community's various peer review techniques do not purport to anoint particular results as finally settling contested questions.

Unfortunately, policymakers often seem to conflate peer review with science itself, which in turnmay lead them to exaggerate the possible utility of independent expert scrutiny of decisions based on science. Recently, in the course of testifying in favor of S. 746, one scientist who had previously served as a high-ranking official at a pair of federal agencies explained to members of a congressional subcommittee that "peer review is the surety bond of science."¹⁸ Glowing endorsements of this sort exaggerate the potential value of independent expert scrutiny in reforming regulatory decisionmaking. Mark Powell offered a more cautious assessment of this development in his recent book Science at EPA: Information in the Regulatory Process:

Various forms of external scrutiny, including peer review, . . . have a relatively good track record of ensuring some level of quality of regulatory analysis. However, given the scientific uncertainties involved in most environmental regulatory decisions, peer review and official science advisory panels are no panacea—generally, reviewers can reveal uncertainties, but only additional research can resolve them. Relying too heavily on external reviewers and advisers to render judgments that inherently involve policy choices can result in shifting problems rather than solving them and abdicating policy formulation to unaccountable experts.¹⁹

Powell does, however, recognize the virtues of peer review at the EPA: "The agency's official science advisers serve as important sources of scientific information, as a quality assurance mechanism, and as a source of scientific credibility and legitimacy for decisionmaking."²⁰

If understood as a supplement to current opportunities for supervising agency actions, then the benefits of early brainstorming should counterbalance concerns about the associated costs of this additional procedure. If, however, viewed as a substitute for public involvement in, or judicial review of, agency decisionmaking, then regulatory peer review will not live up to its advance billing. This Article suggests ways of integrating external scientific scrutiny with existing rulemaking procedures and avenues for judicial review so as to maximize the potential utility of a peer review requirement.

Benefits of Regulatory Peer Review

Given serious resource constraints, agencies may do an inadequate job of addressing complex scientific information [30 ELR 10608] if left to their own devices. A regulatory peer review requirement may help counteract such agency tendencies toward superficiality. But peer review in the scientific community does more than attempt to screen out work of poor quality; it also creates opportunities for dialogue and feedback designed to improve scientific work. When adapted to the regulatory process, external peer review should engage agency staff in collaborative brainstorming with nongovernmental experts. Many of the existing mechanisms for peer review, at least those that federal agencies have established on their own initiative, serve such a consultative function.²¹

At one time, proposals for a "science court" attracted significant attention.²² If such an expert tribunal engaged in rather formal but passive scrutiny when an agency solicited input on technical questions, regulatory peer review should aspire to informal but more active deliberation. When structured to adjudicate scientific disputes between the agency and interested parties, as done by the U.S. Environmental Protection Agency's (EPA's) Scientific Advisory Panel (SAP) when reviewing proposals to cancel pesticide registrations,²³ peer review has not served the regulatory process as well as some had hoped. In contrast, when agencies consult outside experts before publicly committing themselves to a particular course of action, the proceedings become less adversarial and more productive, as reflected in the successful use of advisory committees by the Food and Drug Administration (FDA).²⁴ Indeed, one might say that soliciting expert input at the front-end of the regulatory process more closely resembles peer collaboration than peer review.

In some respects, regulatory peer review mandates have parallels to other innovations in notice-and-comment rulemaking. In the 1970s, Congress "hybridized" the procedures used by some agencies for the issuance of rules. For instance, the Consumer Product Safety Commission (CPSC) must issue advanced notices of proposed rulemaking. This provides interested parties with an earlier opportunity to comment and a rule emerges after a process based on mandatory consultations with industry.²⁵ More recently, Congress and the President have encouraged federal agencies to use negotiated rulemaking, in which agencies assemble interested parties in an effort to seek a preliminary consensus that provides the basis for a notice of proposed rulemaking.²⁶ Finally, recent amendments to the Regulatory Flexibility Act added a requirement that, for proposed rules likely to have a significant economic impact on a substantial number of small entities, EPA and the Occupational Safety and Health Administration (OSHA) must "convene a review panel" composed of agency employees to consider any comments received from small business representatives. A report for the rulemaking record must then be prepared.²⁷

Whatever their merits and limitations, these methods share an important feature: they attempt to improve rulemaking by eliciting feedback early in an agency's decisionmaking process. Ideally, regulatory peer review should operate in the same fashion. What distinguishes other forms of hybridized rulemaking from independent expert scrutiny—namely, the involvement of a wide range of interested parties—may also save a regulatory peer review mandate from the shortcomings experienced with these other methods for eliciting external input.

By not insisting on interest group representation or open access to deliberations, regulatory peer review requirements would differ from many of the agency advisory committees currently in use. Under the Federal Advisory Committee Act (FACA), agencies must provide for broad-based representation on advisory committees that they charter and must invite the public to attend meetings.²⁸ The courts have expanded the coverage of the statute to include a number of different mechanisms used for soliciting expert input.²⁹ In justifying the FACA exemption in S. 746, the Senate committee emphasized that "peer review is not intended to provide policy advice or analysis to an agency, and it is not a political debate among interested parties."³⁰ Congress made the same basic pointwhen it empowered EPA's Science Advisory Board (SAB): "The Board is not intended as a forum to be used by outside interests to criticize the workings of the Agency."³¹ In fact, President Reagan vetoed an amendment [30 ELR 10609] that would have required "fair balance" on the SAB, explaining that the peer review function "must remain above interest group politics."³²

Although it seems unrealistic to think that one can entirely separate scientific advice from policy, either in theory or in practice, the inevitable intrusion of policy does not require structuring the expert peer review process as a political forum, at least so long as other existing mechanisms for participation and accountability remain in place. If Congress acceded to critics' demands for open meetings and interest group representation, legislative over-specification of the procedures for peer review will make this hybridized process unduly cumbersome. If understood as a structured form of brainstorming by a group of scientists from different backgrounds, peer review cannot function as effectively if the process becomes excessively proceduralized.

Objections to Regulatory Peer Review

Critics of regulatory peer review requirements make a number of interrelated objections. First, they fear that peer review will exacerbate delays in rulemaking because of the additional time required for external scrutiny.³³ Second, they worry that agencies will have to divert scarce resources from enforcement and related activities to pay for these addedprocedures.³⁴ Third, they suspect that peer reviewers will have ties to industry that will influence their judgments.³⁵ Fourth, they object to the secrecy of peer review deliberations insofar as regulatory reform proposals exempt these panels from FACA.³⁶ Fifth, they expect that negative comments from peer reviewers will either stall valuable rulemaking initiatives prematurely or else make final rules more vulnerable to reversal on judicial review.³⁷

Few observers doubt that rulemaking has become "ossified" during the past few decades,³⁸ though disagreements persist about the precise causes and appropriate responses to this development. If the above-mentioned criticisms of mandatory peer review have merit, agencies may face additional disincentives when they contemplate issuing or revising health and safety rules—what some observers have christened "paralysis by analysis." Indeed, critics suggest that these sorts of reform proposals are designed for precisely that purpose, offering regulatory relief for industry in the guise of more rational procedures.³⁹ Procedural reforms often do disguise legislative policy preferences about the substance of agency decisionmaking. Although they may appear to improve judgments, better procedures are rarely neutral in their effects.⁴⁰

Even if one does not ascribe to the cynical explanation of peer review as regulatory relief, agencies may well attempt to circumvent such reforms by opting for less formal mechanisms not subject to these new requirements. Not only would such efforts at evasion mean foregoing the benefits of external peer review, but agency decisionmaking would also become less accountable to existing procedural safeguards for public participation and review by all three branches of the government. Nonetheless, given all of the existing disincentives as well as the offsetting advantages to agencies, such as ease of enforcement when they announce formally binding regulations, it seems unlikely that a peer review requirement alone would tip the balance against the issuance of rules. Moreover, the additional time and effort invested early in the rulemaking process may provide later dividends if regulatory peer review reduces the likelihood of successful judicial challenges.⁴¹

[30 ELR 10610]

The fears of secrecy seem exaggerated and they appear to spring from a basic misconception about the intended role of external peer review. As suggested previously, these panels should not provide still another forum for all interested stakeholders to influence the agency. Because agencies will publish the comments prepared by their external advisors, persons excluded from the deliberations will have an opportunity to respond. An insistence on open meetings may bog down peer review of agencies' judgments.⁴² The Freedom of Information Act (FOIA) recognizes the hazards of interfering in an agency's internal decisionmaking process by exempting pre-decisional deliberative process materials from disclosure requirements,⁴³ and at least one court has extended this exemption to cover unfinished scientific reports requested by a regulatory agency.⁴⁴ In a similar vein, regulatory peer review will function most effectively if protected from premature disclosure of working documents.

Concerns about the identity of peer reviewers also seem misplaced. In selecting members, agencies will attempt to identify those nongovernmental scientists who can offer their independent perspectives. Presumably, as has happened with existing advisory committees and boards, most of the members will come from academia. Critics suggest that even these researchers will harbor biases because the industry funds much of their work,⁴⁵ but, absent evidence of direct support of research relevant to the particular regulatory questions under review, this represents a fairly tenuous "taint" at worst. Ultimately, the drafters of S. 746 decided to defer to the conflict of interest rules that federal agencies currently apply to members of their existing advisory committees.⁴⁶ Although even misplaced fears of bias may weaken public confidence in the results of external peer review, which could undermine some of its value to the regulatory process, critics have overdrawn the claim insofar as they suggest that essentially no scientists have the independence necessary to serve as referees.

Concerns about the potential for added administrative burdens do, however, deserve serious attention. If it does not help steer an agency early in the process, peer review of regulatory decisionmaking may become an ominous hurdle for agencies to surmount, both in terms of the difficulty of undergoing that scrutiny and because of the prospect of judicial invalidation triggered by the inevitable criticisms from expert peer reviewers. A critical peer review may amount to a de facto veto and derail a rulemaking proceeding if the agency chooses to credit the expert objections and withdraws a proposal.⁴⁷ Even if a negative peer review does not discourage an agency from pursuing an initiative, the process may lead to delays in rulemaking.⁴⁸ Moreover, if agencies sense that they must get the blessing of outside scientists before proceeding with a rule, they may decide to settle for second-best regulatory options simply because these generate the least disagreement among the experts. Some may applaud this temporizing effect, but such a role for the referees would magnify the force of the previously mentioned criticisms that they lack accountability.

Making the Most of Regulatory Peer Review

Even if one can design mechanisms for external scientific scrutiny so as to minimize some of these shortcomings, one must also work to ensure that it fits comfortably with existing procedural requirements. Risk assessments undoubtedly can benefit from some peer review. Nonetheless, this expert scrutiny should not become a straightjacket for agency decisionmakers, and courts must appreciate the limitations of external peer review when asked to resolve challenges to agency rules that are subject to a risk assessment requirement. No matter how thorough their consideration, independent experts cannot certify the accuracy of an agency's scientific judgments.⁴⁹ Scientists who serve as members of agency advisory panels have only limited time that they can devote to this task,⁵⁰ so peer reviewers may, on [30 ELR 10611] the one hand, miss basic flaws in a risk assessment or, on the other hand, fail to raise fairly nuanced concerns about the interpretation of complex data.

Thus, peer review is best understood as a supplement to, rather than a substitute for, existing forms of external scrutiny.⁵¹ It cannot replace notice-and-comment rulemaking procedures or the possibility of judicial review of a regulation. Instead, by offering agency officials a preview of likely objections, independent scientific experts can help them anticipate and hopefully minimize weaknesses in a risk assessment or other predicate for a regulation, thereby sharpening subsequent reviews by interested private parties and officials in the three branches of government.⁵² All of this assumes, of course, that the risk assessment document serves as a predicate for rulemaking. In some cases, agencies issue free-standing risk assessments that have equally significant regulatory consequences but reduced opportunities for public participation or judicial review. In other cases, an agency may decide to abort a rulemaking initiative based on a failure to persuade its independent peer reviewers, which would also largely escape public and judicial scrutiny.

As suggested previously, scientific advisory committees or panels may have only limited utility if their input is not sought until fairly late in an agency's decisionmaking process. If conducted early in an agency's risk assessment, independent peer review can provide valuable expertise and diverse perspectives, and it can focus attention on at least some of the data gaps in time for corrections.⁵³ Early peer review also should minimize the temptation for scientists to become distracted with the opportunity to offer advice on questions of policy more properly left in the agency's domain.

Several commentators have described risk assessment as a "trans-scientific" exercise because of the inevitable data gaps and uncertainties that require policy or value judgments on which science has little to offer.⁵⁴ As the Supreme Court has recognized in a related context, "it is not infrequent that the available data does not settle a regulatory issue and the agency must then exercise its judgment in moving from the facts and probabilities on the record to a policy conclusion."⁵⁵ For this reason, peer review seems more appropriate with respect to the inputs for a risk assessment, as done in the Safe Drinking Water Act Amendments of 1996,⁵⁶ or the Clean Air Scientific Advisory Committee's (CASAC's) review of the criteria documents that provide the foundation for the national ambient air quality standards (NAAQS).⁵⁷ In addition, independent experts could offer valuable assistance to agencies in formulating generic risk assessment guidelines. With regard to specific risk assessments for particular rulemaking initiatives, however, some peer reviewers predictably will disagree about how best to interpret ambiguous research or resolve uncertainties.⁵⁸ For that reason, agencies should not feel hamstrung by failures to convince their expert advisory panels. Just as an editor of a scientific journal retains the prerogative to ignore comments provided by a referee, federal agency officials must not cede their power to pursue rulemaking to independent (and democratically unaccountable) peer reviewers.⁵⁹

Rulemaking Procedures

Panel peer review cannot serve as a substitute for notice-and-comment rulemaking procedures with respect to the adequacy of an agency's risk assessment. Although now viewed as a mechanism for compiling an administrative record for purposes of facilitating substantive judicial review, the Administrative Procedure Act's (APA's) rulemaking procedures originally were designed to help provide additional information and perspective to agency officials. Judge David Bazelon, for instance, expressed a strong preference for enhanced procedures rather than more searching judicial review of scientific decisions by agencies:

In cases of great technological complexity, the best way for courts to guard against unreasonable or erroneous administrative decisions is not for the judges themselves to scrutinize the technical merits of each decision. Rather, it is to establish a decision-making process which assures a reasoned decision that can be held up to the scrutiny of the scientific community and the public.⁶⁰

[30 ELR 10612]

In particular, he urged that agencies use "technical advisory committees comprised of outside experts with differing perspectives" as one way of encouraging a "genuine dialogue" about scientific issues.⁶¹

In a sense, the public comment period itself represents an independent peer review process.⁶² Just as editorial peer review provides only a first cut on the reliability of new research submitted for publication, followed by a far less structured but ultimately more important opportunity for post-publication peer review, panel reviews of draft risk assessments cannot provide a definitive seal of approval (or disapproval) of any risk assessment. Instead, just as referees can help authors improve a manuscript before publication, regulatory peer review may help the agency better anticipate adverse public comments.⁶³

Conversely, the notice-and-comment process does not make prior independent peer review redundant or an unnecessary source of additional delays. Truly disinterested experts do not typically go to the effort of filing comments unless an advocate has hired them for that purpose.⁶⁴ Regulatory peer review is best understood as a collaborative process where scientists from different disciplines are able to hash out disagreements with agency staff prior to the publication of a proposed rule.⁶⁵ Although one could avoid the delays associated with a sequential process by requesting peer review concurrently with the invitation for public comment, the extra effort invested at the outset may provide a net benefit by reducing the prospect of objections to a regulation that later may trigger time-consuming and resource-draining litigation.⁶⁶ If undertaken late in the rulemaking process, external peer review also can increase the possibility of procedural objections. For instance, courts may conclude that information brought to the agency's attention late in the process might undermine the right of interested parties to receive notice and a meaningful opportunity for comment.⁶⁷

Substantive Judicial Review

In reviewing agency action, courts often find it difficult to assess controverted scientific and technical issues. Because they generally lack the expertise to evaluate the basis for a rule or order in such cases, judges increasingly have deferred to the agency's substantive judgments. When an agency is making predictions "at the frontiers of science," the Supreme Court has stressed that "a reviewing court must generally be at its most deferential."⁶⁸ Courts have also come to appreciate the inevitable political character of these judgments.⁶⁹ Instead of struggling to critically evaluate the substantive merits of a rule, reviewing courts may focus on questions of procedural adequacy.⁷⁰ They also may take cues from the views of experts disclosed in the administrative record.

Some commentators have gone so far as to suggest that regulatory peer review could substitute for judicial review. For example, Professor Jasanoff has concluded that, for purposes of substantive review by the courts, expert scrutiny "certifies that the agency's scientific approach is balanced and rational and that its conclusions are sufficiently supported by the evidence."⁷¹ Conversely, she suggested that [30 ELR 10613] agency departures from peer review might presumptively invalidate agency action.⁷² In either case, such an approach would give essentially determinative weight to the conclusions of nongovernmental scientists. After devoting much of her important book on this subject to explaining the contingency of scientific knowledge and complaining that proponents of a technocratic model had "oversold" the possibilities of regulatory peer review, Professor Jasanoff's apparent exuberance about displacing the courts in this fashion comes as something of a surprise.

Although provocative, such a view misconceives the respective spheres of competence of these different mechanisms for scrutinizing agency decisions. A peer review panel should not become a science court, brought in at the end of the process and asked to resolve the conflicting positions of adversaries about issues that often transcend science. Such an approach seems problematic for at least two reasons: it occurs too late in the process, and it asks scientists to resolve policy questions.⁷³ Engaging in peer review early in the decisionmaking process would offer the best opportunity for encouraging genuine and meaningful deliberation about scientific uncertainties without usurping the responsibility for resolving nonscientific questions that Congress has delegated to agencies subject to supervision by the courts.

Because judicial review will remain important, courts will have to determine how to weigh comments from independent experts about the reliability of an agency's risk assessment. In one case, a federal court suggested that peer review should have a significant impact on its substantive review of an agency rule:

When new data or, as in this case, new mathematical extrapolations, provide the basis for the new rule, independent peer reviews may be extraordinarily helpful to the court. . . . Precisely because the data has [sic] not been scrutinized, however, the court has particular interest in having access to both favorable and unfavorable peer reviews.⁷⁴

Although such judicial interest in the comments of peer reviewers seems understandable, this deference to peer review reports may be misplaced.

Unless courts appreciate the limited function served by panel peer review of risk assessments, judicial review of agency regulations will not improve and, instead, may undermine the continued value of the peer review process. In one significant recent decision, a judge invalidated an EPA risk assessment report finding that environmental tobacco smoke (ETS) was a human carcinogen. The report was invalidated on procedural grounds because EPA apparently had failed to convene a representative committee as required under the Radon Research Act, as well as on substantive grounds. The judge also focused in part on critical peer reviewers' comments from a review conducted by EPA's Indoor Air Quality Committee.⁷⁵ But the court's resolution of the substantive challenge conflicted with that committee's ultimate conclusion, which generally had concurred with EPA's listing decision.⁷⁶ Although a reviewingcourt need not give presumptive weight to these favorable peer reviews and remains free to decide that EPA's decision was substantively flawed based on materials in the administrative record,⁷⁷ the court's opinion suggests one of the hazards of making peer review comments part of the administrative record—namely, the tendency to focus on the inevitable, usually constructive, criticisms found in any peer review report and taking them out of context. It also suggests a danger that courts will become unduly preoccupied with questions of procedural compliance.⁷⁸

In some ways, this question parallels an issue that the courts have struggled with in reviewing agencies' adjudicatory decisions. In evaluating an administrative record, courts must decide how much weight to assign the initial decision of an administrative law judge (ALJ) or hearing officer who heard the evidence and may have made credibility determinations to resolve conflicts in the evidence. After initially suggesting that federal agencies would have to give particular attention to an ALJ's determinations and explain a contrary decision, the Supreme Court decided that these initial decisions, while part of the record for review, do not deserve any special weight or otherwise limit the agency's [30 ELR 10614] ultimate decisionmaking authority in a case.⁷⁹ A similar approach to the comments of an independent peer review panel seems appropriate.

If an agency is specifically obligated to respond to all significant peer review comments (or the legislation otherwise elevates the importance of this step in the risk assessment process), then these comments may be given special solicitude by a reviewing court. At the other end of the spectrum, one might provide that the risk assessment itself and the peer review comments would not become part of the administrative record for purposes of judicial review, much as happens with internal staff reports not relied on by an agency. There is, of course, some danger that such an arrangement would grant agencies license to ignore valid scientific criticisms without any fear that a reviewing court would call the agency to task for dismissing such concerns. A better compromise would make the peer review panel's report part of the record but also instruct courts not to assign it any special weight. If simply made part of the record for judicial review, they should have no greater force than comments submitted by interested parties during the public comment period.⁸⁰

One can imagine different scenarios for substantive judicial review involving agency rules where peer review comments are simply made a part of the administrative record. A critical peer review should not derail the agency when defending a regulation in court,⁸¹ though the agency may have to justify its reasons for disagreeing with the advice of the independent experts,⁸² just as it must respond to significant public comments. Conversely, a favorable peer review should not provide an impenetrable shield against objections pressed by individuals who were not able to present their concerns directly to the peer review panel. In any event, a favorable peer review report would not insulate an agency's rule from judicial invalidation when challenged on other than substantive grounds.⁸³

Thus, apart from its potential contribution to the quality of agency decisionmaking, independent peer review may affect the course of judicial review in a variety of ways. At one extreme, legislation could invite reviewing courts to enforce strict compliance with new procedural hurdles, such as insisting on what may often be largely duplicative peer reviews of an agency's draft and final risk assessments. This may not appreciably improve agency decisionmaking, and it could certainly worsen the ossification of rulemaking. Under a more moderate scenario, legislation would strive to minimize the risk of undue judicial scrutiny of compliance with procedural mandates, but courts may come to view these procedural hurdles as codifying a substantive standard for nonarbitrary agency decisionmaking.⁸⁴ If this happens, judges again would engage in "hard look" review for agency decisionmaking at "the frontiers of science," a decidedly less deferential stance than the courts have announced for such cases. Finally, one can imagine that courts would neither review for compliance with new procedural requirements (leaving that task for other branches), nor adopt a more rigorous stance when engaging in substantive review, but instead undertake substantive review under the existing and fairly deferential standards with the benefit of more complete information and elucidation. Peer review by a panel of scientific experts can help agencies generate and refine such information; it cannot substitute even partially for continued but deferential public and judicial scrutiny of health and safety regulations.

Conclusion

Proponents of regulatory reform should not oversell the utility of independent peer review of technical judgments made by agencies. Notwithstanding its centrality as a quality control mechanism for the scientific community, peer review does not provide a bromide for what some observers think ails our government—namely, overzealous bureaucrats issuing health and safety rules based on a misunderstanding or conscious disregard of the best available data. Involving outside scientists in the process undoubtedly will promote enhanced deliberation and reflection, and these peer reviewers may help steer agencies clear of embarrassing and costly mistakes, but ultimately the independent experts cannot and should not displace the broader deliberative process about hard policy questions that science cannot answer. The New Deal faith in expertise continues to have strong appeal, especially as contrasted with the pluralist vision of agencies simply brokering compromises between competing interest groups. So long as independent peer review does not become a substitute for public participation or judicial review, it may provide a forum for genuine deliberation that can facilitate subsequent steps in the administrative process and help to better focus other forms of external scrutiny of agency decisionmaking.

1. See SHEILA JASANOFF, THE FIFTH BRANCH: SCIENCE ADVISERS AS POLICYMAKERS 32-34 (1990); id. at 16 ("One refrain heard with increasing frequency in recent years is that agencies should ensure that their decisions are peer-reviewed in accordance with the normal practices of the scientific community."); see also Thomas S. Burack, Of Reliable Science: Scientific Peer Review, Federal Regulatory Agencies, and the Courts, 7 VA. J. NAT. RESOURCES L. 27 (1987) (evaluating agency experiences under older peer review mandates).

2. Regulatory Improvement Act of 1999, S. 746, 106th Cong. (1999).

3. See id. § 625(a) (calling for peer review of newly mandated risk assessments as well as review of cost-benefit analyses of major rules reasonably likely to have an annual economic effect of $ 500 million).

4. Id. § 625(b)(1)(A).

5. Id. § 625(d).

6. See id. § 625(a).

7. Id. § 625(b)(1)(A)(ii). Members of existing agency advisory boards or panels established by statute would satisfy the requirement of independence. See id. § 625(f). Peer review members shall "be governed by agency standards and practices governing conflicts of interest of nongovernmental agency advisors." Id. § 625(b)(1)(B).

8. See id. § 625(b)(1)(A).

9. Seeid. § 625(b)(1)(A)(i) (calling for the inclusion of participants "with expertise relevant to the sciences or analyses involved in the regulatory decisions"); id. § 625(b)(1)(D) (demanding that peer reviews "contain a balanced presentation of all considerations, including minority reports"); id. § 625(c)(1) (requiring certification only of the independence and expertise of the peer reviewers). As further confirmation that the drafters intended broad disciplinary rather than stakeholder representation, the bill would exempt peer review bodies from the Federal Advisory Committee Act (FACA). See id. § 625(e); see also infra notes 28-29 and accompanying text (discussing FACA).

10. See Regulatory Improvement Act of 1999, S. 746, 106th Cong. § 625(b)(1)(D), (b)(2) (1999).

11. See id. § 625(b)(2)(B). The rulemaking record must also include a statement by an official who was not involved in that proceeding concerning (1) compliance with the requirements for peer reviewer independence and expertise, and (2) the adequacy of the agency's responses to the peer review comments. See id. § 625(c).

12. See id. § 625(b)(1)(C) (calling on the peer reviewers to "provide for the timely completion of the peer review including meeting agency deadlines"); id. § 625(b)(3) ("If the head of an agency, with the concurrence of the Director [of the Office of Management and Budget], publishes a determination in the rule making file that a cost-benefit analysis or risk assessment, or any component thereof, has been previously subjected to adequate peer review, no further peer review shall be required under this section . . . ."); id. § 625(h) ("Nothing in this section shall require more than one peer review of a cost-benefit analysis or a risk-assessment during a rule making. A peer review required by this section shall occur to the extent feasible before the notice of proposed rule making.").

13. NATIONAL RESEARCH COUNCIL, RISK ASSESSMENT IN THE FEDERAL GOVERNMENT: MANAGING THE PROCESS 156 (1983) [herein-after NRC REPORT].

14. 50 Fed. Reg. 52893, 52896 (1985); see also CARNEGIE COMM'N ON SCIENCE, TECH. & GOV'T, NEW FRONTIERS IN REGULATORY DECISION MAKING: THEROLE OF SCIENCE AND TECHNOLOGY 90 (1993) ("External science advisory boards serve a critically important function in providing regulatory agencies with expert advice on a range of issues.").

15. See PRESIDENTIAL/CONGRESSIONAL COMM'N ON RISK ASSESSMENT AND RISK MANAGEMENT, RISK ASSESSMENT AND RISK MANAGEMENT IN REGULATORY DECISION-MAKING 103 (1997) (concluding that peer reviews "enhance the credibility of agency decisions and . . . improve their technical quality").

16. For a more comprehensive treatment of these questions, see Lars Noah, Scientific "Republicanism"?: Expert Peer Review and the Quest for Regulatory Deliberation, 49 EMORY L.J. (forthcoming Dec. 2000).

17. See Lars Noah, Sanctifying Scientific Peer Review: Publication as a Proxy for Regulatory Decisionmaking, 59 U. PITT. L. REV. 677, 680 (1998) (suggesting that legal institutions have "expressed significant and arguably unwarranted confidence in the content of published scientific articles").

18. Regulatory Improvement Act of 1999 (S. 746): Hearings Before the Senate Comm. on Governmental Affairs, 106th Cong. (1999) (statement of Lester M. Crawford, Director, Center for Food & Nutrition Policy, Georgetown University) [hereinafter 1999 Hearing]; see also Regulatory Improvement Act of 1998 (S. 981): Hearings Before the Senate Comm. on Governmental Affairs, 105th Cong. 110 (1998) (statement of Bruce Alberts, President, NAS):

Many scientists and engineers who have devoted their careers to working on environmental problems are puzzled as to why anyone might oppose S.981. It is unclear why anyone would argue against obtaining better scientific information and analysis to aid federal officials and inform the American public when important regulatory decisions are made.

[hereinafter 1998 Hearing].

19. MARK R. POWELL, SCIENCE AT EPA: INFORMATION IN THE REGULATORY PROCESS 4 (1999); see also id. at 139 ("External peer reviewers and science advisers can serve an important quality control function, . . . but peer review is no cure-all and should not be viewed as a means of arbitrating policy decisions.").

20. Id. at 38.

21. See STEPHEN BREYER, BREAKING THE VICIOUS CIRCLE: TOWARD EFFECTIVE RISK REGULATION 68 (1993) (mentioning the creation of EPA's Science Advisory Board and other specialized advisory committees, and noting that "those who have studied the work of these committees find that they have led to significant improvements, . . . [in part because] consultation is often private and nonadversarial").

22. See James A. Martin, The Proposed "Science Court," 75 MICH. L. REV. 1058, 1069-86 (1977) (summarizing this suggestion, and emphasizing its adversarial character); Jeffrey N. Martin, Procedures for Decisionmaking Under Conditions of Scientific Uncertainty: The Science Court Proposal, 16 HARV. J. ON LEGIS. 443, 454-68 (1979) (contrasting the science court proposals with the use of scientific advisory committees). But cf. Joel Yellin, Science, Technology, and Administrative Government: Institutional Designs for Environmental Decisionmaking, 92 YALE L.J. 1300, 1307-12, 1332 (1983) (disputing the premise underlying the science court proposal that one can separate technical and political questions, and questioning the value of an adversarial process).

23. See JASANOFF, supra note 1, at 126, 139-41, 143-46 (documenting the adversarial style of SAP reviews); id. at 247 (characterizing the SAP as "a miniature science court").

24. See id. at 162-65, 178-79 (praising the FDA's use of advisory committees).

25. See 15 U.S.C. § 2058(a); see also Carl Tobias, Early Alternative Dispute Resolution in a Federal Administrative Agency Context: Experimentation With the Offeror Process at the Consumer Product Safety Commission, 44 WASH. & LEE L. REV. 409 (1987).

26. See 5 U.S.C. §§ 561-570; see also Jody Freeman, Collaborative Governance in the Administrative State, 41 UCLA L. REV. 1 (1997); Symposium, Negotiated Rulemaking, 46 DUKE L.J. 1255 (1997).

27. See Pub. L. No. 104-121, § 244(a), 110 Stat. 864, 867 (1996) (to be codified at 5 U.S.C. § 609(b) & (d)).

28. See 5 U.S.C. app. II § 2; see also Cargill, Inc. v. United States, 173 F.3d 323, 332-39 (5th Cir. 1999) (reviewing the National Institute for Occupational Safety and Health's compliance with FACA's fair balance requirements); Nicholas A. Ashford, Advisory Committees in OSHA and EPA: Their Use in Regulatory Decisionmaking, SCI. TECH. & HUMAN VALUES, Winter 1984, at 72, 76-80 (applauding FACA's application to scientific advisory committees); Steven P. Croley & William F. Funk. The Federal Advisory Committee Act and Good Government, 14 YALE J. ON REG. 451 (1997).

29. See Animal Legal Defense Fund, Inc. v. Shalala, 104 F.3d 424, 429-31 (D.C. Cir. 1997) (holding that FACA applied to the National Academy of Sciences' (NAS') guidance concerning the care of laboratory animals); National Nutritional Foods Ass'n v. Califano, 603 F.2d 327, 329, 335-36 (2d Cir. 1979) (holding that FACA applied to an ad hoc FDA advisory group); Food Chem. News v. Young, 709 F. Supp. 5, 6-9 (D.D.C. 1989) (holding that an expert scientific panel under contract to prepare a food safety report for the FDA was subject to FACA), rev'd, 900 F.2d 328 (D.C. Cir. 1990).

30. S. REP. NO. 106-110, at 48 (1999) ("Moreover, the Committee believes that the FACA exemption will reduce the potential rigidity, time, and expense of peer reviews.").

31. H.R. CONF. REP. NO. 95-722, at 16 (1977), reprinted in 1977 U.S.C.C.A.N. 3293, 3295.

32. President's Message to the Senate Returning S. 2577 Without Approval, 18 WEEKLY COMP. PRES. DOC. 1376, 1376-77 (Oct. 25, 1982) ("To require that the Board become a political entity, with representatives from various special interests, would completely undermine the use of scientific knowledge in EPA rulemaking.").

33. See Regulatory Improvement Act of 1997 (S. 981): Hearings Before the Senate Comm. on Governmental Affairs, 105th Cong. 116 (1997) (statement of Sally Katzen, Administrator, Office of Information & Regulatory Affairs, OMB) ("Peer review, like other salutary steps in the regulatory process, is not cost-free, either in terms of agency resources or in time. In fact, six-months to a year would be the norm, not the exception for each review.") [hereinafter 1997 Hearing].

34. See 1999 Hearing, supra note 18 (statement of David C. Vladeck, Director, Public Citizen Litigation Group) ("This [peer review] process will be extremely time-consuming—and tens of thousands of dollars per panel will be diverted from other important priorities. For agencies already cash-starved by budget reductions, having to bear this burden will sap their ability to do their work.").

35. See id. (statement of Patricia G. Kenworthy, Vice President, Governmental Affairs, National Environmental Trust); 1998 Hearing, supra note 18, at 139-40 (statement of Karen Florini, Senior Attorney, Environmental Defense Fund); 1997 Hearing, supra note 33, at 218-20 (statement of David G. Hawkins, Senior Attorney, Natural Resources Defense Council); id. at 268-70 (statement of David C. Vladeck, Director, Public Citizen Litigation Group).

36. See 1999 Hearing, supra note 18 (statement of David C. Vladeck, Director, Public Citizen Litigation Group) ("The potential for self-interested members of peer review panels to abuse their positions is made all the more severe because the panels are permitted to operate in secret, closed-door sessions, with no public oversight of the process.").

37. See id. ("The agency is required to respond, in writing, to significant peer review comments, giving peer review panel members enormous leverage in the rulemaking process. It is counter to basic democratic principles to empower a select few to be given a privileged place in the rulemaking process . . . ."); 1998 Hearing, supra note 18, at 139 (statement of Karen Florini, Senior Attorney, Environmental Defense Fund) ("The term 'peer review,' with its connotations of disinterested scientific objectivity, cannot fairly be applied to a statutorily mandated process in which the regulated industry can seek to undercut the factual basis for the rulemaking.").

38. See Thomas O. McGarity, Some Thoughts on "Deossifying" the Rulemaking Process, 41 DUKE L.J. 1385 (1992); Lars Noah, Doubts About Direct Final Rulemaking, 51 ADMIN. L. REV. 401, 403-05 (1999); Mark Seidenfeld, Bending the Rules: Flexible Regulation and Constraints on Agency Discretion, 51 ADMIN. L. REV. 429, 455-58 (1999).

39. See 1999 Hearing, supra note 18 (statement of Franklin E. Mirer, Director, Health & Safety Department, United Auto Workers); see also JASANOFF, supra note 1, at 171 (noting "the perception of public interest groups that regulatory peer review is simply a ploy to delay or avoid politically unpalatable decisions"); Burack, supra note 1, at 56 (suggesting that EPA's initial opposition to the establishment of the SAP may have arisen from concerns about "an unstated political objective" by Congress of slowing efforts to cancel the registrations of hazardous pesticides).

40. See JERRY L. MASHAW ET AL., ADMINISTRATIVE LAW 152 (4th ed. 1998) (explaining that some statutes "have clear substantive objectives, though they often employ ostensibly 'procedural' means"); Lars Noah, Sham Petitioning as a Threat to the Integrity of the Regulatory Process, 74 N.C.L. REV. 1, 19-20, 58 (1995) (explaining that expanded opportunities to participate in agency decisionmaking have facilitated anticompetitive manipulation of procedures, whether Congress intended that result or not); see also Cass R. Sunstein, Congress, Constitutional Moments, and the Cost-Benefit State, 48 STAN. L. REV. 247, 271 (1996) (applauding "technocratic" aspects of recent regulatory reform proposals, but noting the "reactionary" purposes underlying additional procedural burdens).

41. See 1998 Hearing, supra note 18, at 86 (statement of Lester M. Crawford, Director, Center for Food & Nutrition Policy, Georgetown University) ("[A] requirement that these issues be addressed in the regulatory review process will accelerate, not impede, the development of useful regulations. This is because these issues are practically always raised and are very often the difference in being given permission to finalize one or more steps in the regulatory approval process."); id. at 103-04 (statement of Robert W. Hahn, Resident Scholar, American Enterprise Institute, & Robert E. Litan, Director, Economic Studies, Brookings Institution):

There is a good case to be made that peer reviewed regulatory analyses actually will expedite the implementation of rules that are vital to protecting the environment, health and safety . . . . Rules that are supported by well-done regulatory analyses that have been peer reviewed are much more likely to withstand that opposition, either in Congress . . . or in the courts.

id. at 104 ("Indeed, well-done analyses actually can deter legal challenges because rules need only pass an 'arbitrary and capricious' test, which should be easy for an agency to meet if its regulatory analyses have passed peer review with flying colors."); Burack, supra note 1, at 48-49 (same).

42. See Philip M. Boffey, Scientists and Bureaucrats: A Clash of Cultures on FDA Advisory Panel, 191 SCIENCE 1244, 1244-46 (1976).

43. See 5 U.S.C. § 552(b)(5); Wolfe v. HHS, 839 F.2d 768, 774-76 (D.C. Cir. 1988) (en banc); Hunt v. U.S. Marine Corps., 935 F. Supp. 46, 51-52 (D.D.C. 1996); Russell L. Weaver & James T.R. Jones, The Deliberative Process Privilege, 54 MO. L. REV. 279, 309 (1989).

44. See Chemical Mfrs. Ass'n v. CPSC, 600 F. Supp. 114, 117-18 (D.D.C. 1984). As the court explained, "agency scientists may, for example, discuss hypotheses which have not matured, and can be effectively shared only with peers in regular and confidential communication." Id. at 118 (adding that "disclosure of an internal hypothesis or the data related to its formation followed by a demeaning attack on that hypothesis before the author has finally formed a conclusion would have an obvious chilling effect on the persons still in the process of forming the opinion").

45. See 1999 Hearing, supra note 18 (statement of David C. Vladeck, Director, Public Citizen Litigation Group).

46. See S. REP. NO. 106-110, at 47 & n.89 (1999); see also 7 U.S.C. § 136w(d)(1) (requiring adoption of conflict of interest rules for the SAP); Eliot Marshall, Science Advisers Need Advice, 245 SCIENCE 20, 20-22 (1989) (reporting on conflict of interest allegations made against two members of the SAP).

47. See Burack, supra note 1, at 52-54 (describing the fate of EPA's proposed rule on airborne radionuclides, explaining that, after the SAB severely criticized the scientific basis for the notice of proposed rule making, the "EPA had little choice but to withdraw" it); see also JASANOFF, supra note 1, at 82 ("Agencies subject to supervision by scientific experts will seldom feel confident enough to act unless their scientific assessments survive peer review.").

48. See Thomas O. McGarity, The Courts and the Ossification of Rulemaking: A Response to Professor Seidenfeld, 75 TEX. L. REV. 525, 535 (1997) ("Greater certainty always requires additional information and analysis, and more data gathering and analytical exercises take time and resources. Unfortunately, in most modern rulemaking contexts (especially those involving health and environmental protection), uncertainty is the order of the day.").

49. See JASANOFF, supra note 1, at 237 ("Just as peer reviewers generally cannot detect fraud or fabrication in scientific publication, so advisory committees will fail to uncover most instances of intentional misrepresentation in regulatory science . . . ."); id. at 238 (cautioning that "regulatory peer review remains only a partial prophylactic against methodological disputes").

50. See POWELL, supra note 19, at 139 ("Most of the scientists are serving essentially on a pro bono basis and have a very limited amount of time that they can sacrifice . . . . Much of the time served on expert panels is consumed by meetings and travel, thus limiting the time for independent analysis.").

51. See JASANOFF, supra note 1, at 234 ("Contrary to the presumptions of the technocratic model, negotiation among scientists alone is not sufficient to ensure the public and judicial acceptability of agency decisions based on regulatory science . . . ."); see also AFL-CIO v. Marshall, 617 F.2d 636, 652 (D.C. Cir. 1979) (suggesting that to demand an "explanation [of the scientific] basis of the agency's decision not only facilitates proper judicial review but also provides the opportunity for effective peer review, legislative oversight, and public education.").

52. See DAVID L. FAIGMAN, LEGAL ALCHEMY: THE USE AND MISUSE OF SCIENCE IN THE LAW 181 (1999) ("In practice, CASAC exists as much to provide Congress and EPA's critics with grounds on which to challenge proposed regulations as it does to give EPA guidance. In that capacity, CASAC serves as an integral part of the checks and balances existing between the branches of the federal government.").

53. See Burack, supra note 1, at 30 ("If peer review occurs early in the decisionmaking process, it can strengthen the record, narrow the controversial issues during public comment, and help in reaching sound decisions.").

54. See John S. Applegate, The Perils of Unreasonable Risk: Information, Regulatory Policy, and Toxics Substances Control, 91 COLUM. L. REV. 261 (1991); Sanford E. Gaines, Science, Politics, and the Management of Toxic Risks Through Law, 30 JURIMETRICS J. 271 (1990); Thomas O. McGarity, Substantive and Procedural Discretion in Administrative Resolution of Science Policy Questions; Regulating Carcinogens in EPA and OSHA, 67 GEO. L.J. 729 (1979); Wendy E. Wagner, The Science Charade in Toxic Risk Regulation, 95 COLUM. L. REV. 1613 (1995).

55. Motor Vehicle Mfrs. Ass'n v. State Farm Mut. Auto. Ins. Co., 463 U.S. 29, 52, 13 ELR 20672, 20678 (1983).

56. Pub. L. No. 104-182, § 103, 110 Stat. 1613, 1621 (codified at 42 U.S.C. § 300g-1(b)(3)(A)(i), (B)(v), ELR STAT. SDWA § 1412).

57. See JASANOFF, supra note 1, at 112-13.

58. See id. at 79, 98, 157; id. at 81 ("Many of the experimental methods and analytical techniques used in assessing health and environmental risks remain controversial and are not based on a broad scientific consensus."); POWELL, supra note 19, at 140 ("If scientific consensus is viewed as a prerequisite to regulatory action, then balance can be problematic because a well-balanced committee [i.e., one with broad disciplinary representation] is less likely to produce a meaningful consensus.").

59. See FAIGMAN, supra note 52, at 188-89, 200 ("The principal danger of this sort of expert assistance is that policy makers will abdicate their responsibilities to it . . . ."); NRC REPORT, supra note 13, at 159 ("A scientific review panel's critique of an agency's risk assessment should not be binding; that is, the agency should not be obliged to revise its risk assessment if the panel regards it as deficient. . . . The authority for [agencies'] actions must remain their own.").

60. International Harvester Co. v. Ruckelshaus, 478 F.2d 615, 652, 3 ELR 20133, 20150 (D.C. Cir. 1973) (Bazelon, C.J., concurring); see also David L. Bazelon, The Judiciary: What Role in Health Improvement?, 211 SCIENCE 792, 792 (1981) ("Full disclosure [by agencies of the bases for their decisions] will undoubtedly improve the quality of information by exposing weaknesses to peer review, legislative oversight, and public scrutiny."); David L. Bazelon, Risk and Responsibility, 205 SCIENCE 277, 279 (1979) (suggesting that "openness will promote peer review of both factual determinations and value judgments").

61. See Natural Resources Defense Council v. NRC, 547 F.2d 633, 653, 6 ELR 20615, full text of opinion is available from ELR Document Service, ELR Order No. C-1065 (D.C. Cir. 1976) (Bazelon, C.J.), rev'd sub nom. Vermont Yankee Nuclear Power Corp. v. Natural Resources Defense Council, 435 U.S. 519, 549, 8 ELR 20288, 20295 (1978); see also David L. Bazelon, Science and Uncertainty: A Jurist's View, 5 HARV. ENVTL. L. REV. 209 (1981).

62. See United States v. Nova Scotia Food Prods. Corp., 568 F.2d 240, 252 (2d Cir. 1977) (analogizing the opportunity to comment on scientific data during rulemaking to the editorial peer review process).

63. See POWELL, supra note 19, at 176 (describing one situation where a failure to undertake external peer review of a proposal caused an agency to go back and waste time undertaking analyses in response to objections raised in public comments); Burack, supra note 1, at 107-08 ("While not a surrogate for public comment, peer review may help narrow the issues that an agency will take seriously during the notice and comment period. In short, peer review affords agencies the opportunity to avoid mistakes.").

64. See 1999 Hearing, supra note 18 (statement of John D. Graham, Director, Harvard Center on Risk Analysis):

Today, any scientist has an opportunity to participate in either formal or informal rulemakings but the best scientists are unlikely to participate unless they are invited by the federal government to serve on a peer review panel or similar body. Some scientists currently serve as hired consultants to specific stakeholder groups but the testimony of stakeholder groups is not a substitute for independent, objective peer review.

65. See POWELL, supra note 19, at 137:

EPA's official science advisory panels and public comment under the [APA] provide means of independent scientific quality control, with the agency's science advisers widely regarded as playing the more important and often constructive role. The growing influence of the external scientific community through avenues such as the [SAB] represents one of the most significant changes in EPA's use of science over the past ten years.

66. See 1998 Hearing, supra note 18, at 71 (statement of Milton Russell, Senior Fellow, Joint Institute for Energy & Environment) (suggesting that "good analysis up front shortens subsequent delay").

67. See Corrosion Proof Fittings v. EPA, 947 F.2d 1201, 1212, 22 ELR 20037, 20041 (5th Cir. 1991); Portland Cement Ass'n v. Ruckelshaus, 486 F.2d 375, 392-94, 402, 3 ELR 20642, 20650-52, 20656 (D.C. Cir. 1973).

68. Baltimore Gas & Elec. Co. v. Natural Resources Defense Council, 462 U.S. 87, 103, 13 ELR 20544, 20548 (1983); see also Andrew D. Siegel, The Aftermath of Baltimore Gas & Electric Co. v. NRDC: A Broader Notion of Judicial Deference to Agency Expertise, 11 HARV. ENVTL. L. REV. 331 (1987).

69. See Chevron, U.S.A., Inc. v. Natural Resources Defense Council, 467 U.S. 837, 865-66, 14 ELR 20507, 20514 (1984); Sierra Club v. Costle, 657 F.2d 298, 408, 11 ELR 20455, full text of opinion is available from ELR Document Service, ELR Order No. C-1244 (D.C. Cir. 1981) ("we do not believe that Congress intended that the courts convert informal rulemaking into a rarified technocratic process, unaffected by political considerations").

70. See supra notes 60-61; see also Richard B. Stewart, Vermont Yankee and the Evolution of Administrative Procedure, 91 HARV. L. REV. 1805, 1812 n.34 (1978).

71. JASANOFF, supra note 1, at 241 ("Approval by an advisory committee . . . is equivalent to a judicial determination that the agency action is in compliance with applicable standards of substantive rationality."); see also id. at 248-49 (suggesting that favorable peer reviews should render substantive judicial review "redundant"); id. at 249 ("When a smoothly functioning advisory process is in place, . . . the spectacle of courts immersing themselves in technical data may gradually become as much an artifact as aggressive judicial overruling of congressional enactments became in the aftermath of the New Deal."); Frank B. Cross, Shattering the Fragile Case for Judicial Review of Rulemaking, 85 VA. L. REV. 1243, 1309 (1999) (arguing at length against substantive judicial review, and asking in that connection "why administrative law provides for judicial review dialogue but not for review by, say, the [NAS] or the National Academy of Engineering, which may have far more knowledge to offer in a dialogue").

72. See JASANOFF, supra note 1, at 62:

The arguments for close judicial scrutiny are likely to be strongest when one or more of the following conditions are met: the agency and its advisers disagree in their assessment of the evidence; the agency acts contrary to the recommendations of a peer-review panel; there is evidence of procedural impropriety in the review process.

73. See POWELL, supra note 19, at 139 ("If determining whether the data and analysis are adequate for regulatory decisionmaking is the problem, then peer review does not solve the problem.").

74. Asbestos Information Ass'n/North Am. v. OSHA, 727 F.2d 415, 421 n.15 (5th Cir. 1984); see also id. at 421 ("The extent to which the [OSHA's] supporting evidence has survived public and scientific scrutiny . . . will affect the weight given to it by an inexpert judiciary."); American Trucking Ass'n v. EPA, 175 F.3d 1027, 1059, 29 ELR 21071, 21082 (D.C. Cir. 1999) (Tatel, J., dissenting in part):

Bringing scientific methods to their evaluation of the [EPA]'s Criteria Document and Staff Paper, CASAC provides an objective justification for the pollution standards the Agency selects. . . . Other federal agencies with rulemaking responsibilities in technical fields also rely heavily on the recommendations, policy advice, and critical review that scientific advisory committees provide.

JASANOFF, supra note 1, at 58 ("Prior scrutiny by nonagency scientists might satisfy the courts that the agency's resolution of disputed technical evidence was backed by more than mere claims of administrative expertise."); id. at 249 (urging "courts to adopt a highly deferential posture with respect to a scientific record that has undergone thorough peer review").

75. See Flue-Cured Tobacco Coop. Stabilization Corp. v. EPA, 4 F. Supp. 2d 435, 28 ELR 21445 (M.D.N.C. 1998).

76. See id. at 459, 462-63, 28 ELR at 21456, 21457. For other examples of courts invalidating agency regulations notwithstanding generally favorable peer reviews of underlying risk assessments, see Gulf S. Insulation v. Consumer Prod. Safety Comm'n, 701 F.2d 1137, 1141-42, 1146-48, 13 ELR 20850, full text of opinion is available from ELR Document Service, ELR Order No. C-1318 (5th Cir. 1983) (urea formaldehyde foam insulation).

77. See American Trucking, 175 F.3d at 1035-36, 29 ELR at 21072 ("EPA cited the consensus of the [CASAC] that the standard should not be set below 0.08. That body gave no specific reasons for its recommendations, so the appeal to its authority . . . adds no enlightenment."). In fact, the CASAC did not reach any consensus on the significant questions embedded in the ozone and particulate matter (PM) NAAQS rulemaking. See Hearings Before the Subcomm. on Commercial & Administrative Law of the House Judiciary Comm., 105th Cong. 89-94 (1997) (statement of George T. Wolff, Chair, CASAC's Panel on Ozone & PM).

78. See Aeschliman v. NRC, 547 F.2d 622, 630-31, 6 ELR 20599, 20603 (D.C. Cir. 1976) (remanding licensing orders in part because of an ambiguity in an advisory committee's report addressing some of the technical issues), rev'd sub nom. Vermont Yankee Nuclear Power Corp. v. Natural Resources Defense Council, 435 U.S. 519, 556-57, 8 ELR 20288, 20297 (1978) (harshly criticizing this aspect of the lower court's decision).

79. Compare Universal Camera Corp. v. National Labor Relations Bd., 340 U.S. 474, 488 (1951) ("The substantiality of evidence must take into account whatever in the record fairly detracts from its weight.") with Federal Communications Comm'nv. Allentown Broad. Corp., 349 U.S. 358, 364 (1955) (rejecting suggestion that the Federal Communications Commission could overrule a hearing examiner's findings only if clearly erroneous); see also Allentown Mack Sales & Serv., Inc. v. National Labor Relations Bd., 118 S. Ct. 818, 824-25 (1998) (rejecting the National Labor Relations Board's decision to adopt an ALJ's finding that lacked support in the record).

80. See S. REP. NO. 105-188, at 74 (1998) (minority views of Sens. Lieberman, Akaka, Durbin, Toricelli, and Cleland):

We believe that peer review can play an important role in improving an agency's risk assessment prior to the notice of proposed rulemaking. But we do not believe that the peer reviewers' comments should be entitled to special deference by a court in determining whether an agency's actions are arbitrary or capricious.

81. See Burack, supra note 1, at 108 (suggesting that courts take into account an agency's failure to heed peer review objections, and hold action arbitrary and capricious if the agency fails to explain its reasons for ignoring those objections, but warning against the creation of a presumption of invalidity in such circumstances "for it would allow peer review panels to exercise a form of veto power").

82. See International Harvester Co. v. Ruckelshaus, 478 F.2d 615, 649, 3 ELR 20133, 20148-49 (D.C. Cir. 1973).

83. See American Trucking Ass'n v. EPA, 175 F.3d 1027, 1036, 29 ELR 21071, 21072 (D.C. Cir. 1999) ("The dissent stresses the undisputed eminence of CASAC's members, . . . but the question whether EPA acted pursuant to lawfully delegated authority is not a scientific one.").

84. See Hon. Patricia M. Wald, Environmental Postcards From the Edge: The Year That Was and the Year That Might Be, 26 ELR 10182, 10188 (Apr. 1996) (warning that a risk assessment mandate might lead courts to adopt a "checklist mentality"). So far, a few courts have resisted this temptation, at least with respect to an independent peer review mandate. See American Petroleum Inst. v. Costle, 665 F.2d 1176, 1188-89, 11 ELR 20916, 20921 (D.C. Cir. 1981) (holding that EPA should have requested SAB review of the proposed revisions to NAAQS for ozone and not just of the underlying draft criteria documents, but concluding that this procedural error was harmless); Lead Indus. Ass'n v. EPA, 647 F.2d 1130, 1171-72, 10 ELR 20643, 20665-66 (D.C. Cir. 1980) (rejecting procedural objection to EPA's failure to submit the criteria document for lead to CASAC's predecessor where the SAB had previously reviewed it).