Junk science is not just the domain of civilian litigants. Government agencies are no strangers to it. Unfortunately, courts, Congress and the public too often let it slide.
In fact, traditional standards of proof for judicial review are so lenient, and courts can be so highly deferential to agency science, it is a wonder that the regulatory process ever hits its target at all. A recent Columbia Law Review article by Professor Wendy E. Wagner brilliantly analyzes the junk that regulatory agencies (sometimes) pass off as science. In her article, titled "The Science Charade in Toxic Risk Regulation,"[1] Professor Wagner posits that the failures of regulatory science are at least partly attributable "to a pervasive ?science charade,' where agencies exaggerate the contributions made by science in setting toxic standards in order to avoid accountability for the underlying policy decision."[2]
I agree. My comments below are intended as an argument in favor of importing the Supreme Court's decision in Daubert [3] into administrative law. Daubert reasoning should apply when federal judges review final agency actions under the Administrative Procedure Act (APA), and other statutes that authorize federal agencies to regulate the environment, health or safety.
Before you get mad that someone wants to mix the laws of evidence with principles of administrative law, let me explain why red meat litigators as well as APA geeks should care about applying Daubert in the regulatory arena.
First, we're all citizens, and subject to government authority. When agencies issue environmental safety and health regulations, they are exercising delegated legislative powers from Congress. If agency bureaucrats undergird their rules with junk science, Congress gets a pretty shabby return on its delegation investment. Moreover, the public's right to the accountability of government officials is eroded. This is bad for democracy. It also constrains the country's economic performance.
The second reason real litigators ought to worry about judicial review under the APA is that regulatory decisions based on regulatory science are often precursors to meat and potatoes civil litigation. There is a close nexus between the regulatory judgments of the Environmental Protection Agency, Food and Drug Administration and the Occupational Safety and Health Administration and tort litigation involving asbestos, dioxin, environmental tobacco smoke, breast implants, etc. So-called "endocrine disruptors," now all the rage in Washington, D.C., will provide the next wave of tort actions. Remember the one word of career advice passed along to Dustin Hoffman in the movie The Graduate -- "PLASTICS." Well, in the sequel, the new career advice gets translated as "ENDOCRINE DISRUPTORS."
My point is not to suggest that there are no bona fide scientific issues with endocrine disruptors -- isolated reports of low sperm counts and small penises in alligators could be harbingers of more serious problems. Instead, my point is that government agencies have a substantial responsibility to Congress and the public to do a better job of getting their science right. I believe the courts can help more than they do now. Daubert may be a useful tool in fundamentally changing how judges look at scientific issues. Before proceeding, let me provide a little more background on judicial review under the APA. This will help explain why regulatory science is fertile ground for becoming a junk yard.
Real litigators are used to standards of proof like "beyond a reasonable doubt," "clear and convincing evidence," or at least, " a preponderance of the evidence" (i.e., better than 50-50). Frankly, plaintiffs' lawyers have it easy. Administrative law geeks don't beat an agency in court unless the plaintiffs prove the government's decision or action was either "arbitrary or capricious," or, for some claims where Congress decided to cut the private bar some slack, not supported by "substantial evidence" in the record. This high litigation hurdle applies to industry and public advocacy plaintiffs alike.
Under either standard, however -- "arbitrary and capricious" or "substantial evidence" -- the challenging party does not prevail if the best she can do is show the agency was wrong. A plaintiff must show that the agency was "really, really wrong." This is not the technical term, but it comes pretty close to describing the effective standard of proof in a lot of APA cases.
Another point to bear in mind is that the pro-agency standards of proof are applied on top of other judicial doctrines that grant -- properly in most cases -- considerable deference to the facts found, assumptions made and decisions reached by agency regulators.
I submit that the scientific "gate keeping" function of Daubert should also influence APA litigation. Transplanting this doctrine into administrative law should help diminish the excessive deference too many courts now accord agency science. Agency science is no more or less hallowed than anything else the government does. It deserves judicial scrutiny commensurate with the importance of the scientific and regulatory issues at stake.
Judge Leventhal articulated this argument superbly in his concurring opinion in the D.C. Circuit's 1976 decision in Ethyl Corp. v. EPA.[4]
Congress has been willing to delegate its legislative powers broadly -- and courts have upheld such delegation -- because there is court review to assure that the agency exercises the delegated power within statutory limits, and that it fleshes out objectives within those limits by an administration that is not irrational or discriminatory. Nor is that envisioned judicial role ephemeral, as Overton Park makes clear.
Our present system of reviewing assumes judges will acquire whatever technical knowledge is necessary as background for decision of the legal questions. It may be that some judges are not initially equipped for this role, just as they may not be technically equipped initially to decide issues of obviousness and infringement in patent cases. If technical difficulties loom large, Congress may push to establish specialized courts. Thus far, it has proceeded on the assumption that we can both have the important values secured by generalist judges and rely on them to acquire whatever technical background is necessary.[5]Scientific gate keeping under Daubert can fulfill Judge Leventhal's vision. More robust judicial gate keeping will deter agencies from improperly exercising delegated authority and imposing binding legal norms on the public based on: exaggeration or diminution of risks; failure to disclose material uncertainties or weakness in the available data; reliance on hidden policy agendas; and, adoption of overly conservative risk models or models that do not apply to the facts at hand.
Simply stated, courts need to do a more aggressive job of ferreting out administrative scientific failures. You might ask at this point, "my goodness, this is our government, could they really be so messed up?" Even if the answer were yes, it would not mean that politics and ideology had joined profit and greed as motives to fudge science.
We must acknowledge that lawyers, litigation and judicial review are partially responsible for the pressure on agencies to dress up uncertainty and equivocation as research and analysis in order to prevail in court. This is most unfortunate. Agencies deserve to be upheld when: (a) they admit publicly what they know and what they do not know about the science; (b) they regulate as best they can on the basis of disclosed policy judgments; and (c) Congress has authorized the extent of discretion being exercised.
Given my argument that Daubert should spill over to diminish judicial deference to agency science -- that is, bring it back from its current Olympian heights to a more work-a-day status -- I should highlight a few "high deference" cases to provide a flavor of the status quo.
In Color Pigment Mfrs. Ass'n v. OSHA, [6] the Eleventh Circuit held that OSHA had presented substantial evidence to justify including cadmium pigments in its regulatory standard governing occupational exposure to cadmium. Not even the following concatenation of contingencies could deter the court from deferring to the agency's decision to regulate.
No definitive study confirming or refuting the hypothesis that cadmium pigments are equally as toxic and carcinogenic as other cadmium compounds has been presented to us. Given the absence of definiteness on the issue, the volume of evidence that points at least implicitly to the dangers of cadmium pigments, and the serious potential health risks present if cadmium exposure is as great in pigment form as in other compounds, we believe that OSHA was justified in choosing to include cadmium pigments in the [standard], despite the existence of an equally rational alternative.[7]In NRDC v. EPA [8], citizens groups unsuccessfully challenged EPA's approval of water quality standards for dioxin set by Maryland and Virginia under the Clean Water Act ("CWA"). EPA relied on toxicological studies in reviewing the standards set by Maryland to determine that they were consistent with the CWA and adequately protective of human health. The U.S. District Court for the Eastern District of Virginia noted that not only is APA "arbitrary and capricious" review "highly deferential" in itself, but "where the information at issue is highly scientific in nature, the basis of the agency action is entitled to even greater deference."[9]
In Huls America, Inc. v. Administrator, EPA,[10] the D.C. Circuit went even further. It upheld EPA's decision to continue listing isophorone diisocyanate ("IPDI") as an extremely hazardous substances ("EHS") under section 302 of the Emergency Planning and Community Right to Know Act ("EPCRA"). Appellant Huls argued, among other things, that EPA used artificially high concentrations of IPDI in the animal tests it used to reach its conclusion that IPDI should be listed, and that the other characteristics of IPDI -- non-volatility, non-flammability, low reactivity with water -- make it a low overall risk. Nonetheless, the court upheld EPA's use of high dose animal tests and the sole criterion listing standard, stating, "we will give an extreme degree of deference to the agency when it is evaluating scientific data within its technical expertise."
It would not be accurate, of course, to leave the impression that agencies today always win regardless of what they do or how badly they do it. This is certainly not true. The D.C. Circuit, in particular, may be returning to the "probing scrutiny" advocated by Judge Leventhal in the 1976 Ethyl Corp. decision. The Supreme Court, however, is still more deferential than the D.C. Circuit. Perhaps less so, however, than when it issued the landmark Chevron decision in 1984. In any case, I particularly enjoyed Judge Silberman's analysis and rhetoric in Chemical Mfrs. Ass'n v. EPA.[11]
In CMA v. EPA, the D.C. Circuit vacated EPA rule's designating methylene diphenyl diisocyanate ("MDI") a high risk pollutant under the Clean Air Act. To determine the likely level of exposure to a hazardous air pollutant, the agency developed a generic air dispersion model, from which it predicted the concentration of a hypothetical air pollutant emitted from a typical industrial facility under average meteorologic conditions. The air dispersion model rested on the assumptions that the pollutant would be a gas emitted at 20 degrees Celsius from a single 10-meter stack and would disperse as a gas through the atmosphere. Petitioner CMA presented evidence, uncontested by the agency, that MDI is a solid at 20 degrees Celsius, does not tend to evaporate, and disperses as an aerosol.
Striking down the rule, the court stated:
It necessarily follows that in designing the model the agency must have broad discretion to make simplifying assumptions . . . If we are to earn our keep, however, judicial deference to the agency's modeling cannot be utterly boundless.
. . . [T]he EPA has not pointed to any record evidence that shows a rational relationship between the generic air dispersion model and the physical properties of MDI, even in the face of the specific scientific evidence to the contrary adduced by the CMA...
When the CMA presented specific objections to the EPA's application of its model to MDI, the EPA responded in a high-handed and conclusory manner. For example . . . the EPA accepted the CMA's point that MDI is a solid at the ambient temperature . . . at which the model assumes that the generic pollutant is a gas, . . . yet it dismissed the CMA's objection . . . with the facile assertion that ?[i]t is likely that MDI is emitted at temperatures higher than ambient and so would disperse much like any other pollutant" . . . . (In fact, the record shows that MDI is still a solid at 37·C). This is . . . [a] ?speculative factual assertion' . . . [a]nd, we may add, . . . bespeaks a ?let them eat cake' attitude that ill-becomes an administrative agency whose obligation to the public it serves is discharged if only it avoids being arbitrary and capricious. [12]
Thank heavens the D.C. Circuit will not let an agency get away with regulating a gas when the substance is really a solid.
What can Daubert do to shift administrative law away from the "extreme deference" to agency science promised in Huls America, and hold agencies more accountable for their "let them eat cake" attitude? I suggest that importing Daubert could advance the following salutary policy objectives.
- Change judicial attitudes in favor of delving somewhat more deeply into the science underlying important regulatory decisions.
- Enhance development of more consistent criteria for judges reviewing agency science.
- Compel agencies to distinguish hidden policy choices and assumptions from actual science.
- Allow more effective examination of an agencies's assumptions.
- Allow more thorough evaluation of the relevant uncertainties.
- Facilitate the ability of Congress and the public to determine whether regulatory decisions are based on true agency expertise and congressionally authorized judgment, or not.
Bringing Daubert into administrative law would empower judges to exercise a gate keeping responsibility to bring the curtain down on the science charade that too often gets by as agency science. Unfortunately, the one court that has been invited to apply Daubert under the APA, Sierra Club v. Marita,[13] declined to do so. It opined that Daubert was not sufficiently deferential to the agency's science. However, Daubert's criteria are most simpatico with the Supreme Court's judicial review criteria for APA litigation as set out in Motor Vehicle Mfrs. Ass'n v. State Farm Mut. Auto Ins. Co.[14] So there is some hope future APA decisions will not banish Daubert.
If you are still queasy about the notion of exposing federal agencies to greater scientific scrutiny, remember, when real litigators win a case, they succeed in assigning or avoiding liability, and obtaining or fending off money damages. When administrative law geeks win big, all we get is a remand. The Agency always gets to try and try again.
References
- 95 Column L. Rev. 1613 (1995).
- Id at 1617.
- Daubert v. Merrell Dow Pharmaceuticals, 509 U.S. 579 (1993) (determining admissibility of scientific evidence under Federal Rules of Evidence).
- 541 F.2d 1 (D.C. Cir.) (en banc), cert. denied, 426 U.S. 941 (1976).
- Id. at 68-69.
- 16 F.3d 1157 (11th Cir. 1994).
- 16 F.3d at 1161 (emphasis added).
- 806 F. Supp. 1263 (E.D. Va. 1992), aff'd 16 F.3d 1395 (4th Cir. 1993).
- 806 F. Supp. at 1272 (emphasis added).
- No. 95-5282 (D.C. Cir. May 10, 1996).
- 28 F.3d 1259 (D.C. Cir. 1994) (emphasis added).
- 28 F.3d at 1264-65 (emphasis added).
- 7th Cir. 1995.
- 463 U.S. 29 (1983).
Mr. Raul practices administrative and environmental law with Beveridge & Diamond, P.C. in Washington, D.C. He previously served as General Counsel of the Office of Management and Budget, General Counsel of USDA, and Associate Counsel to President Reagan. This article is adapted from remarks delivered by Mr. Raul on August 5, 1996 to the Sections of Science and Technology, Litigation, SONREEL, and Tort and Insurance Practice at the American Bar Association annual meeting in Orlando.
Material presented on this home page constitutes opinion of the author.
Copyright © 1996 Steven J. Milloy. All rights reserved. Site developed and hosted by WestLake Solutions, Inc.