"Regulatory Daubert": A Proposal to Enhance Judicial Review of Agency Science by Incorporating Daubert Principles into Administrative Law
Raul, Alan Charles, Dwyer, Julie Zampa, Law and Contemporary Problems
In the landmark decision of Daubert v. Merrell Dow Pharmaceuticals, Inc., (1) the United States Supreme Court empowered federal judges to reject irrelevant or unreliable scientific evidence. Daubert has given the judiciary a mandate to foster "good science" in the courtroom and to reject expert testimony not grounded in scientific methods and procedures. Federal agencies--in particular, the Environmental Protection Agency ("EPA")--have been widely criticized for lacking a commitment to sound science. Too often, federal courts have upheld agency decisions that are based on faulty scientific evidence or unsupported assumptions and conclusions. The courts' responsibility to engage in meaningful judicial review of agency action, together with Congress's recent directive calling for agencies to issue data quality guidelines, strongly points to the need for a mechanism to enable more rigorous, consistent review of agency science.
Daubert provides a suitable framework for reviewing the quality of agency science and the soundness of agency decisions consistent with the standards established for review of agency rulemakings under the Administrative Procedure Act ("APA"). (2) While the Daubert line of cases arises in connection with the admissibility of evidence in litigation governed by the Federal Rules of Evidence, the same "good science" rationale should also apply to judicial review of the science underlying regulatory decisionmaking. Indeed, if private litigants are entitled to rules requiring sound science to protect parochial interests, certainly the public should be equally assured that good science is the foundation for national action. The regulatory science used to justify agency decisions that commit society's public resources and allocate social priorities should be no less rigorous than the litigation science that is currently tested according to the methods and procedures prescribed in Daubert.
To date, two courts have specifically rejected the application of Daubert in cases governed by the APA, (3) and no court has directly accepted the proposition that Daubert principles should apply in the context of administrative law. In 1999, however, a D.C. Circuit judge on the panel that decided American Trucking Associations, Inc. v. EPA cited Daubert favorably in his dissenting opinion. (4) In Daubert itself, the Supreme Court cited and relied on a scholarly book by a former EPA official discussing regulatory science. (5) One would therefore imagine that the Supreme Court would be receptive to extending Daubert principles to review of federal agency science. Indeed, the criteria for judicial review of agency action established in Motor Vehicle Manufacturers Ass'n v. State Farm Mutual Auto Insurance Co. (6) and its progeny under the APA are substantively consistent with the "relevance" and "reliability" themes of Daubert.
This Article argues that Daubert principles should apply to the review of agency rulemaking under the APA because these principles are consistent with the APA requirement that agencies engage in reasoned decisionmaking, would assure better documentation of agencies' scientific decisions, and would enhance the rigor and predictability of judicial review of agency action based on scientific evidence. Judges would evaluate the scientific methods and procedures employed by agencies but would not substitute their own policy preferences or conclusions for those chosen by the agencies. The fundamental goal of "regulatory Daubert" is, quite simply, to encourage reviewing judges to be less deferential, and thus more probing, of agency science and related administrative justifications for regulatory action.
Incorporating Daubert principles into administrative law would improve agency decisionmaking and enhance accountability. Agencies would be compelled to identify the most reliable and relevant scientific evidence for the issue at hand and disclose the default assumptions, policy choices, and factual uncertainties therein. …