Public Participation in Risk Regulation

During the years that have intervened since the consumer/environmental decade of 1967-1977, the basic principle that the "public" ought to play a role in regulatory decisions involving health and environmental risks has not been seriously questioned. That a social consensus has solidified around this principle is in many ways remarkable, given the prevailing model of administrative decision making that dominated thinking about that subject from the New Deal to the mid-1960s.¹ The New Deal institution-builders placed great stock in expertise as the answer to social problems, and the great repositories of expertise were the brand new agencies that were created during the heyday of the New Deal.

Under the New Deal model of administrative decision making, agency experts would write standards, issue permits and generally implement broad statutory grants of power to regulate "in the public interest". The agency was the public's representative, and it could be trusted to reign in corporations' uninhibited pursuit of their own private interests. Perhaps the best symbol of this view of regulation still stands outside the offices of the Federal Trade Commission in Washington, D.C. in the form of a statute in which a determined and very muscular man struggles to control a beautiful, but high-strung stallion.

Over the years, however, the New Deal model encountered significant impediments. First, after the "happy hot dogs" who founded the agencies moved on to more lucrative or more powerful positions, top-flight experts were generally unwilling to work in regulatory agencies. The United States bureaucracy never acquired the status of the civil service in some European countries. Regulatory agencies simply could not compete in monetary rewards or prestige with universities and private industry. More importantly, it soon became apparent, as the New Deal model was expanded beyond the traditional areas of regulated utilities and antitrust policing, that the experts did not have all of the answers. Many regulatory problems were too complex even for the experts, and most had political components that simply defied rational analysis. Finally, critical evaluations of the agencies in the mid-sixties by Congressional committees and self-proclaimed public interest groups like "Nader's Raiders" revealed that far from hewing to the original model, the regulatory agencies had evolved during a couple of decades of benign neglect into "captive" entities that were for the most part playing the tunes called by the regulated industries. It was as if the horse and the man struck a bargain by which the horse would allow the man to ride on its back so long as they both went where the horse wanted to go.

Under the considerable pressures of the late sixties and early seventies, the New Deal model rapidly gave way to the "interest representation" model, in which regulatory decisions are arrived at through the interplay of contending interest groups. The regulatory process in this model thus becomes a surrogate for the political process itself. Congress, which for a number of reasons cannot bring itself to resolve definitively difficult political clashes, delegates controversial issues to the regulatory agencies for resolution by the erstwhile "experts", knowing full well that expertise is incapable of providing final answers. The agencies which inherit the political debate then attempt to resolve individual cases by finding a middle ground among the contending forces. Public participation, therefore, becomes an essential ingredient of the policy making process.²

At about the same time that the courts were discovering and implementing the interest representation model, Congress was becoming very active in enacting substantive reform legislation aimed at health and environmental problems. Once again, on the surface of things, it appeared that the best way to establish and implement health and environmental standards was to delegate the standard-setting function to experts. Only experts could understand the complex scientific and economic considerations that were necessarily a part of the decision making process. But any attempts to invoke the New Deal model for solving the newly found health and environmental problems quickly ran up against the experience that the reformers had encountered with the old-line New Deal agencies. Once burned, they were unwilling to place their trust in experts to decide sensitive questions involving public health and the environment.

The debate over the proper role for expertise in risk regulation has never been fully resolved. Both environmentalists and industry frequently invoke the wisdom of experts when they perceive that the judgment of the experts will help advance their views of the issue. And both denigrate expertise when it does not point toward substantive ends that they prefer. Yet many, if not most, important health and environmental questions are in fact not resolvable by the experts. The available information and the state of the scientific art is often so poor that the experts can at best hazard highly uncertain educated guesses. The tough issues are by default resolved on the basis of outcome-oriented policy considerations.³ Sometimes this explicit invocation of policy considerations to resolve highly technical regulatory issues is done explicitly for all the world to see. All too often, however, it is done behind closed doors.

The demands for public participation in risk regulation, therefore, stems from a distrust of experts, a corresponding distrust of regulatory decision makers, and a conviction that most important risk regulations issues are not resolvable solely by reference to expertise. Yet as the "interest representation" model has continued to evolve in the context of risk regulation, it has become increasingly clear that the "public" is not a monolithic entity. This should come as no surprise; we live in a pluralistic society in which one person's gain is another's loss. It has also become apparent that there are several models available for structuring public participation into the regulatory decision making process.

The purpose of this paper is to identify some of the different "publics" that claim rights of participation in risk regulation and to explore briefly the legitimacy of those claims. The paper will also suggest several models for public participation in risk regulation and examine the advantages and disadvantages of each. The paper accepts the premise that public participation is necessary and desirable, but it suggests that, in the more complicated world of the 1990's, serious thought should be given to the appropriate role for different publics in different regulatory proceedings.

The early debates about public participation in the administrative process focused on the threshold question whether any entity other than the regulated industry and the agency (and in some contexts the union of the regulated company) had "standing" to participate in the regulatory process. The debate took place in relatively formal administrative contexts, such as Atomic Energy Commission licensing proceedings and Food and Drug Administration formal rule making hearings. Since the proceedings closely resembled judicial proceedings, the agencies and the commentators easily adopted the "standing" metaphor that had already received considerable judicial gloss. In time, the more perceptive observers pointed out that the judicial wisdom on "standing" before courts was generally irrelevant to the question of who should participate in administrative agency decision making, because the separation of powers considerations that informed judicial analysis of the problem were inapplicable to a purely administrative action.⁴ Nevertheless, some of the policy considerations that motivate courts to deny standing to parties that lack a sufficient interest in a judicial proceeding are relevant to the question whether formal administrative hearings should be open to everyone, and the agencies tended to focus on the "interest" of the person seeking to participate. Paralleling a similar judicial debate, and with some judicial prodding,⁵ the agencies finally came around to the view that someone who purported to speak for the "public interest" should have an opportunity to participate. This view had to overcome, however, considerable resistance on the part of agencies and the regulated interests, who often insisted that the agency itself was an adequate guardian of the public interest.

Two things happened in the consumer/environmental decade to change permanently the debate over public participation in agency decision making. First, the courts expanded rather dramatically the range of interests that could seek judicial relief. Aesthetic and environmental interests were welcome in court along with the economic interests that had always dominated courtroom adjudication. Second, the nature of the administrative decision making process changed dramatically, as agencies (with both judicial prodding and congressional insistence) began to make regulatory policy explicitly through rule making and other less formal policy making tools. By the end of the 1970s it was clear that representatives of "public interest" groups could present their views of the public interest in most regulatory forums that addressed risk regulation. But with the corresponding proliferation of "public interest" groups, it became increasingly less clear what the "public interest" in a particular issue was and who appropriately spoke for that interest. By the end of the 1980s, it is becoming increasingly apparent that the public has a multiplicity of interests and that no single group can cl aim lay exclusive claim to the role of spokesperson.

Perhaps the most universally accepted representative of the "public" is an individual who is merely interested in the subject matter of the administrative decision making process and desires to offer views or proof relevant to that proceeding. Very few individuals have such an interest in governmental decision making in the abstract to justify the expenditure of time and money that often attends such participation, but some individuals can be so acutely affected, either economically, morally or aesthetically, by the predictable outcome of the process that they are willing to invest their own time or money in an effort to affect the result.

Other individuals have a professional interest in the technical aspects of particular administrative proceedings and are willing to spend time and/or resources participating in the administrative process to some degree to share their perspectives with the agency. For example, scientists conducting research in areas related to the issues raised in a standard-setting proceeding might desire to participate as a professional matter. At the very least, such professionals usually want the agency to ask their advice about matters upon which they claim expertise. For the most part, these participants do so out of a sense of obligation to share their talents with society or perhaps out of a desire to have some influence on momentous decisions.

It is not at all unusual for a group of individuals who share a common interest in the subject matter of an administrative proceeding to form a loosely coordinated group solely for the purpose of affecting its outcome. For example, a group of residents near a proposed hazardous waste incinerator may have few interests in common but nevertheless band together to urge the relevant agency to refuse the permit. When the stimulus disappears, the group typically dissolves, although there are numerous examples of such ad hoc groups achieving some degree of permanence. The tendency of such groups is to focus very intensely upon a particular decision or issue to the exclusion of similar decisions in other contexts. National public interest groups find it notoriously difficult to mobilize such grass roots organizations into an effective constituency, because they lack continuity and tend to splinter easily once the focus is broadened. They are generally very effective for a brief period of time, however, in raising the public consciousness about the particular issues that are of concern to them. For decisions involving risks and benefits that fall primarily upon local communities, such groups can very persuasively claim a legitimacy that national public interest groups often lack.

National public interest groups are often created to deal with large issues of nationwide importance or to address local issues that recur throughout the country. Although "fly-by-night" national public interest groups are not unheard of, the national groups tend to have a higher degree of stability than local groups, and they generally have more resources available to them, often coming from large national foundations. Although national groups tend to focus on broad national issues, they can on occasion participate vigorously in local decisions, where they must occasionally rebut suggestions that they are outsiders and troublemakers.

Not long after such national public interest groups as the Sierra Club, the Natural Resources Defense Council, and the Environmental Defense Fund began to participate actively in environmental decision making at EPA, business-oriented public interest groups began springing up to advocate less stringent regulation and to intervene on the part of regulated industries. Although often accused of being mere industry fronts, some of these groups no doubt genuinely desire to advance an alternative definition of the "public interest" that views unfettered free enterprise as a worthwhile social goal. These newer groups have attempted to persuade regulatory decision makers that pro-consumer and pro-environmental groups do not have an exclusive claim to be representatives of the public interest, and they seem to take particular pleasure in invoking health and environmental statutes to impede governmental action aimed at protecting health and the environment.⁶

D. Regulated Industry and Trade Associations

It perhaps goes without saying that the regulated industries themselves and their representative trade associations desire to participate in the administrative decision making process insofar as it affects their economic interests. Indeed, the Administrative Procedure Act and numerous agency procedural regulations were designed largely with the prospect of participation by the regulatee in mind.

Labor unions often play an active role in administrative proceedings that protect wo rkers, even nonunionized workers. Thus, it is not surprising that labor unions play a prominent role in decision making in the Occupational Safety and Health Administration. When OSHA regulates toxic substances in the workplace, the labor unions and the environmental groups often have common interests and present a united front. On the other hand, when the issue is perceived to be jobs versus wildlife or aesthetic values, labor unions often pragmatically side with employers in resisting costly environmental restrictions that might cause companies to close up shop.

F. Competitors

One seldom mentioned representative of the "public interest" is the competitor of the regulated entity or industry. A paper manufacturing plant that has expended millions of dollars installing new pollution control technology has an obvious economic interest in ensuring that its competitor is required to install the same technology. The railroad industry, which must comply with numerous environmental controls, does not want the government building dams that facilitate barge traffic.⁷ Thus, railroad companies are sometimes willing to participate with environmental and farmer groups in administrative proceedings and even assume a considerable share of the financial burden of public participation in such proceedings.

Public participation in some form has become an accepted norm of administrative decision making. It tends to bear out in practice the aspiration that ours is a government by the people. No individual is so insignificant that he or she cannot demand an opportunity to be heard before the government acts to his or her detriment. Public participation is an essential component of good government. Members of the public can make relevant facts available to decision makers and allow them to view issues from different perspectives. If the public officials are attentive, better decisions should result. Public participation also has a desirable tendency to open up governmental decision making to public scrutiny. The public in general becomes better informed about how decisions are made in the real world and about what factors actually motivate bureaucratic policy makers. Finally, broad public participation tends to produce a better record for judicial review. For all of these reasons, public participation tends to legitimate administrative decision making.

Public participation is not, however, an unalloyed good. Allowing individuals and groups to challenge an agency's decision making forces the agency to expend resources defending itself that might otherwise be spent pursuing its statutory mission. Similarly, it requires regulatees in the private sector to expend additional resources beyond those required to deal with the agency itself. Public participation is time consuming. For a harried bureaucrat or a company executive whose interest payments mount while a permit application is pending, the time consumed in public participation is perceived as delay. Indeed, potential participants can use the threat of delay to extract substantive concessions from the agency and regulators. Although it is unfair to characterize all time consumed in public participation as "delay",⁸ public participation undeniably slows down the governmental wheels.

To the extent that effective public participation requires a two-way exchange of information, it can be inconsistent with a regulatee's legitimate interest in protecting valuable trade secrets and financial information. This clash of interests becomes particularly acute in the context of health and environmental regulation, where the identity of a chemical and the content of any health and safety data concerning that chemical are critical to effective participation by environmental groups, but can also yield undeserved commercial advantages to competitors. For example, the identity of the segment of DNA inserted into that of a plant genetically engineered to secrete an insecticide is of obvious interest to both competitors and to persons concerned with the environmental effects of widespread use of the plant. This clash of interests suggests that not all participants should be treated equally in all aspects of regulatory decision making.

Fortunately, a sufficiently large variety of vehicles for channeling public participation into the decision making process exists to facilitate a fair and efficient balance of the advantages of participation against its disadvantages. The remainder of this paper will examine several models of public participation and suggest some factors that might guide a policy maker's choice among them.

As might be expected in a large and diverse federal system, public participation in administrative decision making comes in many shapes and sizes, ranging from a virtual delegation of decision making power to specified members of th e public to complete exclusion. While the following discussion of six discrete mode ls of public participation should convey a sense of the wide variety of participation modes, it nevertheless masks to some extent the rich diversity that exists in the real world. Indeed, the models might more properly be viewed as locations along a spectrum of possible variations. The purpose of the effort is not to see how many discrete models can be identified; it is to convey to policy makers a feel for the choices that exist and the considerations that might inform those choices.

At one extreme on the spectrum of public participation paradigms is the traditional exclusionary model. Under this model the regulated industry and the agency fought out regulatory battles among themselves in the context of a fairly formal regulatory process. If, as was often the case, they made their peace on terms acceptable to both entities, the matter was closed without further inquiry into whether the agreed-upon solution was in the public interest. The assumption was that the agency was the exclusive guardian of the public interest, and any self-proclaimed representative of the public interest was an officious intermeddler. The only members of the public who were invited to participate in the process were the regulated industry and perhaps unions and competitors, depending on the issues.⁹

The exclusionary model has been thoroughly rejected in nearly all regulatory agencies. Still the model has appeal in some contexts. For example, although agencies could not lawfully attempt to exclude interested members of the public who would be adversely affected by the outcomes of formal adjudicatory proceedings, they might legitimately attempt to limit that participation to exclude irrelevant evidence and arguments and to prevent duplicative cross-examination by public interest intervenors.¹⁰ Even in less formal contexts, agencies might exclude especially confrontational forms of participation in the interest of decorum. For the most part, however, the exclusionary model does not fit well into currently accepted norms of public participation in informal decision making structures. There is very little lost by allowing an interested (even an officious) person to write down his views and insist that they be read by some responsible official. Similarly, it is rarely burdensome for an agency to allow interested individuals or representives of groups to testify briefly at informal "vent your spleen" hearings. Since most health and environmental agencies make important, binding decisions through informal rule making, this suggests that the exclusionary model is by-and-large inappropriate for risk-oriented decision making.

There is one small area, however, in which the exclusionary model may have an appropriate role, even in informal administrative proceedings. As previously mentioned, health and environmental decision makers often need to examine and evaluate commercially sensitive information. Members of the public desiring to participate in the decision making process will understandably want to see the information upon which the agency relies, and they usually consider this an essential element of effective public participation. The problem arises from the fact that competitors are also members of the public, who have an appropriate role to play but no legitimate interest in using the information to advance their own economic interests. Although procedural tools, such as protective orders, are sometimes available in more formal contexts, there is no easy way other than exclusion to prevent the information from coming to the attention of a competitor who might use it to the disadvantage of the regulatee. In this relatively limited situation, the policy in favor of protecting the regulatee's investment in producing the information argues in favor of invoking the exclusionary model, for such competitors.

The question remains whether other members of the public who might either intentionally or inadvertently reveal commercially sensitive information to competitors ought to be deprived of the information thereby be effectively excluded from full participation in the decision making process. This issue has plagued several health and safety agencies, and it has never been resolved to the satisfaction of all concerned.¹¹ Clearly, a solution must involve a balancing of the regulatee's interest in maintaining the confidentiality of commercially sensitive information against the interest of members of the public in having access to the data that underlies agency decision making.¹²

At the opposite end of the spectrum is the confrontational model. Although it too is one of the least desirable models, it also has ample historical precedent. Indeed, the confrontational model often results from an excessively stringent application of the exclusionary model. When an individual or group feels that it is being excluded from the decision making process or that its interes ts are not being adequately considered by an agency, it can simply confront the agency at every available opportunity with any available tools, including civil disobedience. During the 1980s, the picket lines manned by outraged neighbors of proposed hazardous waste dumps became familiar sights, as did shouting matches at city council meetings during debates about whether municipal solid waste incinerators should be constructed to reduce demand on limited landfill capacities. More serious confrontations result from attempts by activists to disrupt private initiatives that might pose risks to health or the environment. It is not unusual to hear of an individual chaining himself to a tree that is scheduled for cutting or pouring glop from a company's outfall onto the desk of its chief executive officer.¹³ Recently, a group of environmental activists in Austin "occupied" several small caves on private property in protest against scheduled development of the property that would pose a risk to several endangered insect species that inhabited the caves.¹⁴ On relatively rare occasions, the confrontations can even become violent, as the confrontation between the environmental organization Greenpeace and the French government over atmospheric testing of nuclear weapons near New Zealand a few years ago, tragically demonstrated.¹⁵

Although such confrontations are usually intended to be very public, they are not designed to be participatory and they are definitely not conducive to informed dialogue about risks and the measures that can he taken to reduce risks. Occasionally a decision maker will be persuaded to agree to talk with the activists in order to induce them to cease their disruptive activities, but rarely will they be asked to participate in the actual decision making process; nor would they necessarily be willing to participate, even if asked.

The confrontational model reflects a high degree of distrust among the activists in the process itself. Because they strongly perceive the process to be failing, the activists seek a wider audience. The model requires a high degree of commitment to the issues at stake, because individuals often participate only at considerable risk to their own economic and even physical well-being. Generally, only highly emotional issues or matters of high principal are likely to inspire sufficient devotion to justify such risks. Regulatory decisions that affect purely economic interests do not often invoke the confrontational model.

The confrontational model does appear to be effective in addressing matters of intense local interest. Even though a regulatory body is putatively obliged to render a decision based upon identified statutory or regulatory criteria, all agencies are politically responsible and therefore abhor confrontation. Confrontation can be very effective in persuading agencies to reconsider decisions previously made. When the activists attract the sympathy of a large number of otherwise inactive members of the public (as, for example, in mounting resistance to a hazardous waste dump), they are often highly successful in achieving their desired results, though not always (or even usually) through the existing decision making structures.

The model is not invoked frequently by national "public interest" groups to address broad issues of national importance. Perhaps the threat of arrest and prosecution is too large for national public interest group leaders, or it may be that confrontation is an ineffective tool for an entity that plans to keep its concerns on the public agenda for the long haul and therefore must be able to deal on a civil, if arms length, basis with regulatory decision makers and representatives of regulated entities. The relatively recent advent of the environmental group Earth First!, which advocates confrontational tactics and whose newsletter contains a suggestion box entitled, "Ecotage", and the successful confrontational style of the Green Party in Europe, may signal a change in this perception. At present, however, the confrontational model is largely restricted to local activities, and they are usually employed by individuals or local ad hoc groups.

Under the adversarial model, all interested persons have a right to participate in the decision making process by submitting facts, evidence, views and arguments. Each of the parties may make its own arguments and attempt to rebut the arguments of other participants. The agency assumes the position of neutral decision maker, dispassionately weighing the arguments of both sides of the relevant issues in accordance with objective criteria. Because courtroom adjudication is the origin of the adversarial model, it is usually invoked in the context of formal agency adjudication, which draws heavily upon the court room analogy. Parties are usually represented by counsel, and they have a right to submit direct testimony, usually through experts. Though cross-examination is not absolutely required, it is often allowed. A transcript of the agency hearing is usually prepared. Since the agency relies heavily on the parties for the information and analysis upon which it bases its decisions, the adversarial model is often used in the context of licensing products and technologies, such as pesticides and nuclear power plants where the applicant is required to submit information in accordance with predetermined data requirements. The model demands that the decision maker explain his or her decision to the winners and the losers, often according to a precise formula, requiring, e.g., "findings of fact" and "conclusions of law". This element enhances the legitimacy of the process and helps to ensure that the decision was in fact based on the neutral application of particular criteria to particular facts or circumstances.

The adversarial model presents a forum for structured debate among designated representatives of relevant interest groups. In regulatory proceedings involving risk management, scientific expertise has a role to play and scientists with different views may normally present their own views and criticize the views of others through the presentation of expert testimony. Yet there is no attempt to achieve a scientific consensus on important scientific or technical issues. Indeed, the process is usually dominated by lawyers, who are familiar with such structured decision making tools. Although the outcome of proceeding under the adversarial model can be a "settlement", which represents an agreed-upon compromise, the model presumes that there will be winners and losers.

The adversarial model traditionally dominated administrative decision making at the federal level, prior to the advent of the "rule making revolution" of the 1970s. Although anyone who is sufficiently aggrieved may participate in agency decision making under the model, the primary players tend to be the agency staff, the regulated industry and one or more national public interest groups or unions. The expense of formal participation, which as a practical matter usually requires a lawyer, usually dissuades persons without a strong interest in the outcome from participating. Sometimes financial limitations prevent individuals or groups with strong interests but few resources from participating. The expectation among all parties is that the agency's determination will be appealed to a court for a structured review by non-scientists in an adversarial context.

The adversarial model fosters a "we" versus "they" attitude among the participants that impedes compromise and facilitates feelings of disgruntlement. Rarely is everyone satisfied by the outcome, and there can even be a good deal of dissatisfaction on all sides about relatively trivial procedural rulings during the proceedings. Although the agency can attempt to cast its decision in scientific or technical terms, the sophisticated players at the national level know that policy considerations usually predominate, and they try their best to hold the decision maker accountable in both the reviewing courts and the courts of public opinion. At the same time, the adversarial model can broaden the range of fact and extend the policy options available to the decision maker. It also ensures that all affected interests have a relatively complete opportunity to convey facts and express views.

A close cousin of the adversarial model is the "due consideration" model, under which all interested parties are welcome to submit views and the agency is obliged to give "due consideration" to all relevant facts and arguments. As the model has evolved through the last two to three decades an additional requirement that the agency explain why it chose the option that it adopted has become an essential element of the model. The model obviously describes informal rule making under section the Administrative Procedure Act.¹⁶ Since a very large part of decision making involving risk management at the federal level is done in the context of informal rule making, the due consideration model is very relevant to risk management.

The due consideration model resembles the adversarial model in that it allows every interest to make its case for why it is right and the other parties are wrong. It also presumes a neutral decision maker who is swayed only by facts and analysis. Since it makes no attempt to bring the affected interests together to achieve an amicable resolution of the issue, it yields winners and losers.

The model differs from the adversarial model, however, in several subtle ways. First, the agency usually takes a position in advance of the public hearing and invites public comment on the agency's position as well as on the issues generally. Second, the parties therefore do not define the relevant issues to quite the same extent that they do in the adversarial model. Third, the model generally does not adopt all of the procedural protections of the adversarial model, and it is therefore corr espondingly less burdensome for the agency. Fourth, the agency relies more on its own sources of information and analysis than under the adversarial model, in which the parties are expected to be the primary sources of information. Finally, the agency is only required to give due consideration to all of the facts and analysis submitted by outsiders and explain its chosen course of action.

The due consideration model is not nearly as procedurally ponderous as the adversarial model, but it lacks the adversarial model's ability to cut through to the nub of arguments having surface plausibility but little factual support. The due consideration model is better adapted to issues that are policy dominated and for which factual accuracy is not essential. From the standpoint of the participants, the due consideration model is not as satisfying as the adversarial model or the mediation model to be discussed next. The participants are less directly involved in the actual decision making process: Suspicions are easily raised that the agency is not really giving due consideration to other points of view, especially when it adopts the option that it initially proposed.

Under the mediation model, representatives of groups with an interest in a regulatory decision meet together, often with the aid of a "mediator" or "facilitator", to present facts and arguments to one another and to try to reach an agreement on the ultimate result. The regulatory agency, which may either participate actively in the discussions or play the mediating role, then attempts to implement the agreed upon solution. Public participation may be invited at this step, but it is largely pro forma. At the federal level the mediation model has been pursued actively through the concept of "regulatory negotiation".¹⁷ Negotiated rule making involves broad discussions among all interested parties with the goal of arriving at a consensus on a proposed rule.¹⁸ The agency must appoint an official to convene and organize the negotiations, and it can also appoint a mediator to facilitate agreements.¹⁹ Ground rules for the negotiations are established by the participants at the outset. The negotiators then meet until they have reached agreement, until they have agreed that they will not reach agreement, or until a previously imposed deadline. Depending on how intensely the negotiators go about their work, the entire process can consume from six months to a year.²⁰ The Conference then recommended procedures to facilitate regulatory negotiation.²¹

The Conference's early optimistic assessment of negotiated rule making as a vehicle for facilitating dispute resolution in a highly participatory context presumed that most interest groups genuinely accepted the legitimacy of risk regulation by federal agencies and were willing to work within the current statutory and administrative framework. With some exceptions, these assumptions no doubt hold for major companies, trade associations, universities, and national public interest groups. When the negotiations are successful, the agency should be able to promulgate the proposed rule with substantial cost and time savings. Other advantages include the avoidance of litigation, the possibility of more accurately identifying the real concerns of the parties (because they do not engage in the posturing that occurs in litigation), the opportunity to identify the intensity of the parties' concerns over various issues, and the legitimacy the promulgated rule will enjoy because it was a joint product of the agency and the parties.²²

Some argue, however, that even at the national level, the mediation model is not likely to be very effective. For example, some argue that there is too much distrust between management and labor for this idea to function effectively in OSHA. Others point out that negotiated rule making will not be successful unless the parties believe that, if they do not negotiate an agreement, a decision will occur anyway. Most would agree with Harter that the mediation model has the greatest probability of success when:²³

(l) each party has power to influence the outcome of a pending decision; (2) the number of parties is small enough that negotiations can effectively take place; (3) issues are sufficiently "mature" that the parties are ready to decide them; (4) there is sufficient pressure to resolve the matter because otherwise the agency will do so; (5) the parties have something to gain from the negotiations; (6) there are not fundamental value conflicts between the parties which prevent the negotiation over the outcome; (7) there are a sufficient number of contested issues that trade-offs are possible between the parties; (8) the resolution of the dispute depends on policy decisions for which there are no objective solutions; and (9) the framework of the negotiations include an effective method to implement decisions reached during them.

During their 90 day tenure, the committee meets with two of the developer's representatives.²⁵ The applicant has representatives at the committee meetings. He has a duty to assign a person with a position of authority.²⁶ At all meetings, the applicant will provide for a technical advisor and a second representative, who is not to be an attorney.²⁷ At the end of 90 days, the local review committee prepares a report pinpointing both the resolved and unresolved issues. The report is then submitted with the permit application to the relevant state agency.²⁸ The committee's report must identify the issues that were resolved and those that were left unanswered. Essentially, it is a fact-finding report that may prove useful to both the community and the agency that considers the technical merit of the application.²⁹ Nevertheless, the Keystone Center has proposed, and the SNR of Texas has adopted, a mediation process for channeling local disputes over siting hazardous waste disposal facilities that has been enacted into law in Texas. The Keystone process is a method by which a developer who wishes to construct a solid waste facility may undergo negotiations with representatives of the local community in an attempt to mitigate public opposition.³⁰ A "site-specific" team is appointed to review and perhaps modify the developer's proposal but it has no authority to submit a recommendation.³¹ Rather, the team is "to assist both permit applicants and the public in identifying items of mutual concern in the siting of environmentally sound hazardous waste facilities".³²

At the core of the Keystone process is the idea of facilitating dialogue between the applicant and the public.³³ The ultimate goal of the committee is to prepare a report detailing citizen concerns and the manner in which the applicant will deal with them.³⁴ By working with the public early in the process, the group can work to reduce the potential for misinformation about local projects for which risk perceptions are very high.³⁵

The Keystone process has been tested on at least two occasions. In 1985, the Gulf Coast Waste Disposal Authority (GCA) proposed to construct an incinerator at its existing landfill and waste water treatment plant in Texas City. The committee in Texas City acted fairly well as a conduit between the community and the developer. Of 24 issues raised, the applicant satisfactorily responded to 17, and 7 were left as "unresolved". The appropriateness of the selected site, however, was not an item that the the committee addressed.³⁶

According to one participant, the process was successful in three regards. First, more than half of the issues were resolved by the process.³⁷ Second, there was a change of attitude of the participants: A survey revealed that the members of the committee had a greater trust in the applicant after the process. There was also a shift in the attitudes regarding the facility in that concern about transportation decreased with a corresponding increase in concern over non-local waste. Third, the participants indicated that they were greatly pleased with the process. ³⁸

The project, however, failed in its ultimate objective of eliminating the need for lengthy hearings. A long public hearing on the facility is inevitable.

The second project on which the Keystone process was used in Texas concerned a secret request from an industry wanting to construct a hazardous waste management facility in East Texas. The developer contracted directly with the Keystone Cent er to facilitate public participation and to provide necessary clerical help.³⁹ Under the contract, the members appointed were to be trained by the Center. Unfortunately, city officials knew nothing about the project until its existence was leaked to a local newspaper; nor did they know any of the Texstar executives or the Keystone Group members. Not surprisingly, this colored local attitudes toward the process. Although some of these feelings were resolved, the committee eventually disbanded when Texstar refused to give them exclusive control over the selection of a facilitator.⁴⁰

At the national level, the mediation model attempts to facilitate compromises among well-informed representatives of contending groups. The model facilitates the flow of information, but its primary purpose is not scientific dialogue. The model recognizes that few of the important issues will be resolved by technical expertise and that most require political compromise. At the local level, the mediation model must educate as well as facilitate agreements of a political nature. One of the primary goals of the Keystone process, for example, is to educate worried local activists about the hazards of solid waste disposal facilities. The Texstar experience indicates, however, that building trust is an especially important function of the mediation model at the local level. People will refuse to be "educated" by what they regard to be a biased source.

The advisory committee model closely resembles the mediation model, except that it relies heavily on scientific and technological expertise. Under that model, the decision maker appoints a committee of disinterested experts to advise the agency on technical issues. The experts deliberate in accordance with scientific norms, analyze the scientific reports, debate about the proper interpretations and inferences to give to and draw from the scientific data, and ultimately advise the decision maker on how he should resolve the issues. While there may still be a public proceeding using either the adversarial or due consideration models, the advice of the experts weighs very heavily in the agency's ultimate decision.

Since technical considerations are often close to the heart of risk management, this model is attractive to decision makers who are not scientifically trained. Even though many science policy disputes that arise in connection with risk assessment and risk management are policy dominated, the advisory committee model can be attractive to policy makers who do not want to "take the heat" for resolving policy questions. If the issues can be made to appear to be technical in nature and if a group of disinterested and highly qualified experts expresses a view on the issue after due deliberation, then the decision maker can avoid the accountability that goes along with tough decision making. The decision maker, to some degree, loses control over the outcome of the decision making process, but even that disadvantage can be reduced by choosing experts for the advisory committee whose views on the technical issues will yield results that are in accordance with the decision maker's policy preferences.

On the public participation spectrum, the advisory committee model is located close to the exclusionary model. Only credentialed experts are invited to participate on the advisory committees, and the experts are not necessarily chosen to reflect different scientific perspectives. In some manifestations, the advisory committee model incorporates some participatory elements. For example, the Federal Advisory Committee Act requires that federal advisory committees be "balanced", but that has not been read to imply that such committees be chosen from expert representatives of particular interest groups. The Occupational Safety and Health Act requires that advisory committees on individual standard setting initiatives be composed of at least one designee of the Secretary of Health and Human Services, at least one representative of state health and safety agencies, and an equal number of representatives from labor and management.⁴¹ Although the OSHA advisory committee structure is more participatory than the others, the agency has used it only rarely, in part because advisory committees so composed could rarely reach consensus. One observer reported that advisory committee meetings "did little more than provide a forum for the contending parties -- labor and employers -- to argue with each other."⁴² Thus while the OSHA model provided effective public participation, it did not comport with the image of an advisory committee as a body of experts dispensing unbiased advice.

Administrative policy makers have a wide variety of models from which to choose in selecting a decision making vehicle that addresses the complex scientific and technical issues and emotionally charged p olicy questions that arise in risk assessment and risk management. Which model the policy maker chooses should, to some degree, reflect the extent to which the agency desires public input into the decision making process and the extent to which countervailing policies preclude public participation.

^* Professor McGarity teaches at the University of Texas School of Law. His B.A. (physics) is from Rice, and his J.D. is from the University of Texas.

¹ Also consider that the Administrative Procedure Act, codified at 5 U.S.C. §§ 551 et seq., has been essentially unchanged since its passage in 1946.

² See generally, STEWART, THE REFORMATION OF AMERICAN ADMINISTRATIVE LAW 33 HARV. L. REV. 1669 (1975); T.J. LOWI, THE END OF LIBERALISM 125,126,148-49, 155 (1969) -- (describing in disparaging terms "Interest Group Liberalism").

³See Latin, Good Science, Bad Regulation, and Toxic Risk Assessment, 5 YALE J. ON REG. 89 (1987); McGarity, Substantive and Procedural Discretion in Administrative Resolution of Science Policy Questions: Regulating Carcinogens in EPA and OSHA, 67 GEO. L. J. 729 (1979).

⁴ See generally, Crampton, The Why, Where and How of Broadened Public Participation in the Administrative Process, 60 GEO. L. J. 525 (1972); Gellhorn, Public Participation in Administrative Proceedings, 81 YALE L. J. 359 (1972).

⁵ See, e.g., Office of Communication of United Church of Christ v. FCC, 359 F.2d 994 (D. C. Cir. 1966); National Welfare Rights Organization v. Finch, 429 F.2d 725 (D. C. Cir. 1970).

⁶ See, e.g., Castle v. Pacific Legal Foundation, 445 U.S. 198 (1980).

⁷ See, e.g., Atchison, Topeka & Santa Fe Ry. Co. v. Calloway, 382 F. Supp. 610 (D. C. Cir. 1974).

⁸ I agree with Jacks that time consumed by public participation in administrative decision making constitutes "delay" only if it performs no function or is dysfunctional. Jacks, The Public and the Peaceful Atom: Participation in AEC Regulatory Proceedings, 5 TEX.. L. REV. 466, 506-511 (1974).

⁹ Association of Data Processing Service Organizations v. Camp, 397 U. S. 150 (1969).

¹⁰ See Jacks, supra note 8, at 511-513.

¹¹ The Federal Insecticide, Fungicide and Rodenticide Act, codified at 7 U. S. C.§ 136 et seq., for example, provides that health and safety data are not confidential business information and are therefore disclosable. 7 U. S. C.§ 136h (d)(1) (1980). Companies, however, may not use the information without providing compensation to the producers of the data. 7 U. S. C. § 136a (c) (1 )(D) (ii) (1980).

¹² See McGarity & Shapiro, The Trade Secret Status of Health and Safety Testing Information: Reforming Agency Disclosure Policies, 93 HARV. L. REV. 837 (1980).

¹³ See, e.g., 8 From Greenpeace Held for Plugging Waste Line, New York Times, April 23, 1985, at A15, col. 1.

¹⁴ See Environmentalists Vow to Continue Occupation of Caves, Austin American-Statesman, Sept. 1,1988, at B22, col. 5.

¹⁵ 1 Killed as 2 Blasts Sink Greenpeace Protest Ship, Los Angeles Times, July 11, 1985, at A12, col. 1.

¹⁶ 5 U.S.C. § 553 (1977).

¹⁷ See generally, Harter, Negotiating Regulations: A Cure for Malaise, 71 GEO. L. J. 1 (1982); Perritt, Analysis of Few Negotiated Rulemaking Efforts, reprinted in 1985 Reports and Recommendations of the Administrative Conference of the United States 637. The Administrative Conference of the United States has endorsed one subspecies of regulatory negotiation (viz. "negotiated rule making"). The Conference noted that:

[p]articipants in rulemaking rarely meet as a group with each other and with the agency to communicate their respective views so that each can react directly to the concerns and positions of the others in an effort to resolve conflicts." The Conference concluded that "if the parties in interest were to work together to negotiate the pros of a proposed rule, they might be able to identify the major issues, judge their importance to the respective parties, identify the information and data necessary to resolve the issues, and develop a rule that is acceptable to the respective winners, all within the contours of the substantive statutes.

¹⁸ Administrative Conference of the United States, Recommendation No. 85-5, Procedures for Negotiating Proposed Regulations, 1985 ACUS 23-27 (hereinafter 1985 ACUS Recommendation); Administrative Conference of the United States, Recommendation No. 82-4, Procedures for Negotiating Proposed Regulations, 47 Fed. Reg. 30708 (1982) (codified at 1 C.F.R. 𨶉.82-4 (1990)).

¹⁹ 1985 ACUS Recommendation, supra, at 26; see generally, Harter supra note 17, at 67-82.

²⁰ Administrative Conference of the United States,Recommendation No. 82-4, Procedures for Negotiated Rulemaking, supra note 18.

²¹ Id.

²² See Harter, supra note 17, at 28-31.

²³ See Harter, supra note 17, at 42-52.

²⁴ 25 TEX.. ADMIN. CODE 325.64 (a) (1986); TEX.. HEALTH AND SAFETY CODE ANN. § 361.063 (a) (Vernon 1990).

²⁵ Keystone Workshop, THE KEYSTONE SITING PROCESS HANDBOOK, 7 (1984) (available from Tex.. Dept. of Water Resources and Tex.. Dept. of Health) [hereinafter KEYSTONE HANDBOOK].

²⁶ Id.

²⁷ KEYSTONE HANDBOOK, supra note 25, at 9.

²⁸ 25 TEX.. ADMIN. CODE 325.64 (b) (1986); TEX.. HEALTH AND SAFETY CODE ANN. § 361.063 (j) (Vernon 1990).

²⁹ KEYSTONE HANDBOOK, supra note 25, at 3. Bowman, The Keystone Process: An Evaluation of its Test Cases 11 (1988) (available from author).

³⁰ The process was developed in response to public opposition to the attempts of the Gulf Coast Waste Disposal Authority (GCA) to site a high-technology hazardous waste facility in southeast Harris County. As a result of the turmoil, the GCA gave grants to the Keystone Center, a non-profit organization that addresses environmental and public policy issues, to sponsor two workshops that would focus on the siting of hazardous waste facilities. Thirty-five participants, mostly Texans, who represented the state and local governments, industry, environmental organizations, public interest groups, law and engineering, attended the workshops. The keystone approach to siting these waste facilities was developed and a report setting forth the procedure was produced. (Siting Waste Management Facilities in the Galveston Bay Area: A New Approach, Report of the Keystone Workshops on Siting Nonradioactive Hazardous Waste Management Facilities, Held 17-20 August 1982 at Keystone, Colorado, and 21-23 October 1982 at West Columbia, Texas). Additionally, the group wrote THE KEYSTONE HANDBOOK, supra note 25.

³¹ The committee's purpose is to "compliment the present permitting process, not to delay it or substitute for it." For example, the committee has no power to recommend alternative sites. Bowman, supra note 29.

³² KEYSTONE HANDBOOK, supra note 25, at 1.

³³ KEYSTONE HANDBOOK, supra note 25, at 4.

³⁴ Id.

³⁵ Id.

³⁶ Bowman, supra note 29, at 8.

³⁷ Bowman, supra note 29, at 10-11.

³⁸ Bowman, supra note 29, at 11.

³⁹ Bowman, supra note 29, at 6.

⁴⁰ Bowman, supra note 29, at 7.

⁴¹ 29 U.S.C. § 656(b) (1985).

⁴² T. GREENWOOD, KNOWLEDGE AND DISCRETION IN GOVERNMENT REGULATION 125 (1984).