Federal regulatory agencies and state common law courts play different but complementary roles in protecting citizens from the undesirable consequences of unregulated private markets. Congress created the regulatory agencies to provide ex ante protection by promulgating standards and establishing permit requirements applicable to private sector actors. Both permit requirements and regulatory standards often give targets of regulation a good deal of discretion in determining exactly how to comply. Nevertheless, agencies attempt to articulate the outer boundaries of legally permissible activities with some precision, because the government will ultimately bear the burden of proof in any subsequent enforcement action. State common law courts also craft rules for determining liability after a product or activity has caused harm, and in many cases those common law standards are designed to discourage socially undesirable behavior by forcing those who engage in it to compensate the victims. The broad standards that common law courts prescribe, such as the “reasonable person” test for negligence, are not stated with a great deal of precision, and their breadth and vagueness can leave prospective plaintiffs and potential defendants with little guidance as to whether a court will permit a jury to find that any particular product or activity violates the relevant standard.

Federal regulation and state common law litigation are both information-intensive exercises, imposing diverse burdens on regulators, litigants, and courts. Federal regulatory schemes typically impose threshold informational burdens but are far less comprehensive or consistent in tracking postaction developments.

Deconstruction Writ Large

When a new scientific study suggests that an industry's products or activities may be causing unanticipated adverse effects on health or the environment, a typical reaction by that industry or other affected stakeholders is to attack the messenger. This chapter details the strategies that risk-producing industries have employed to deconstruct scientific studies and to create the impression among regulatory agencies and the general public that such studies are no more than “junk science.” These strategies depend on a number of legal processes and tools that not only legitimate partisan, biased attacks on research, but also provide a broad array of mechanisms to facilitate these attacks.

Scientific studies documenting risk pose a serious threat to an industry's economic well-being, motivating companies and industry trade associations to mount aggressive efforts to belittle that threat by demanding the underlying data, reanalyzing those data, finding flaws with the study's conception and execution, hiring scientists to publish letters to the editors of scientific journals, withdrawing support from the academic institutions that employ the authors of the study, and challenging the integrity of the scientists in available forums. The scientists who produce such studies are rarely prepared to ward off such attacks on their work and even on their personal integrity. Unlike the industries that launch the attacks, they cannot afford to pay other scientists to respond to the attacks, hire expensive lawyers to defend their studies in court, and retain public relations firms to defend their reputations in academia and the public at large.

The team approach is by far the most prevalent decisionmaking model in federal regulatory agencies. The primary subordinate decisionmaking entity in this model is a team, or work group, composed of representatives from all of the institutional subunits within the agency that have an interest in the outcome of the rulemaking process. Typically, a team is composed of representatives from the program office, the research and development office, the policy office, the Office of General Counsel (or Solicitor's Office), the enforcement office, and one of the agency's regional or field offices. The team meets periodically to discuss regulatory options, to examine problem areas, to respond to requests of upper-level decisionmakers, and to resolve disputes among team members. Although the program office usually has responsibility for the rulemaking effort, the group often delegates tasks to individual members of the team or to small subcommittees. Copies of memoranda, rulemaking and regulatory analysis documents, and other decisionmaking documents are circulated to team members for review and comment. All team members are regarded as coequal participants in the decisionmaking process, and there is usually strong pressure, deriving from both the group psychology of the effort and from upper-level decisionmakers, for the team to reach consensus on important questions.

Mechanisms for resolving disputes on teams vary. Sometimes disputes are resolved by votes, with the minority acquiescing in the majority's decision. Usually, dissenters are allowed to elevate disputes through formal or informal channels to higher levels for resolution. A team member, however, can only elevate dissents a limited number of times without fracturing the overall consensus necessary for the effective operation of the team.

In the late 1960s and early 1970s, astute observers of federal regulation concluded that the cumbersome adjudicatory procedures with which agencies implemented most regulatory actions were ineffectual. Responding to this criticism, many agencies began to use informal procedures to promulgate rules of general applicability. Rather than “put on a case” before an Administrative Law Judge using expert testimony subject to cross-examination and various evidentiary limitations, agencies began to experiment with “paper” rulemaking hearings in which the agency offered a proposed rule and supporting data and analysis for broad public comment. At roughly the same time, Congress enacted a new generation of safety and environmental statutes that empowered, and often required, agencies to govern through informal rulemaking. The “rulemaking revolution” that resulted had the potential to expand enormously the federal government's regulatory powers.

Three themes of regulatory reform

The rulemaking revolution had been under way for less than a decade when regulated industries and some academics began to complain that the federal agencies were going too far. Operating beyond the range of effective political control, they were irrationally imposing burdensome requirements on companies without taking into account their costs or assessing their corresponding benefits. These criticisms and their associated prescriptions for change paraded under the broad banner of regulatory reform.

Some of the critics longed for less burdensome times when administrative agencies were more sympathetic to regulated industries. Unconvinced of the social desirability of government intervention into the marketplace, these critics argued that the only effective solution to the regulatory morass was to remove the dead weight of government regulation from the back of American industry. In this view, regulatory reform meant “regulatory relief.”

The three regulatory reform themes described in Chapter 1 – rational decisionmaking, bureaucratic accountability, and regulatory relief – merged in the mandates of Executive Order 12,291 and the Regulatory Flexibility Act that agencies analyze the impacts of their regulations on the regulated entities. Many regulatory reformers believe that regulatory analysis will inspire agencies to promulgate more logical (and less burdensome) rules in a way that ensures accountability to the President, the Congress and the public. These are high expectations indeed for a conceptual framework, but to some extent they are warranted. Comprehensive analytical rationality has many virtues, and it has great potential for enhancing the quality of regulatory decisionmaking. This chapter will first examine the concept of regulatory analysis and attempt to place it in the context of a larger body of experience and thought devoted to the broader practice of policy analysis. Next, the chapter will discuss the virtues of analysis and suggest how regulatory analysis can advance the goals of rational decisionmaking, bureaucratic accountability, and regulatory relief. We will save for Chapter 9 the discussion of the impediments to using regulatory analysis and the inherent limitations in the conceptual framework that make it less useful and desirable in real-world decisionmaking contexts.

Essential aspects of regulatory analysis

The enormous expansion of government since World War II has precipitated many ambitious attempts to manage governmental decisionmaking. The success of quantitative techniques to manage the war effort made students and practitioners of government optimistic that similar techniques could aid policymakers in other large bureaucratic contexts.

Rather than leaving regulatory analysis entirely to the agencies, Executive Order 12,291 assigned an important review role to the Office of Management and Budget (OMB). It empowered OMB to designate any minor rule as “major” and to waive regulatory analysis requirements for any major rule. OMB was given authority to identify “duplicative, overlapping and conflicting rules” and rules that were “inconsistent with the policies underlying” the Executive Order and to require “appropriate interagency consultation to minimize or eliminate such duplication, overlap, or conflict.” Finally, the Executive Order required OMB to “[m]onitor agency compliance” with its requirements and to “advise the President with respect to such compliance.” AH regulations and all RIAs accompanying major regulations are reviewed by desk officers in OMB's Office of Information and Regulatory Affairs (OIRA), and most RIAs are also reviewed by a separate group of “superanalysts” in OIRA that set the analytical agendas for all executive agencies. Disputes between OMB and the agencies have been resolved by a vague process that includes the White House. In practice, the vast majority of disputes are resolved in the informal give and take of day-to-day interactions.

The relationship between OMB and most executive agencies cannot fairly be characterized as a cooperative pursuit of common ends. To the contrary, it is typically highly adversarial and often acrimonious. According to a former Deputy Administrator of OIRA: “We yell and scream, jump up and down, do whatever we can to get them to listen to us.” Disputes between OSHA and OMB have been especially bitter.

Many agencies did not modify existing institutional arrangements to meet the analytical requirements of the Executive Orders and the Regulatory Flexibility Act (RFA). Several simply added the task of preparing regulatory analysis documents to the responsibilities of the technical staff without creating a separate regulatory analysis office. The same employees who draft the rulemaking documents also draft the regulatory analysis documents. The regulatory analysis then winds its way up through the hierarchy of the program office along with the other rulemaking documents. Under this “hierarchical model” the technical staffers and their superiors at each level in the agency hierarchy can become aware of the relevant information and analysis, but there is no separate institutional entity to skeptically review rulemaking documents and otherwise carry the flag for comprehensive analytical rationality.

Both of the agencies chosen here as examples of the hierarchical model are lodged in the United States Department of Agriculture (USDA). USDA is so decentralized that in the words of one USDA employee, it is “like a supermarket.” As a practical matter, the individual services within USDA are relatively autonomous. All documents that are published in the Federal Register, however, must receive clearance for legal sufficiency from the centralized Office of the General Counsel. USDA has a large staff of agricultural economists in the Economic Research Service, who assemble data relevant to the agricultural economy and make projections about future crop yields, prices, and exports. Although the Service plays a fairly small role in preparing regulatory analyses, informal discussions between its economists and regulatory analysts in the agencies occur frequently.

For many decades worker exposure to toxic substances in the workplace has taken a heavy toll. Despite the efforts of the Occupational Safety and Health Administration (OSHA) to force employers to control workplace exposure to toxic chemicals, workers still played a guinea pig role for the rest of society. As science yielded more information about chemical risks, workers began to insist that they be apprised of the risks they faced at work.

Without information about risks, workers lack the freedom to make informed career choices. They also undervalue their services in wage negotiations, because they fail to attach a risk premium to unsafe workplaces. On the other hand, evaluating and communicating risks is quite expensive, and employers are naturally reluctant to spend money informing employees when the predictable consequence will be attempts by employees to bid up wages.

Regulatory background

In 1974, the National Institute for Occupational Safety and Health (OSHA's sister agency in the Department of Health and Human Services) published a “criteria document” urging OSHA to promulgate a hazard communication standard. After an OSHA-appointed advisory committee made similar recommendations, Ralph Nader's Health Research Group in 1975 petitioned the agency to issue a standard. Two years later, OSHA published an Advance Notice of Proposed Rulemaking (ANPR) discussing some of the important issues and soliciting public comment on whether it should mandate risk communication. After spending four more years studying the responses to the ANPR, OSHA published a Notice of Proposed Rulemaking for a generic hazard identification standard as one of the many “midnight regulations” issued in the waning moments of the Carter Administration.

The National Highway Traffic Safety Administration (NHTSA) was created in the mid-1960s after Ralph Nader convinced the nation that traffic safety depended as much upon the “nut on the wheel” as the “nut behind the wheel.” Although state highway safety agencies are responsible for ensuring visual acuity for the human half of the man-machine combination, NHTSA must ensure that the machine half does not unnecessarily obstruct the driver's view of the road and of approaching hazards. The long-studied, but never implemented “Field of Direct View” (FDV) standard was NHTSA's attempt to establish requirements for a minimum field of view for the driver, for light transmittance through windshields, and for the permissible size of obstructions (e.g. roof pillars and rear-view mirrors) in the driver's field of view.

Regulatory background

Shortly after its creation, a NHTSA contractor reported that automobile rear-view mirrors were a major obstruction to driver vision and that roof pillars in both the front and the rear of some automobiles created substantial “blind spots.” The agency also became concerned that windshield tinting could cause night visibility problems. The agency in April 1972 proposed a standard that would set requirements for: (1) obstructions in the driver's fields of view; (2) light transmittance levels of windshields; (3) visibility of the vehicle's corners; (4) visibility of ground surface targets for trucks, buses, and other such passenger vehicles; and (5) view obstruction by sun visors.

Comments on the proposal were “strong, critical and helpful.” Manufacturers agreed “in principle” with some of the proposals, but they disagreed with the agency and among themselves as to how visibility requirements should be achieved.

Given the impressive array of debilitating limitations explored in Chapter 9, it should not surprise the reader to discover that analysis suffers considerably in its real world application. Despite the practical problems that plagued the analytical enterprise in the case studies examined in Chapters 3 to 7, they represent success stories; the agencies threw enough resources into the endeavor to give the analytical effort a fighting chance. The vast majority of regulatory analyses do not receive the kind of detailed attention that the agencies devoted to the analyses described in those chapters. Yet, most executive agencies attempt to honor the spirit of the analysis requirements, and some have devoted serious attention to beefing up their analytical capabilities.

This chapter will explore the commitment to analysis that executive agencies typically make on paper and in the real world. Its primary focus will be on EPA, OSHA, NHTSA, FA A, and four agencies in USDA. It will examine in a general way the analytical sophistication of agency regulatory analysis documents, paying particular attention to how they go about quantifying costs and benefits, stating assumptions, and characterizing uncertainties. It will also probe the extent to which regulatory analysts attempt to identify and evaluate innovative options that may go beyond the agency's statutory authority. Finally, it will examine briefly the impact of regulatory analysis on public participation in agency decisionmaking. We shall see wide variations among agencies, and even within agencies, in the level of analysis achieved in a typical rulemaking. None of the agencies, however, comes close to meeting the comprehensive analytical rationality ideal.

One of EPA's most important functions under the Clean Air Act is to promulgate and revise National Ambient Air Quality Standards (NAAQSs) for ubiquitous air pollutants. The primary NAAQS for a pollutant specifies an ambient concentration of the pollutant (expressed as micrograms of pollutant per cubic meter of air (μg/m3)) that will protect the public health (including the health of susceptible groups such as infants, asthmatics and the elderly) with an adequate margin of safety. A secondary NAAQS is set at a level sufficient to protect the public welfare. In setting the primary standards, EPA examines the human health effects of pollutants, while secondary standards focus on effects on wildlife, visibility, crops, man-made materials, and the general ecology. States must write State Implementation Plans (SIPs) that impose emissions limitations on individual sources sufficient to meet the NAAQSs by specified deadlines.

Regulatory background

One of the very first NAAQSs was for “total suspended particulates” (TSP), a generic term that refers to any particles that stay in suspension in the air long enough to be captured by a designated measuring device. This includes a wide variety of solid and liquid particles that vary widely in size, stability and toxicity. Particulate matter from coke ovens and diesel trucks contains highly carcinogenic poly cyclic aromatic hydrocarbons; particulate matter from West Texas cotton fields is mostly composed of relatively harmless silicates. Congress in 1977 told EPA to reevaluate the TSP NAAQSs by 1980 and at five-year intervals thereafter.

The March 209 1984 Proposal

EPAs first step in revising the particulates NAAQS was to prepare a criteria document setting out the available health and welfare information about the pollutant.

Regulatory analysis is currently in a state of awkward adolescence. It has emerged from its infancy, but it has not yet matured. It is often noisy and clumsy, and it generally commands little respect. Yet, despite its considerable shortcomings, it has important virtues. Every decisionmaker wants to make rational and informed decisions, and regulatory analysis can be very useful in sorting out the pros and cons of regulatory options. Perhaps more importantly, it can encourage the decisionmaker to articulate policy preferences and demonstrate to the public how those policy preferences were applied in important rulemaking initiatives. Like most adolescents, regulatory analysis also has great potential for the future. If the public, and particularly the beneficiaries of regulation, become convinced that it is not being used cynically to reach particular substantive results, regulatory analysis can become an effective tool for improving regulatory decisions and for enhancing the accountability of the bureaucracies to the public that they serve.

The conclusions of a high-level USDA employee who had an opportunity to observe the regulatory analysis process for many years at close range may provide an apt summary. This employee is enthusiastic about the theoretical value of regulatory analysis, but pessimistic about its current efficacy in the real world. Although some agencies have a strong analytical orientation, he believes that most bureaucrats, like most other other people, are not comfortable with thinking analytically. They bring their individual experience and intuition to bear on a problem, and when they are presented with a regulatory analyst's work product, they immediately search for the bottom line before agreeing with or critiquing that analysis.

Petroleum refiners have used tetraethyl lead since 1923 as an additive to boost the octane rating of gasoline and reduce “knocking.” By the mid-1970s, approximately 90 percent of all gasoline manufactured in the United States contained tetraethyl lead. Although large gasoline refiners produced most leaded gasoline, a number of smaller companies refined leaded gasoline and supplied tetraethyl lead to large refineries.

Unfortunately, lead can be extremely toxic, especially to children, in high concentrations. At very high blood concentration levels of approximately 120 micrograms of lead per deciliter of blood (μg/dl), lead can cause severe and irreversible brain damage. Children with blood lead levels of over 80 to 100 μg/dl can suffer permanent cognitive impairment. Lead can also have reproductive effects on both men and women. At much lower blood levels of approximately 40 μg/dl, lead can impair the synthesis of hemoglobin and cause anemia.

Lead is ubiquitous in the human environment. It is present in food, water, air, soil, dustfall, paint, and other materials with which humans come into contact. Human exposure to lead can come from a variety of sources, including ingestion of food grown in leadcontaminated soils, ingestion of lead-based paint, inhalation of airborne lead, and ingestion of dust contaminated with airborne lead, the latter route being common among infants with pica, an affliction which causes them to eat dust and dirt. A major source of human exposure to lead is airborne lead, and approximately 90 percent of airborne lead comes from automobiles.

Regulatory background

Congress enacted Section 211 of the Clean Air Act specifically to empower EPA to regulate fuel additives like tetraethyl lead.

Although agencies were never free to disregard the impact of their rules on regulatees and the public, formal requirements that agencies prepare documents detailing those impacts have existed for about two decades. The idea probably originated with the National Environmental Policy Act of 1969 (NEPA). NEPA's real bite was in its requirement that agencies prepare an Environmental Impact Statement (EIS) for every proposal for legislation or other major federal action significantly affecting the quality of the human environment. The EIS was required to describe: (1) the environmental impact of the proposed action; (2) any unavoidable adverse environmental effects; (3) alternatives to the proposed action; (4) the relationship between local short-term uses of environmental resources and the maintenance and enhancement of long-term productivity; and (5) any irreversible and irretrievable commitments of resources. Although agencies initially regarded NEPA as a wasteful impediment to the attainment of their programmatic goals, the courts demanded strict adherence to NEPA's analytical requirements, and the agencies soon began to hire employees with expertise in environmental impact analysis.

As the agencies began to write NEPA into their standard operating procedures, observers noted that compliance with NEPA's analytical requirements did not ensure that agency action comported with NEPA's substantive goals. Indeed, cynics suggested that agency staff often drafted lengthy EISs in excruciating detail to ensure that busy decisionmakers would not read them and therefore would not be influenced by them. Perhaps the most cogent criticism of the environmental impact assessment process was that it presumed an unrealistic decisionmaking process “characterized by abstract rationality and focused on a single responsible decision-maker who, even if he did exist, could hardly be expected to undertake the investigation of alternatives that the Act requires.” …