The Role of Courts In Regulatory Negotiation-A Response to Judge Wald by Philip J. Harter* "Alternative means of dispute resolution" have clearly become an important topic of discussion and practice.' The debate recog- nizes that courts currently play, and will continue to play, a major role in resolving society's disagreements. Moreover, courts es- tablish the context that gives shape and life to the alternatives. The current consideration of "ADR" argues that other means may be more appropriate depending on the issues, the parties, the press of time and money, the state of precedent, and the rela- tionship among the parties. 2 The scope of alternative dispute res- olution has ranged from major, multi-million dollar corporate matters to pesky neighborhood disturbances that cause friction but are unlikely to end up in court short of the violence that may result precisely because they go unresolved. Not surprisingly, one of the ways in which various means of dis- pute resolution-particularly direct negotiations among the af- fected interests aided by a neutral mediator-have been used is to * A.B., Kenyon College; M.A., J.D., University of Michigan. 1. The use of alternative means of dispute resolution has become sufficiently significant to have spawned a large literature. The following, for example, are bibliographies of writing in the field: G. BINGHAM, B. VAUGHN & W. GLEASON, ENVIRONMENTAL CONFLICT RESOLU- TION (1981); U.S. DEPARTMENT OF JUSTICE, FEDERAL JUSTICE RESEARCH PROGRAM, DISPUTE RESOLUTION: TECHNIQUES AND APPLICATIONS (1985); F. SANDER, MEDIATION: A SELECT ANNOTATED BIBLIOGRAPHY (1984). 2. As the opening paragraph of a report prepared by an Ad Hoc Panel on Dispute Reso- lution and Public Policy consisting of senior, nationally prominent individuals represent- ing diverse perspectives says: Society cannot and should not rely exclusively on the courts for the resolution of disputes. Other mechanisms may be superior in a variety of controversies. They may be less expensive, faster, less intimidating, more sensitive to disputants' concerns, and more responsive to underlying problems. They may dispense better justice, result in less alienation, produce a feeling that a dispute was actually heard, and fulfill a need to retain control by not handing a dispute over to lawyers, judges, and the intricacies of the legal process. U.S. DEPARTMENT OF JUSTICE AND THE NATIONAL INSTITUTE FOR DISPUTE RESOLUTION, PATHS TO JUSTICE: MAJOR PUBLIC POLICY ISSUES OF DISPUTE RESOLUTION I (1984). 52 COLUMBIA JOURNAL OF ENVIRONMENTAL LAW [Vol. 11:51 resolve disputes over environmental questions. 3 The issues in- volved include complicated and controversial questions of indus- trial siting and use, planning of control and cleanup measures, allocation of resources, and environmental quality. 4 It has also been proposed, if not yet used fully, for cleaning up Superfund sites.5 Most environmental disputes are highly complex, involve sig- nificant technical questions of risk assessment and its manage- ment, are based on or establish issues of public policy, and affect many parties. As a result, they have all the earmarks of a typical rulemaking proceeding. It is, therefore, a relatively short step from the successful use of ADR in environmental matters to their application in developing regulations, albeit the techniques would need to be adapted to the peculiarities of the regulatory process. To that end, the Administrative Conference of the United States commissioned a study of the use of direct negotiations among the parties in interest in developing regulations, 6 them study formally recommended procedures for doing so. 7 Several agencies have used the process, and several more are well along in planning to use it for new regulations. The Federal Aviation Administration used ADR to revise its rules governing the amount of time a pilot can fly without taking a rest and the minimum periods between duty assignments.8 The original rule had proved enormously controversial, and the FAA had tried unsuccessfully several times before to update it. Its complexity had generated more than 1,000 pages of interpreta- tion. Significantly, especially given its history, once the rule was issued, 9 it was not challenged judicially. The Environmental Pro- tection Agency used regulatory negotiation to develop two rules, 3. G. BINGHAM, RESOLVING ENVIRONMENTAL DISPUTES: A DECADE OF EXPERIENCE (Con- servation Foundation, forthcoming); A. TALBOT, SETrLING THINGS (1983); L. BACOW & M. WHEELER, ENVIRONMENTAL DISPUTE RESOLUTION (1984); L. SUSSKIND, L. BACOW & M. WHEELER, RESOLVING ENVIRONMENTAL REGULATORY DISPUTES (1983); Cormick, The "The- ory " and Practice of Environmental Mediation, 2 ENVTL. PROF. 24 (1980). 4. See generally Bingham, supra note 3. 5. Anderson, Negotiation and Informal Agency Action: The Case of Superfund, 1985 DUKE L.J. 261 (1985). 6. The report was published as Harter, Negotiating Regulations: A Cure for Malaise, 71 GEO. LJ. I (1982). 7. Recommendation 82-4, 47 Fed. Reg. 30,701 (1982) (to be codified at 14 C.F.R. § 305.82-4). 8. Eisner, Regulatory Negotiation: A Real World Experience, 31 FED. B.J. 371 (1984). 9. 50 Fed. Reg. 29,306 (1985) (to be codified at 14 C.F.R. § 121, 135). Regulatory Negotiation one involving the nature of penalties imposed for truck engines that fail to comply with the dictates of the Clean Air Act' 0 and the other to establish the procedures for granting emergency exemp- tions from pesticide registration requirements." As with the FAA, the comments submitted in response to the Notices of Pro- posed Rulemaking were remarkably sparse, and no judicial chal- lenge has been filed against either rule. The Occupational Safety and Health Administration used a variant of the process to ad- dress the occupational exposure of benzene. That rulemaking had extended for nearly ten years and had reached the Supreme Court, 12 and thus had become a particularly difficult regulatory proceeding by the time the negotiations began. The discussions involved four industries and four unions. The parties made sig- nificant headway in developing a consensus, but for complex rea- sons largely beyond the negotiations themselves no final agreement was reached. Nonetheless, the participants viewed the process as worthwhile since it elucidated the needs and concerns of the other parties.' 3 That they made it that far in that rulemak- ing demonstrates the potential of the process. One of the major questions facing ADR in general, 14 and regu- latory negotiation in particular, is the relationship of the decisions reached in the alternative forum to the courts. Just as the courts and administrative agencies had to struggle-and are still strug- gling-to reach an accommodation, the various forms of ADR and the courts will likewise have to develop some sort of alloca- tion of responsibilities. Because courts play such a large role in 10. 50 Fed. Reg. 35,374 (1985) (to be codified at 40 C.F.R. § 86). 11. 50 Fed. Reg. 13,944 (1985) (to be codified at 40 C.F.R. § 166). 12. Indus. Union Dep't. v. Am. Petroleum Inst., 448 U.S. 607 (1980). 13. H. PERRITr, ANALYSIS OF FOUR NEGOTIATED RULEMAKING EFFORTS (1985) (a report to the Administrative Conference of the United States). Moreover, in considering the "success" or "failure" of the negotiations, it is important to note that OSHA missed its own deadline for issuing an NPRM on the standard by more than two years. One cannot divorce negotiations from the broader context in which they operate. 14. The issues involve such questions as whether parties must incur extra costs, some- times including attorneys' fees, if they "appeal" an arbitrator's decision to a court without substantially improving their position; whether the discussions during mediation may be held confidential (see, e.g., AMERICAN BAR ASSOCIATION, SPECIAL COMMITTEE ON DISPUTE RESOLUTION, CONFIDENTIALITY IN MEDIATION (1985)); what is the appropriate use of mas- ters to help settle complicated cases (see, e.g., 8 LEGAL TIMES 2 (October 14, 1985)); in- deed, whether it is appropriate to encourage settlements at all (Schoenbrod, Limits and Dangers of Environmental Mediation: A Review Essay, 58 N.Y.U. L. REV. 1453 (1983); Fiss, Against Settlement, 93 YALE L.J. 1073 (1984)). 1986] 54 COLUMBIA JOURNAL OF ENVIRONMENTAL LAW [Vol. 11:51 supervising or overseeing the regulatory process-surely that is a more accurate description than the term "partnership" some- times employed by the courts 5 - the need to establish that rela- tionship is particularly important for regulatory negotiation since it can have such a strong bearing on what happens before the agency itself. Parties are interested in negotiating a regulation, as opposed to relying on a more adversarial mode, and hybrid rulemaking is surely that, because it allows them to share in making the actual decision and to help develop a better, more informed rule. They would not incur the time, expense, and anguish of making hard choices, however, if their work were to be easily undone. Inap- propriate forms of judicial review could do just that. On the other hand, judicial review can also play a vitally important role not only for ensuring the integrity of the process but also for pro- viding positive incentives to make the rulemaking process work well. Undoubtedly the best approach to judicial review would be to obviate its need because everyone has been sufficiently satisfied with what happened before the agency. This has been our experi- ence so far with all of the rules developed by this process. But, alas, that surely will not continue forever. As Judge Patricia M. Wald points out in her article, Negotiation of Environmental Disputes: A New Role for the Courts?, 16 "negotiation typically does not elimi- nate court involvement altogether; instead it changes the nature and scope of the judicial role."' 17 While the full nature and scope ofjudicial review of negotiated regulations will have to evolve over time, the process can be aided by an analysis of the points likely to be raised. As in good negoti- ations, the needs of both the agencies and the courts, while imple- menting their respective duties, can be examined and an ac- commodation sought. A dialogue before the fact can help clarify the issues and lead to a fuller understanding on everyone's part. To that end, Judge Wald's analysis of the judicial oversight of ne- gotiated regulations is significantly helpful. Since much of her 15. Leventhal, Environmental Decisionmaking and the Role of the Courts, 122 U. PA. L. REV. 509, 554 (1974). But cf. Judge Bazelon's opinion in Environmental Defense Fund v. Ruck- elshaus, 439 F.2d 584, 598 (D.C. Cir. 1971). 16. Wald, Negotiation of Environmental Disputes: A New Role for the Courts?, 10 COLUM. J. ENVrL. L. 1 (1985). 17. id. at 17. 1986] Regulatory Negotiation analysis was in response to my earlier writing, I appreciate the opportunity to join the debate in response; indeed, it was in large measure her suggestion that I do so. OVERVIEW OF THE REGULATORY NEGOTIATION PROCESS While each regulatory negotiation, like any other rulemaking, is likely to be a little different from any other, the process as con- templated by the ACUS recommendations has several critical characteristics that will bear heavily on judicial review and the le- gitimacy of the entire process. As a setting for the analysis of ju- dicial review, an understanding of the process is essential; hence, it is briefly described here.' 8 Once the agency, either on its own initiative or in response to a request by an interested party,' 9 decides that it may be appropri- ate to use regulatory negotiation to develop a proposed rule, it would appoint a neutral convenor to conduct a feasibility analy- sis.20 The convenor would identify those interests that would be significantly affected by the regulation and the issues that need to be addressed by the rule, and in any negotiations, should they be undertaken. The convenor typically does this by starting with the agency and asking it both who would be affected and what issues are important. He or she would then go to the people named by the agency and ask them the same questions. That effort, coupled with independent research, will usually uncover at least the major players and issues. 18. For a more complete description of the process, see recommendation 82-4 of the Administrative Conference of the United States, I C.F.R. § 305.82-4 (1985); Harter, supra note 6; Harter, The Political Legitimacy and Judicial Review of Consensual Rules, 32 AM. U.L. REV. 471 (1983); and Eisner, supra note 8. 19. In at least one instance, the agency was initially hesitant to use regulatory negotia- tion to develop its rule, but a couple of its constituents who were usually on opposite sides of issues persuaded it to explore its feasibility. The agency did so and decided to go for- ward with the process. 20. As will be described later in the text, the convenor has two significant roles: one is to identify the parties that will be significantly affected by the rule and who should partici- pate in the negotiations if they are held. The second is to help those parties analyze whether it is in their interest to use direct negotiations for developing the proposed rule or whether some other process would benefit them more. In both cases, it is helpful to have someone who is neutral with respect to the subject matter of the rule and to any party. Otherwise the convenor's bias could influence analysis concerning whether a party should be included. Moreover, it could interfere with his or her working with a party in exploring what the negotiation process would likely entail, since a party may be reluctant to share concerns or viewpoints if it thought they might be repeated or used inappropriately. 56 COLUMBIA JOURNAL OF ENVIRONMENTAL LAW [Vol. 11:51 The convenor must also assess whether the parties believe it is in their overall best interest to attempt to negotiate a proposed rule or to rely on the normal process. Factors likely to influence that decision are: -whether the issues are mature and ripe for decision; it is diffi- cult to negotiate a solution to a problem that is only dimly un- derstood or that can be put off until another day. -whether there is a reasonable deadline for the negotiations; reaching agreement is a tough task since parties must make hard choices; hence, a deadline is decidedly helpful; 2 1 this can usually be provided by the agency's committing itself to devel- oping the rule through the traditional process should the nego- tiations prove unsuccessful after a specified period. -whether the outcome of the proceeding is genuinely in doubt; various parties may each have enough power-clout in one form or another-so that no one party, including the agency, can totally dominate what happens. 22 -whether any interest would have to compromise an issue that goes to the very core of its existence; no one compromises is- sues like that. -whether there are a diverse number of issues that are likely to be of different value to different interests. 23 -whether there are a limited number, roughly no more than 15-25, of distinct interests that would be substantially affected, and whether they are such that individuals can be selected to represent them.2 4 2 1. John Dunlop, one of the country's true experts on negotiation, points out that labor contracts are typically settled late in the night of the deadline precisely because it would be politically imprudent for any negotiator to return to the constituency well before the be- witching hour claiming that they got all that was possible; the difficult choices come at the end when you must decide among priorities and make the ultimate decision-are you bet- ter off with the deal or with the alternative to the agreement. Dunlop, The Negotiations Alternative in Dispute Resolution, 29 VILL. L. REV. 1421, 1436-1437 (1984). 22. It sometimes comes as a surprise that agencies often lack the ability to issue a rule at will. But, as the FAA example shows, agencies too are buffeted politically and other par- ties are able to derail a regulatory effort they oppose. One should not confuse the agency's authority to issue a regulation with its power to do so. Harter, Dispute Resolution and Administrative Law: The History, Needs, and Future of a Complex Relationship, 29 VILL. L. REV. 1393, 1418-1419 (1984). 23. While the "differential value" placed on different issues sounds exotic, it is not. Indeed, it is the most trivial of requirements. For example, the sale of a simple item re- sults from the customer valuing the product more than the money reflected in its price, whereas the store owner has reverse preferences. The point is, one is not likely to negoti- ate anything unless there is some opportunity for a trade which makes the person better off from his or her perspective. 24. It is, of course, true that even the most dedicated convenor will not be able to iden- tify, let alone talk to, all interested parties. His aim is to pick up those that would likely participate in a typical rulemaking, plus sufficient others so there is a significant likelihood 1986] Regulatory Negotiation If, after that analysis, it appears that regulatory negotiation would be appropriate, the convenor would recommend it to the agency. If the agency decides to go forward, it would appoint a senior agency official-generally the one charged with developing a draft rule within the agency-to be its representative to the ne- gotiations. 25 A neutral mediator, who is responsible for aiding the process of the negotiations, is usually helpful for any complex or controversial rule. Generally the person who served as the convenor will fill that role since he or she has developed an un- derstanding of the issues and a rapport with the parties. The me- diator has no power whatever to decide issues or impose a de- cision; his or her role is one of helping the negotiators themselves reach an agreement by helping to define the issues, generating creative options, carrying messages back and forth, being an "agent of reality" to each party, making it "safe" for the parties to talk directly, as well as facilitating the plenary discussions. The purpose of the negotiating committee is to develop a "con- sensus" on a proposed rule.26 It is surely preferable if the con- sensus extends to the actual language of the important provisions of the rule and its supporting preamble, since many policy choices surface in the drafting. "Consensus", as Judge Wald notes, can that the relevant issues will be raised and resolved during the negotiations. Thus, for example, it would be impossible to have the desires of all "consumers" fully represented, but relatively straightforward to have individuals who advocate the major consumer inter- ests, such as safety or low cost or diversity, participate. It may be that there would be too many such parties to negotiate successfully. In that case, the parties may be interested in forming caucuses for purposes of participation, with a designated individual representing that interest along with "elbow advisors" to provide particular vantage points. Some interests may choose not to participate but to be kept informed of the proceeding; usually, the best way to do that is for someone, generally the mediator if there is one, to ensure that the party most closely allied does keep that interest abreast of developments. 25. It may not be essential for the agency to participate directly in the negotiations, but it surely must have some intimate relationship to them if it is to be able to use the resulting agreement. If the agency fails to have such a connection it is highly likely that the agency will find one reason or another to disown the results. 26. It may be, of course, that the group is unable to reach a consensus on a rule but can agree on the basic provisions of one. The give and take of discussions may illuminate what the various parties "can live with." The group may then turn the issue over to the agency to "arbitrate" their differences by developing a rule within the region of acceptability. The FAA negotiations followed this route. On the other hand, the parties may feel that the provisions of a standard are so intertwined that it must be regarded as a whole rather than a collection of freestanding parts. In that case, the failure to agree would mean the group would report that fact and nothing more to the agency. All bets would then be off with respect to all parts of the regulation: no one would be "bound" in any way to anything said or done during the negotiations. The benzene negotiations adopted this approach. 58 COLUMBIA JOURNAL OF ENVIRONMENTAL LAW [Vol. 11:51 be a difficult term to define in the abstract 2 7 and a difficult status for a reviewing court to discern, whatever the definition. The parties can choose to define it however they see fit, 2 8 but its defi- nition can have a significant bearing on the willingness of parties to participate and perhaps on any subsequent judicial review. Un- less the parties agree otherwise, the definition that has proved the most workable is that "consensus" means that each interest repre- sented in the negotiations concurs in-or at least does not op- pose-the result. 29 That definition has several important consequences. First, no party can be outvoted; hence, each preserves whatever power it has. This has proved critical to most parties' willingness to par- ticipate in regulatory negotiations. Second, it forces the parties to come together to solve a mutual problem--developing the rule. They can no longer act as a group of disparate interests, each of which can dissent from a provision, go back home, and tell their constituents how tough they were. Rather, all partici- pants must decide whether, on the whole, they are better off ac- cepting the agreement or trusting their fate to another process. Third, it means that individual members of an interest group may dissent without destroying the consensus, so long as the interest as a whole concurs. The agency agrees to use the consensus proposal as the notice of proposed rulemaking unless it is outside the agency's authority or something is significantly wrong with it. That should happen rarely, however, since presumably the senior agency official has signed off on it, and he or she should have done his or her homework to ensure that no other relevant agency officials had problems with the proposal as it was developing and before its adoption. Following the convenor's analysis, the agency would begin the actual process by publishing a notice in the Federal Register30 and in substantive publications read by mere mortals who may be in- terested in the subject matter. The announcement would indicate the agency is intending to establish an advisory committee pursu- 27. Wald, supra note 16, at 20. 28. The "Organizational Protocols" that were adopted by EPA's Committee for stan- dards for worker protection from agricultural pesticides appear at the end of this article as an appendix. It provides the framework for the Committee's operations. 29. See Harter, supra note 6, at 92-97. 30. See, e.g., 50 Fed. Reg. 38,030 (1985). Regulatory Negotiation ant to the Federal Advisory Committee Act for purposes of nego- tiating the rule. It would describe the issues likely to be raised and the parties that had been identified by the convenor as being interested in participating. In that way, others can determine whether they would be affected and whether they should par- ticipate, or whether their interests are already adequately represented. If new parties do come forward, the convenor and the agency must decide whether to include them. Because the purpose of the negotiations is to ensure that the relevant issues get raised and resolved, it is far better to err on the side of inclusion. Thus, they should be included unless doing so would result in too many par- ties and it is not likely that a party seeking admission would raise significant new issues. Even in that case, it may be appropriate to form caucuses of allied interest for purposes of participation. 31 Even if a party does not join the committee, it can express its views during the committee's meetings or in response to the re- sulting Notice of Proposed Rulemaking. Thus, no one is shut out. The group needs to decide what information is required to make a responsible decision. One of the major values of this pro- cess is that the parties bring with them, or can otherwise obtain, insight and perspectives for developing a workable solution to the regulatory question. In addition, the negotiating group will typi- cally be furnished with technical information developed by the agency. 32 The Environmental Protection Agency has also estab- lished a "Resource Pool" the committee can draw on for pur- poses of the negotiations, such as for conducting short-term research or analysis.33 The groups operate under the Federal Advisory Committee Act which requires open meetings, although private caucuses can be held.34 Most experts on negotiating emphasize the need for closed sessions to encourage candid exchanges. The "sunshine" of the meetings has perhaps made things a bit more difficult at times but not impossibly so, and closing meetings in which public 31. See supra note 24. 32. See, e.g., 50 Fed. Reg. 42,789 (1985) (OSHA announced its intention to form a nego- tiating committee and its plans to furnish it with the relevant information). 33. See, e.g., 50 Fed. Reg. 38,030 (1985) (announcing EPA's intent to form a committee to negotiate its standards for protecting farm workers from agricultural pesticides and which describes the Resource Pool). 34. Nat'l Anti-Hunger Coalition v. Executive Comm. of the President's Private Sector Survey on Cost Control, 711 F.2d 1071 (D.C. Cir. 1983). 1986] 60 COLUMBIA JOURNAL OF ENVIRONMENTAL LAW [Vol. 11:51 policy decisions are being made has a cost of its own. Therefore, very little effort has been made to close any meeting. If the group is able to reach a consensus, it submits it to the agency as a recommendation for a Notice of Proposed Rulemak- ing (NPRM). The protocols developed for three negotiations have explicitly required each participant to sign the consensus and to ensure that that person has the authority to do so.a 5 In other instances, the approval has been more informal. The agreement has meant that the representative concurs in it, but it does not purport to bind all members of that representa- tive's constituency. Thus, for example, an individual member of the interest group-be it a trade association, a labor union, or an environmental organization-who is dissatisfied could file com- ments with the agency in response to the NPRM or challenge the rule in court. Indeed, the interest as a whole would even disown the earlier agreement and file adverse comments or challenge it judicially. As a practical matter, however, that should rarely hap- pen: if one party disaffects, others will too. Since it is likely that the outcome of the rule was genuinely in doubt when the process began, it probably still is. In that case, trashing the agreement will result in the loss of something that is known in return for a situation where the party could lose even more. Thus, a more desirable condition from everyone's perspective would be for the individual dissidents to challenge while the interest as a whole signals the decisionmaker-be it the agency, OMB, or a court- that it backs the agreement. In that way, the challenger can at- tempt to show that it would be affected differently than others or that its concerns were not raised. That will serve as an important check on those who do participate not to ignore the full range of interests. But, if the others lend their support, the decisionmaker will also know that the challenger does not speak more broadly. That in turn will help illuminate whether the challenger has real needs or whether it simply seeks more than it got at the table. THE NEEDS OF REVIEW Sloughing off a course full of detail,3 6 the Administrative Proce- dure Act (APA) provides that a reviewing court is to "decide all 35. See, e.g., Paragraph III of the Appendix. 36. See Levin, Federal Scope-of-Reviews Standards: A Preliminar, Restatement, 37 AD. L. REV. 95, 104-110 (1985). Regulatory Negotiation relevant questions of law" and "hold unlawful. . .agency ac- tion. .. found to be. . .arbitrary, capricious, an abuse of discre- tion, or otherwise not in accordance with law." 3 7 Thus, as explicated by Overton Park,38 which, although not involving the re- view of rulemaking, has become the foundation of the nature and scope of the judicial review of regulations, a court is to undertake a two step process: The court is first required to decide whether the [agency] acted within the scope of [its] authority. . . . This determination naturally begins with a delineation of the scope of the [agency's] authority and discretion. . . . Also involved in this initial inquiry is a determination of whether on the facts the [agency's] decision can reasonably be said to be within that range. Scrutiny of the facts does not end, however, with the determi- nation that the [agency] has acted within the scope of [its] stat- utory authority. Section 706(2)(A) requires a finding that the actual choice made was not "arbitrary, capricious, an abuse of discretion, or otherwise not in accordance with law." . . . To make this finding the court must consider whether the decision was based on a consideration of the relevant factors and whether there has been a clear error of judgment ... Although this inquiry into the facts is to be searching and care- ful, the ultimate standard of review is a narrow one. The court is not empowered to substitute its judgment for that of the agency. 39 Scope of Authority. Everyone involved in negotiating a regulation must recognize on a fundamental level that the proposal must be within the agency's authority granted by statute if the agency is to use it as the basis for a rule. That is simply a recognition that neither the agency nor certainly the negotiating group is sovereign. Rather, the agency must act within the bounds of discretion provided by Congress if its actions are to be a binding, official position. Thus, just as in the judicial review of a rule developed by another pro- cess, the court's first task must be to determine whether the rule is within the scope of the agency's authority. That process would be very much as it is customarily. Not only is this step required by the APA, it also serves several practical functions with respect to negotiating regulations. 37. 5 U.S.C. § 706(2)(A) (1982). 38. Citizens to Preserve Overton Park v. Volpe, 401 U.S. 402 (1971). 39. 401 U.S. 415-416. 1986] 62 COLUMBIA JOURNAL OF ENVIRONMENTAL LAw [Vol. 11:51 First, the statute, along with any implementing gloss placed on it by the agency and the courts, is the norm against which the negotiations are conducted. As in any negotiation, a party should keep its eye on what it could obtain without an agreement, and it should make a bargain only if it will be better off for doing so. In Fisher and Ury's popular term, the party needs to continually assess its "best alternative to a negotiated agreement" or BATNA. 40 The agency's scope of authority defines the contours of the parties' BATNA in regulatory negotiation. It is important for that reason alone for the courts to enforce that expectation. Were it otherwise, one party could pressure another into giving up something it won legislatively although that party had gained a reasonable expectation of achieving that result.4 1 Second, judicial enforcement of the limitations on an agency's authority also provides a deterrent against that ever present vil- lain of democratic politics-logrolling. Were it not for judicial oversight, it would surely not be inconceivable for the negotiating parties to make impermissible "deals" that are outside Congress's contemplation.4 2 A strikingly clear example of that-which has not arisen-would be if the parties to a negotiation over an occu- pational health standard agreed to a lower level of protection in return for an increase in wages for exposed employees. While that agreement may or may not make sense outside the OSHA context, and both sides have been argued strongly, the theory and nature of OSHA regulation is precisely contrary to such trade- offs. Thus, the judicial role here is to prevent impermissible deals or logrolling by requiring that the resulting rule be within the agency's scope of authority. Third, the judicial role also prevents sell-outs in which a party agrees to something short of what it should obtain under the gov- erning law. Perhaps the greatest concern here would be in the 40. R. FISHER & W. URY, GETTING TO YES 101 (1981). 41. To be sure, a party may still suffer some other sanction such as the inability to achieve the regulation in a timely fashion. That is, whereas one party may have an expec- tation of a particular type of decision, another party may have the power to delay it or to condition it on other provisions. Thus, the first party's BATNA may be a moderate regula- tion in a timely fashion, no regulation at all, or a stringent one after a protracted fight, all depending on how the pieces are reconciled. 42. Judge Wald raised this spectre. See Wald, supra note 16, at 22. Regulatory Negotiation agency's hasty acceptance of a negotiated deal in return for some other, intangible benefit.43 Thus, the judicial role in defining the scope of the agency's au- thority not only serves its usual role, but also provides additional incentives for the parties. While this role is an important and nec- essary one, the courts should be decidedly circumspect in exercis- ing this duty too aggressively. The court should, of course, be on the alert for logrolling and the like, but, that aside, it should be reluctant to second-guess a diverse group of interests-represent- ing a wide spectrum of concerns44 -that has concurred in the out- come, and whose individual members feel that on the whole the rule meets their needs. The process itself helps ensure that the proposal is within the agency's authority: if the agreement does not reflect a major pro- vision of the statute or is otherwise outside the agency's authority, presumably some party would be better off if it did and that party would likely insist on compliance. While that may sound a bit utopian, it is a measure of the extent to which the agreement is out of conformance with the statute. It may be that a bit of play is needed in the joints of a statute. The parties are typically better able to determine "what works" within the theory of the statute and hence what is the best way of achieving its overall goal. Or, a provision may be included in a statute to benefit a particular in- terest. If that interest does not insist on its full exercise as part of the agreement, the rule should still be accepted without that in- terest's being estopped from insisting upon the full application of the provision in future rulemakings. That they agreed indicates the interest achieved the protection sought in the statute45 which would otherwise come through a rigorous application of "scope of authority" review. 43. Although not in the regulatory negotiation context, the controversy over "sweet- heart" deals with respect to Superfund cleanups would be a prime example. Note, how- ever, that "sell-outs" should not occur if in fact all the major parties are at the table and concur. Surely neither the local municipalities nor environmentalists would have con- curred on those Superfund deals if they were indeed tainted. The process itself, therefore, is constructed to address the issue. That a court will review for scope-of-authority is thus an incentive to ensure that the proper parties are included. 44. See supra note 24 and accompanying text. 45. Such an issue arose in the benzene negotiations when corporate interests would have preferred to waive a provision that was intended to protect them. See Perritt, supra note 13. 1986] 64 COLUMBIA JOURNAL OF ENVIRONMENTAL LAW [Vol. 11:51 Thus, the court needs to define the-authority of the agency but provide a little leeway to accommodate practical interpretations and implementation. Arbitrary and Capricious. The Supreme Court directed the reviewing courts to make a "searching and careful" inquiry into the facts and to decide whether the decision is based on a consideration of the relevant factors.4 6 More recently, Judge Leventhal's characterization of the judicial process has become the frequently used shorthand. As Judge Wald began her article by pointing out, Leventhal advo- cated-and as a reviewing judge required-that "the agency. . .take[] a 'hard look' at the salient problems and en- gage[] in reasoned decision-making; the decision itself must come 'within a zone of reasonableness.' -47 Whichever term is used, the court is expected to carefully review the factual basis on which the agency based its decision and to carefully analyze whether the agency considered the relevant factors (or relied on an impermis- sible factor) and ultimately reached a decision within the zone of reasonableness under the statute. It should be recognized, however, that in developing this stan- dard of review, the courts adapted a more deferential, less intru- sive oversight to meet the needs of the times. 48 The former standard was simply not up to corralling agency action in the face of complex technical and social science questions or the pervasive influence of regulation. The courts responded with a new form of review by changing the interpretation placed on twenty-five year old statutory commands. This adaptation provides important in- centives to agencies to do a thorough job in developing the underlying facts and to exercise their discretion with care. It therefore serves as a powerful check against the arbitrary use of administrative power. In the same vein, it may be appropriate for courts to alter their means of ensuring that agency action is not arbitrary or capri- cious if the rule is the product of regulatory negotiation. Thus, the destination of Overton Park would remain the same, but the route taken in getting there can and should be changed. That is not because the standard should be any lower, but because the 46. Citizens to Preserve Overton Park v. Volpe, 401 U.S. at 416 (1971). 47. Wald, supra note 16, at 2. 48. DeLong, Informal Rulemahing and Integration of Law and Policy, 6 VA. L. REV. 257, 286 (1979). Regulatory Negotiation nature of review developed to supervise one process may inhibit the use of another that has other means of achieving the same end. Judicial review should therefore be tailored to ensure that those means are fulfilled. Without supervision, the agency may not have adequate incen- tives to delve into the relevant factors-statutory, factual, or polit- ical-that are involved in a rule. The "hard look" form ofjudicial oversight addresses that lack by providing the stimulus of the pain of reversal unless the agency does its homework. But, the directly affected parties are in a far better position than the agency or a reviewing court to determine what the "relevant factors" are and the weight to be accorded each; they can determine how much information is needed to resolve the outstanding factual issues; they are in the best position to determine the appropriate polit- ical trade-offs-all within the context of the agency's authority. 49 And again, if there were a "relevant factor" that would affect the outcome of a rule, it would be in the interest of some party to raise it for discussion and insist that it be adequately considered lest a consensus fail. That the rule reflects a consensus of the affected parties therefore goes a long way towards meeting the goals of Overton Park and ensures that the rule is neither arbitrary nor capricious. The issue then becomes whether the rule does indeed reflect that consensus. The very fact that we are discussing judicial re- view posits that at least someone is unhappy with it and hence that it does not reflect unanimity, which is, after all, one definition of consensus. What, therefore, is the court to do? The issue in contest may, of course, have been addressed di- rectly in the preamble to the rule so that the review would be much the same as in a typical rulemaking. This emphasizes that a negotiated rule will still have a preamble that explains the basis and purpose of the rule and the reasons the various choices were made as they were. If the issue is not explicitly developed, the court could still sat- isfy itself in large part by a procedural review of the process. The question is whether the parties to the consensus reflect suffi- 49. That said, it is highly unlikely that a rule produced through regulatory negotiation will be at an extreme of the agency's authority. Rather, its emphasis is more likely to be on finding practical solutions that accommodate the various interests' needs, and that will rarely be on the boundary. Thus, if it is important for an interest to press all the way, this process is probably not an appropriate vehicle. 1986] 66 COLUMBIA JOURNAL OF ENVIRONMENTAL LAW [Vol. 11:51 ciently diverse perspectives that the full range of issues would be raised in the negotiations. A beginning point would be to examine the process by which the negotiating committee was established. That would surely entail an examination of the issues raised in the Notice of Intent and its list of interests that would be represented on the commit- tee. Thus, the court would determine whether the notice was suf- ficient to alert an interest as to whether it might be affected and whether someone with allied needs would be present so it could be relatively comfortable that its issues of concern would be raised. Indeed, one of the major advantages of negotiated rulemaking is that by means of the convening process and the notice of in- tent, a concerted effort is made to identify the affected parties and to encourage them to participate directly in the development of a rule. Regulatory negotiation is therefore likely to result in broader participation than the traditional process. Thus, more is- sues are likely to be confronted. To ensure that a party's concerns are considered by the negoti- ating committee in the first instance, or by the agency in response to comments filed on the NPRM, 50 the court should not reward a party for sitting out the process and challenging only the final rule. Therefore, if the notice and the preliminary outreach 5 l are sufficient to alert those likely to be affected, it would be appropri- ate for the court to provide an incentive to come forward and par- ticipate directly in the discussions by requiring a reasonable explanation of why the challenger believes its interests-those relevant factors-may not have been raised and its failure to come forward. Thus, the court should require a reasonable explanation as to why the petitioner did not "exhaust its administrative reme- dies" 5 2 by participating before the agency. Parts of an earlier article have been construed as my arguing that a party should be "bound" by an agreement so long as its 50. The agency would likely seek the views of the negotiating committee on the com- ments so that it could take them into account when writing the final preamble and revising the rule to reflect meritorious comments. 51. It may be appropriate for an agency to include a list of the parties that were con- tacted by the convenor, along with a brief description of their respective interests, in the record of a negotiated rule. That would aid the court in determining whether any interest was excluded from the discussions that would be affected by the rule. 52. Vermont Yankee Nuclear Power Corp. v. Natural Resources Defense Council, 435 U.S. 519 (1978), requires that a party challenging a rule must raise the issue first before the agency. Regulatory Negotiation interest was represented at the table. 53 After having seen these things in action and after reflecting on Judge Wald's article, I am prepared to confess error on that score. "Interests" are too slip- pery for that, and determining just how far a consensus goes can be as elusive as the term itself.54 It now seems to me that a failure to provide an adequate explanation should not be regarded as a fatal preclusion of review, but the petitioner should be put to a relatively higher burden of knocking down the agreement that others forged. That is certainly close, if not identical, to current standards. If such an explanation is provided, the court must still deter- mine whether the decision is within the "zone of reasonableness" or whether it reflects an adequate consideration of the "relevant factors." In determining whether the important factors were con- sidered and resolved appropriately, the court would want to look at the range of interests that were represented and that concurred in the result by signing the document. 55 The petitioner should be required to make some showing that its specific interest was some- how not dealt with adequately, and hence that a relevant factor was not considered, or that the result was a clear error of collec- tive judgment. Knowing that a single member of a broader constituency is challenging a rule should help the court focus on the precise issue in controversy. Either the petitioner feels it would be affected dif- ferently than the others in its constituency, or it disagrees with the balance that was struck. If the petitioner was affected differently than the others at the table, then the court must decide whether, all things considered, the decision is still within that zone of rea- sonableness. That decision would turn on the interplay of those interests that were present and the explanation of the decision in the preamble. If, on the other hand, the challenger's interest was represented, so that it is disagreeing with the resolution of the 53. Wald, supra note 16, at 20-22 (construing Harter, supra note 6, at 102-107). 54. A frequent criticism of courts is that they lack "feel" of a problem and hence set out requirements that are not practical. It is therefore altogether fitting and proper for a judge to point out the impracticalities of just how a court might go about implementing the interest representation model. Such is the difficulty of writing about something never experienced. 55. If the process were more informal, the court would clearly have less assurance of a "consensus" and hence need some independent evidence of it if the court were to grant any deference to the rule as the product of consensus. That could in part by provided by the support shown for the rule by those who did participate. 1986] 68 COLUMBIA JOURNAL OF ENVIRONMENTAL LAw [Vol. 11:51 issues, the court should recognize that negotiated rulemaking is like legislation-a method for reaching a political accommoda- tion 56-and reject the challenge. The Thorny Problems. In concluding her analysis of the judicial review of negotiated rules, Judge Wald emphasizes three thorny problems: For any particular agency action, interest group pluralists have to determine which interest groups should be represented before the agency, what kind of participation is appropriate, and what sort of procedural rules will govern the ironing out of a consensus among those conflicting group interests. 57 These are surely the major issues, plus perhaps a fourth: how much information should the group be required to develop in reaching a decision. The hybrid process thrives on paper, for the quite legitimate and important reason that it is a means of con- trolling discretion that is needed precisely because parties do not participate directly. But, much of the analysis of the hybrid pro- cedure is largely a surrogate for direct participation; thus, it should not be necessary if the parties are there. Since, however, judicial review has come to focus on the paper record, so here too we may need to adapt. We will, as Judge Wald suggests, have to evolve a response to each concern. In the meantime, there are some preliminary answers or guideposts. First, to meet the dictates of Overton Park, sufficiently diverse interests need to be represented to ensure that the relevant fac- tors will be raised and considered. On a practical level, the con- vening process needs to be one that will help identify the parties that will be significantly affected and will also provide notice to those who were not picked up by direct contacts. Thus, the an- swer to the question of which groups should be represented is that all who represent parties that would be significantly affected by the rule are welcome. 58 The process, within the confines of available slots, is one of self selection; if too many wish to partici- pate, it is usually accomplished through caucuses or through en- couraging participation at meetings without formal membership on the committee. 56. Scalia, Rulemaking as Politics, Chairman's Message, 34 Ao. L. REV. V (1982). 57. Wald, supra note 16, at 24. 58. See Paragraph I of the Appendix for an example of how one negotiating committee defined those eligible for membership. Regulatory Negotiation Second, each interest should have equal access to the table. It may be that a coalition of allied interests will need to be formed and that the caucuses will need to select designated representa- tives. The individual representatives may be supported by "team" members who have particular viewpoints or an expertise that can be drawn upon as needed. Third, a workable definition of "consensus" is that each inter- est represented at the table concurs in the result and that each representative signs the document. Each party then has the power to insist on sufficient information to make an informed choice as balanced against all other demands on resources and on the full consideration of those factors the party thinks are relevant. Summary Overall, courts should be sensitive to the fact that there may be other ways than developing a massive record and extensive paper explanations of agency actions to determine whether the result is within the zone of reasonableness under the statute. If a range of interests is broad enough so that one may be confident that the relevant issues are raised and thrashed out, the court should at least require a challenger to come up with a good explanation as to why it "ain't necessarily so." 59 In that way, the court will trans- mit the proper incentives to all parties to develop a practical regu- lation within the confines of the statute. Finally, courts should remember the final dictate of the Overton Park test: "[T]he court is not empowered to substitute its judg- ment for that of the agency." It should be even more so to the extent it is a collective judgment of interests spanning the issue. 59. Wald, supra note 16, at 22. 1986] 70 COLUMBIA JOURNAL OF ENVIRONMENTAL LAW [Vol. 11:51 APPENDIX* ENVIRONMENTAL PROTECTION AGENCY STANDARDS FOR WORKER PROTECTION FROM AGRICULTURAL PESTICIDES NEGOTIATING COMMIITEE ORGANIZATIONAL PROTOCOLS I. PARTICIPANTS Any interest that would be substantially affected by an EPA standard governing worker exposure to pesticides may be repre- sented on the negotiating committee. Parties may group together into caucuses to represent allied interests. After negotiations have begun, additional parties may join the Committee only with the concurrence of the Committee and if within the number of parties permitted under the Committee's charter. A senior representative of each party or their alternate must attend each full negotiating session. The designated repre- sentative may be accompanied by such other individuals as the representative believes is appropriate to represent his/her inter- est, but only the designated representative will have the privilege of the floor during the negotiations without Committee approval. Participants are expected to represent the concerns and posi- tions of the interests of their constituents and to ensure that any agreement developed by the Committee is acceptable to the or- ganization from which the participant comes. II. DECISIONMAKING The Committee will operate by consensus. Committee deci- sions will be made only with the concurrence of all interests represented. Smaller working groups may be formed to address specific is- sues and to make recommendations to the Committee. Working groups are not authorized to make decisions for the Committee as a whole. Their meetings will be held between the full sessions. Working group meetings will be scheduled in the same location and time whenever possible. All committee members will be noti- fied of all working group meetings. * Oct. 17, 1985 Regulatory Negotiation Meeting agendas will also be developed by consensus. The Committee will determine if information requests are reasonable and relevant. If a deadlock or impasse is declared by any party, the facilitator will be available to help the deadlocked parties to try to resolve the impasse. A caucus can be declared by any participant at any time. III. AGREEMENT EPA is committed to using any consensus reached in these ne- gotiations as the basis of a proposed rulemaking. The agreement reached will take the form of a written statement that will be signed by all parties. IV. SAFEGUARDS FOR THE PARTIES All participants must act in good faith in all aspects of these negotiations. Specific offers, positions, or statements made dur- ing the negotiations may not be used by other parties in any other forum or as a basis for future litigaiton. Parties may withdraw from the negotiations at any time without prejudice. Personal at- tacks and prejudiced statements will not be tolerated. The pro- ceedings will not be electronically recorded. No discussions characterizing the position of any party will be held with the press until the conclusion of the negotiations, even if a party withdraws. All parties agree not to withhold relevant information. If a party believes it cannot or should not release such information, it will provide the substance of the information in some form (such as by aggregating data, by deleting non-relevant confidential in- formation, or by providing summaries) or a general description of it and the reason for not providing it directly. All parties agree not to divulge information shared by others in confidence. V. OPEN PROCESS The negotiations will be conducted under the Federal Advisory Committee Act (FACA). Negotiating sessions will be announced in the Federal Register prior to the meeting and, consistent with FACA requirements, will be open to the public. Minutes of Com- mittee meetings will be kept and made available to the public on request. 1986] 72 COLUMBIA JOURNAL OF ENVIRONMENTAL LAw [Vol. 11:51 VI. SCHEDULE Negotiating sessions will initially be held once each month. The first negotiating session is scheduled for November 4, 1985. Unless otherwise agreed upon, a deadline of four months for the negotiations will be established. The location of the meetings will be decided by the group. The participants may discontinue negotiations at any time if they do not appear productive. VII. FACILITATOR A neutral facilitator retained by EPA will work with all the par- ties to ensure that the process runs smoothly. -