The principal features of American rulemaking -- information, interest participation, and judicial review -- are readily apparent from the hazardous waste example. First, the rulemaking proceeding and judicial review generated considerable information about the BIF Rule. Just from reading the proposed rule, first supplement, second supplement, final rule, and court opinion we have learned a lot about the science, efficiency considerations, industry costs, and health risks that went into choosing the BIF Rule as the means of ensuring that hazardous waste, when used as fuel, is burned safely. And it is not difficult to imagine the additional reams of paper and, presumably, information contained in a rulemaking record that includes EPA-commissioned scientific studies, comments from over thirty major participants, and individual responses to those comments.
Second, a variety of interests had a say in rulemaking. The hydrocarbon standard was added (in the first supplement) as an alternative regulatory solution to the hazardous organic compound problem because firms in the cement kiln industry complained that, even though their emissions were not necessarily hazardous, it would be impossible for them to comply with the carbon monoxide standard. The EPA also listened to those concerned that the initial proposal was not protective enough of human health: it added a control on particulate emissions because of comments contending that they also posed a health risk. In court, both industry representatives and environmental protection groups challenged the regulation.
Finally, the judiciary played an active role in rulemaking. When the governing statute was ambiguous or failed to address the question at hand, the Horsehead court carefully examined the EPA's regulatory choices for reasonableness. It analyzed the organic compound emissions standards to ensure that the agency had, in actual fact, given notice. And it picked through the record, checking the evidence pointed to by the EPA in support of the third emissions standard, to decide the arbitrary and capricious claim. In sum, the agency was forced to explain its decision in terms comprehensible to the general public and answer objections from rulemaking participants.
Information, participation, and judicial review are inextricably linked. More information about a rule lays the groundwork for interest group critique and thorough judicial review. Interest groups generate information because of their familiarity with the industry and their strong incentives to contest or promote agency action, which in turn permits more extensive court scrutiny. Tough judicial review feeds back into expectations as to how much information must be provided by agencies and how seriously rulemaking participants' objections and suggestions must be taken.
Legal rights to acquire information, participate in the rulemaking process, and obtain extensive judicial review guarantee accountability and interest representation in the United States' system of separation of powers and pluralist interest group politics. As argued above, these elements of American rulemaking allow Congress to monitor and control the administrative process even though the executive branch is independent. Also as argued above, legal rights to information, participation, and judicial review enable the many interest groups that influence the legislative process -- through the Senate, House of Representatives, and the President -- to also influence the administrative process, notwithstanding the President's control of the executive branch.
These features of notice and comment rulemaking, however, also have negative consequences for the administrative process. Critics claim that notice and comment leads administrators to make too many concessions to "special" interests. They also point to the slow pace of American rulemaking and argue that it is a product of the numerous legally guaranteed rights that individuals enjoy in the process. In this section, I draw on the voluminous academic literature on notice and comment to evaluate these claims and conclude that, notwithstanding the costs of the system, notice and comment should be adopted in the Community.
American rulemaking has been criticized for giving too much power to interest groups, in particular special interest groups. This claim is related to the vociferous opposition wet kiln operators mounted to the BIF Rule. Although both industry and environmental protection groups took part in the EPA proceeding, business interests proved most difficult to accommodate and occupied a disproportionate share of the Agency's and the court's time. The first supplement to the proposed rule, in which the Agency answered the bulk of the comments, contained a little for everyone: particulate emissions standards for environmental interests and a new toxic organic emissions standard for business interests. The following four years, however, were spent dealing with complaints from cement kilns operators (especially the wet kiln subset) and other industry interests. Fewer than fifty-seven furnace operators -- and within that set a particularly obstinate group of six to ten wet kiln operators -- caused the bulk of the EPA's worries.147 And the extra time spent trying to find a regulatory solution that would placate wet kiln operators (i.e., the third toxic organic compound emissions standard) was very likely meant to fend off a successful challenge in court. In other words, where the costs of regulation were especially concentrated, interests mobilized against it.
Public choice scholars argue that interest groups such as the furnace operators appropriate public resources for private ends through their involvement in policymaking and are thus ultimately destructive of public welfare. The public choice critique came in response to the view, popular in the 1970s, that the rulemaking process embodies pluralist democracy.148 According to this theory, a rulemaking proceeding can be conceived of as a mini-legislature in which debate generates widely accepted and mutually beneficial public policy. But public choice thinkers, drawing on the work of Mancur Olson and other members of the rational choice school, maintain that pluralist interest politics fail to generate administrative rules in the general welfare.149 Special interest coalitions -- called special because they speak for a small group of firms or individuals and not for larger constellations such as consumers -- benefit at the expense of the environmental protection, consumer welfare, and other public goals that, in theory, statutes and regulations are designed to accomplish. In the rational choice model, lawmakers, whether they be legislators or agency officials, are passive vehicles for interest group preferences and lawmaking is a competitive process among interest groups. Groups with few members who stand to win or lose heavily from regulatory change are most likely to organize and put pressure on politicians and administrators because each single member has a significant stake in the outcome. By contrast, large groups are likely to be poorly organized because the same benefits and costs of regulatory change are spread thinly across a large membership, and individual members do not have the necessary incentives to contribute to the collective effort. Consequently, the argument goes, even though many of the great statutes of the 1960s and 1970s were designed for large and diffuse groups (e.g., consumers and environmentalists), when one looks closely at the text and the regulatory schemes developed to implement the statutes, they tend to work to the advantage of small groups of business and regional interests. A number of empirical studies on agency regulation bear out the rational choice hypothesis.150
Even though the literature has generated a series of reform ideas that would transfer policymaking power to institutions less susceptible to special interest pressure, none is particularly convincing.151 The defects public choice identifies are endemic to politics and no one has, as of yet, convincingly argued that either Congress or the executive branch is better at fending off private interests demands.152 Moreover, the exceedingly dark view public choice takes of interest participation is unwarranted. Notwithstanding its predictions, legislative politics and administrative procedure are used by public interest groups as well as "special" ones.153 Further, public choice theory rests on the assumption that the public welfare, what special interests tend to thwart, may be determined absent reference to politics by summing up individual preferences and utilities and arriving at an optimal policy outcome. Such preferences and utilities, however, are themselves the product of a political process in which interest groups are vital participants.154 They represent their constituencies, which indeed may be broader than they appear (e.g., coalitions between firms that produce environmental technologies and the environmental lobby), and galvanize public opinion. Public life in a liberal democracy without interest groups is not only impossible but also undesirable.
The slowness of American rulemaking is one of its main drawbacks. The four years it took the EPA to complete the BIF Rule (and that is not even counting the time used in drafting the initial proposed rule) and the additional two years it took the court to hand down its decision are not at all uncommon. In this area of administrative law, the adjectives of choice are "ossification"155 and "gridlock."156 Fewer rules is not the only cost of slowness. To avoid rulemaking's cumbersome record-building requirements and tough judicial review, agencies sometimes resort to adjudicatory regulatory techniques such as product recalls.157 Yet adjudication can be an inefficient form of agency action and can cause uncertainty for the regulated community, in violation of rule of law principles. Rather than fostering openness and participation in administration, therefore, a sluggish rulemaking process may actually cause a return to unaccountable and arbitrary agency action.
By common agreement, courts and judicial review must take some of the blame for the snail's pace at which rulemaking proceeds. Although the critical view has always been somewhat exaggerated and was probably truer of courts in the 1970s than at present, it nonetheless captures the relationship between tough judicial review and agency process. In response to each legal challenge, courts require that administrators point to a part of the rulemaking record where they entertained the objection, but marshalled scientific evidence and policy arguments against it. Rulemaking, therefore, has turned into a protracted deliberation before agencies in which regulators must carefully document and explain rules so that, if challenged in court, they can mount a successful defense. Agencies bear an especially high burden of explanation because of the unpredictable quality of so-called "hard look" judicial review. They cannot be certain which issues will appear salient to the group of three non-expert judges who review the rule and therefore the paper trail must cover the entire world of possible issues and non-issues, large and small. Consequently, rulemaking has become a long process of comments, agency answers, second-round comments, second-round agency answers, legal challenge, and, if the agency failed to do a thorough job, a court judgment remanding the rule to the agency for further explanation and development of the evidentiary record, or, albeit rarely, a judgment vacating the rule and forcing the agency to start from scratch.
How can American administrative law be changed to ensure Congressional control and pluralist interest group participation without gridlock? Much of the reform debate centers on reducing the role for judicial review. One option is to relax the standard of review employed by courts. A problem with this approach is that it is difficult to instruct courts to move to a less demanding standard of review. It is enough to recall that so-called "hard-look" review developed within the statutory framework of "arbitrary and capricious" review to question whether a change in verbal formula would have a direct effect on judicial practice.
Jerry Mashaw, co-author of an influential case study on automobile regulation, suggests that the judicial burden can be lightened by changing the availability rather than the nature of review.158 He argues that if the timing of review is altered, litigants will resort to courts when they have meritorious claims and not, as is currently the case, because the odds are such that it pays to litigate rather than comply. Most rules are reviewed prior to enforcement.159 Because fines for non-compliance only start running once a rule comes into effect (and a regulation that requires large capital investments necessarily includes some lead time), a firm that challenges a rule will generally only incur the cost of legal fees. Moreover, should it make the opposite choice and begin investing in compliance, it risks losing market share because other firms that take the judicial review gamble and win will have avoided the compliance investment (generally more costly than legal fees). Assuming a fifty-percent chance that the rule will be vacated or remanded (which roughly reflects the current odds), Mashaw shows that a firm will always choose to pay the legal expenses and sue rather than comply. In other words, if a firm believes that its various grounds for challenging a rule give it at least a fifty-percent chance of winning, then it will go to court, even though a fifty-percent chance, especially in a world of tough judicial review, is hardly a deep-seated conviction that the agency behaved illegally. In sum, the litigation structure along with the lead time built into most regulations encourages those who stand to lose from changes in the status quo to fight all the way to the bitter end.
Mashaw suggests that review be delayed until after rules come into effect so that the choice to litigate carries with it the extra cost of non-compliance fines. In such a system, litigants would only go to court if they had a relatively high probability of winning (to offset the additional cost of non-compliance fines). In other words, they would only bring meritorious claims. Delaying review, according to Mashaw, would also improve the quality of judicial fact-finding. Post-enforcement, courts can see how a rule operates in practice and consequently the limitless universe of ill-defined, possible defects is narrowed to a few concrete flaws. By tinkering with the timing of review, Mashaw hopes to solve the gridlock problem while preserving courts as guardians of the rule of law and the sound use of discretion in administrative decisionmaking.
This proposal too is flawed. Most obvious is the hardship to parties, the concern that led to pre-enforcement review in the first place. Faced with a possibly illegal rule, regulated individuals and firms would have to choose between non-compliance fines and the risks of litigation on the one hand and capitulation on the other. Second, the social learning that supposedly occurs once a rule comes into effect requires considerable monitoring and information-gathering and it is not at all clear that administrative agencies or other actors, such as trade associations, would in fact undertake this task.
147 See Burning of Hazardous Waste in Boilers and Industrial Furnaces, 52 Fed. Reg. 16,982 (May 6, 1987) (regulatory impact analysis); Horsehead Resource Development Co. v. Browner, 16 F.3d 1246, 1266 (D.C. Cir. 1994).
148 This theory is widely accepted as Richard Stewart's brainchild. See Richard B. Stewart, "The Reformation of American Administrative Law," 88 Harv. L. Rev. 1667 (1975).
149 See Mancur Olson, The Logic of Collective Action. Public Goods and the Theory of Groups (1965). For the more recent version of rational choice theory that takes a broader view of the costs and benefits of collective action, see Terry M. Moe, The Organization of Interests: Incentives and the Internal Dynamics of Political Interest Groups (1980).
150 See Ann P. Bartel & Lacy Glenn Thomas, "Direct and Indirect Effects of Regulation: A New Look at OSHA's Impact," 28 J. L. & Econ. 1 (1985); Peter Linneman, "The Effects of Consumer Safety Standards: The 1973 Mattress Flammability Standard," 23 J.L. & Econ. 461 (1980).
151 See Peter H. Aranson, Ernest Gellhorn & Glen O. Robinson, "A Theory of Legislative Delegation," 68 Cornell L. Rev. 1 (1983). Administrative law scholars Aranson, Gellhorn, and Robinson argue that all lawmaking, even, say, detailed rules on the disposal of hazardous waste, should occur in the light of day of the legislature because although special interest deals will still occur, they will be visible to the electorate and legislators will have to take responsibility for them. If legislators are at all concerned about reelection, according to this view, they will think twice before supporting a law that sacrifices the public good for private welfare. On the opposite end of the spectrum, Jerry Mashaw argues that administrators are more likely than legislators to protect the common good because they are directly responsible to the President who in turn is elected by a national constituency in which single interest groups, unable to deliver a sizable chunk of the vote, have little influence. See Jerry L. Mashaw, Greed, Chaos, and Governance 147-55 (1997).
152 In another reform suggestion, administrative and constitutional law scholar Cass Sunstein maintains that agencies and courts can reduce special interest politics through what he calls "deliberative decisionmaking." See Cass Sunstein, "Factions, Self-Interest, and the APA: Four Lessons since 1946," 72 Va. L. Rev. 271, 284-85. With this model, administrators are to put expertise and technical knowledge at the service of public welfare goals and resist the temptation to fall prey to powerful lobbying groups. It has four central features: the pluses and minuses of different regulatory approaches should be fully explored; possible conflicts between different values, say, between a cleaner environment and keeping factories open should be resolved by reference to the enabling statute; to the extent that the statute does not resolve such conflicts, administrative discretion should proceed through identification of the relevant factors and solicitation of the public's views; and the agency's final decision must reflect a reasonable balancing of those factors. With this type of deliberation, agencies, under the supervision of courts, can bypass groups that seek to use rulemaking to their private advantage. The most obvious flaw with this proposal is the trust it places in administrators and courts to understand science, interpret the statute, balance competing values, and arrive at reasonable solutions.
153 See Peter Schuck, Against (And For) Madison: An Essay in Praise of Factions, 15 Yale L. & Pol'y Rev. 553, 574-76 (1997).
154 See generally id. at 587-88 (arguing for a procedural conception of the public interest in which interest groups figure prominently); Alan Schwartz, Statutory Interpretation, Capture and State Tort Law 41-44 (Sept. 28, 1998) (unpublished manuscript) (criticizing capture theory on grounds that it fails to identify a substantive norm against which regulations that have purportedly come under special interest influence may be measured).
155 Paul R. Verkuil, "Rulemaking Ossification--A Modest Proposal," 47 Admin. L. Rev. 453 (1995); Richard J. Pierce, Jr., "Seven Ways to Deossify Agency Rulemaking," 47 Admin. L. Rev. 59 (1995).
156 Jerry L. Mashaw, Greed, Chaos, and Governance: Using Public Choice to Improve Public Law 203 (1997).
157 See Jerry L. Mashaw & David L. Harfst, The Struggle for Auto Safety 147-56 (1990) (describing the National Highway Traffic Safety Administration's resort to automobile recalls after failed attempts at rulemaking).
158 See Jerry L. Mashaw, Greed, Chaos, and Governance 166-90 (1997).
159 See Abbott Laboratories v. Gardner, 386 U.S. 136, 149 (1967). Abbott Laboratories held that pre-enforcement, agency action may be reviewed if the factual and legal issues are sufficiently clear and delaying court consideration would cause hardship to the parties.