| 摘要 | In theory, Congress created regulatory agencies and gave them power because it had neither the time nor the expertise to do the regulating itself. But many of these agencies now find themselves lacking the necessary time and even expertise. Not only have their workloads increased dramatically, but the substantive issues addressed by many of them have become exceedingly complex. Decisions in such controversial areas as nuclear power, environmental protection, public health, and transportation often rely on scientific facts and interpretation. As a result, regulatory officials are turning to technical experts in an effort to arrive at scientific “truth” before taking action.
Obtaining and processing the views of experts has proved to be a vexing problem for most regulatory agencies. Under the current system, these views are often the product of an adversary process that is not particularly well-suited for the examination of complex scientific or technical questions. First, the experts (usually paid by principals in the case) may make biased presentations—though one hopes that this is rare. Second, in areas where there are legitimate differences of opinion among experts, the policy-maker is likely to be distracted by the range of opinion expressed and fail to comprehend the preponderance of that opinion. I have often heard regulators exclaim in despair words to the effect that “expert A says one thing and expert B says just the opposite: how are we to judge?” Third—a related point—the actual differences of opinion among the experts may be small, but because of the way in which the testimony is taken the areas of agreement may not be grasped and the areas of disagreement may be overemphasized.
An important proposal to deal with some of these problems has been advanced by Arthur Kantrowitz, a physicist who heads a private research organization. To make regulation more efficient, Dr. Kantrowitz recommends that, before an agency addresses an issue where technical information will be important to the decision, a panel of experts be assembled to debate the technical questions and to draw conclusions concerning the probable effects of the major policy options. The experts constituting this “science court” would not be there to say whether specific proposals should or should not be carried out; rather, their purpose would be to determine the likely results of alternative regulatory policies. These panels would not recommend, for example, that a particular nuclear power plant should or should not be built, but would confine themselves to estimating such a plant’s probable risks to nearby inhabitants, probable power output, probable operating costs, and so on.
The science court proposal has been the subject of considerable debate and some support in the scientific community. In August 1976 a task force of the Presidential Advisory Group on Anticipated Advances in Science and Technology made an interim report recommending that science court experiments be initiated at a few regulatory agencies. To my knowledge, however, no agency has acted on this recommendation.
A more modest proposal is advanced here—namely, that once agency officials have identified the critical technical questions, experts (not attorneys) be used to cross-examine other experts in adversary proceedings. (This is not to say that no attorneys are experts—for, indeed, many of them are.) This arrangement could have several advantages over the present system. First, the quality and objectivity of analysis might improve. As Gordon Tullock noted in The Organization of Inquiry, the strongest incentive for scientists to maintain high standards is peer group pressure. Second, both agreements and legitimate disagreements of fact and interpretation might be identified more quickly and bounded more succinctly than they are now. The point that economists tend to disagree but can very quickly determine their areas of disagreement is undoubtedly applicable to other professions—which suggests that regulatory proceedings might become less time-consuming if the proposed arrangement were adopted. Third, what the experts say might be of more use to decision-makers if it reached them more directly, without being as extensively filtered through attorney-intermediaries.
POLICIES FOR EXAMINING WITNESSES AT TWENTY-TWO MAJOR FEDERAL REGULATORY AGENCIES
Source: Code of Federal Regulations, and discussions with officials of regulatory agencies.
As shown in the table, the degree to which experts are now involved in direct cross-examination of other experts varies from one regulatory agency to another. Over one-half of the agencies listed prohibit experts (not including agency officials and staff, and formal representatives of parties to the proceedings) from cross-examining other experts. And though the practice is permitted by some agencies, only the International Trade Commission employs it frequently. Thus, a situation in which one expert cross-examines another is clearly the exception. Part of the explanation for this may be agency rules, but part would also appear to be the common reluctance of attorneys who represent parties to allow the direct participation of nonattorneys in agency proceedings (often for good reason).
The diversity of regulated areas and administrative procedures makes it difficult to assess the impact of existing cross-examination policies—as well as the policy proposed here—on the overall quality of decision-making. Moreover, the proposed arrangement would be no panacea: there is always the danger that experts would end up debating esoteric issues having little relevance to the policy issues at stake in the proceeding. Nonetheless, the idea seems worth pursuing. Retaining the basic format of the adversary process while allowing greater interaction among technical experts might lead to a significant improvement in the basis for regulatory decision-making at negligible cost. ■
The author is co-director of the Center for the Study of Government Regulation at the American Enterprise Institute. He wishes to acknowledge contributions by his research assistant, George Vurdelja, and helpful comments from John Barnum, George Eads, and Marvin Kosters. Responsibility for errors and opinion is, of course, the author’s. |
