the agencies guiding them on how to comply.' And the experience with agency flouting of the Regulatory Flexibility Act, 5 U.S.C. §§ 601 et seq., until 1996 when Congress added judicial review of agency compliance with its mandate, 110 Stat. 857, warns of the futility of giving the agencies directions without ensuring their adherence to them. On the other hand, Congress has wisely moved cautiously in adding judicially enforceable burdens on the agencies. For example, when it overwhelming adopted the Unfunded Mandates Reform Act of 1995, 2 U.S.C. §§ 1531 et seq., which similarly sought to force agencies to take account of the implications of their regulations on state and local governments (and private entities), it limited judicial review. That is, the Act made no provision for judicial review of the requirements that an agency select the "least costly, most cost-effective or least burdensome alternative that achieves the objectives of the rule" unless it explains why another choice is superior or is dictated by law. See Daniel E. Troy, The Unfunded Mandates Reform Act of 1995, 49 Admin. L. Rev. 139 (1997). Its provision for judicial review of agency compliance with notice, consultation and impact analysis requirements was restricted to their inclusion in the administrative record for consideration by the reviewing court in determining whether the agency's final action was arbitrary or capricious.

Our experience with this more middle ground of judicial review – somewhere between zero and full judicial review -- is too limited to assess its value. But it suggests that until we

1 I still lament the decision by Congress in 1995 to defund the Administrative Conference of the United States. This small agency, whose annual cost was less than $2 million per year, provided exactly that kind of guidance to agencies and significantly improved the administrative process. (Its legislative mandate was not repealed so it could be reestablished at any time simply by appropriating funds for its support. See 5 U.S.C. §§ 59196.)

know more, we might follow this limited approach as adopted in the Unfunded Mandates Reform Act and as proposed in the Regulatory Improvement Act of 1998 (S. 981). That is, I recommend that S. 1214 be revised as follows: (1) that the federalism analysis be made part of the rulemaking record for purposes of review of a final rule; (2) that judicial review of the rule under the arbitrary and capricious test (of 5 U.S.C. § 706(2)(A)) be limited to the procedures used and to the content of the impact analysis; and (3) that remand or invalidation be required if the agency wholly fails to consult with state/local governments or to perform a federalism impact analysis. However, other aspects of this federalism mandate -- e.g., submission of the federalism assessment to OMB -- should be exempted from judicial review.

Review of Analytical Requirements.

While I support the concept of a federalism impact analysis for agency rulemaking, I am concerned that too many analytical requirements may be burdening the rulemaking process. Thus, I also urge that this Committee study the possible consolidation of the numerous analytical requirements. There are plenty to review. First, there are the recent requirements imposed by the Paperwork Reduction Act of 1995, the Unfunded Mandates Reform Act of 1995 and Small Business Regulatory Enforcement Fairness Act of 1996. Then there are more general laws such as NEPA as well as agency-specific statutes (e.g., Clean Air Act) which establish additional, sometimes overlapping analytical requirements. And finally there are the Executive Orders which, in addition to the existing Federalism order, include: Regulatory Planning and Review (E.O. 12,866), Civil Justice Reform (E.O. 12,988), Family Values (E.O. 12606) and Indian Tribal Governments (E.O. 13,084). We need, in other words, an impact analysis of various impact requirements in order to assess which are working, which can be eliminated, and which can be refined and/or consolidated. Incorporating all into one statute would ease agency administration and provide simpler instruction to affected parties. And perhaps we would not have to revisit old issues (such as judicial review) again and again.

Testimony of Caleb E. Nelson

Associate Professor of Law

University of Virginia School of Law

Thank you for the opportunity to speak with you today about federal preemption of state

law. My testimony will focus on the preemptive effects of federal statutes, but I would be happy to address questions about other types of preemption too. My views should not be attributed to my employer, the University of Virginia School of Law; I offer them in my personal capacity.

Preemption is an important topic. The extent to which federal statutes displace state law affects both the substantive legal rules under which we live and the distribution of authority between the states and the federal government. The sheer volume of litigation about preemption reflects the doctrine's significance; according to Professor Stephen Gardbaum, preemption "is almost certainly the most frequently used doctrine of constitutional law in practice."'

Although the Supremacy Clause of the federal Constitution is the reason that valid federal statutes displace contrary state law, the preemptive scope of any particular federal statute is a maner of statutory interpretation. The Supreme Court recognizes this point; assuming that Congress has acted within its constitutional powers, the Court says that the question in preemption cases "is basically one of congressional intent."? Unfortunately, the rules of construction that the courts apply in preemption cases do not match up very well with the realities of the legislative process.

In this testimony, I will first discuss some general flaws with the courts' rules of construction, focusing particularly on the Supreme Court's tests for "implied" preemption. I will then offer some concrete examples of cases in which the results of those tests seem, in the words of S. 1214, to be "contrary to or beyond the intent of Congress." I will close by discussing the advantages of substituting new rules of construction for the ones that the Court currently applies.

-2

Flaws with the Rules of Construction that the Court Currently Applies in Preemption Cases

To determine the preemptive effect of any particular federal statute, courts need to interpret the statute and decide what legal rules it establishes. Those legal rules might be either "substantive" rules (which tend to regulate primary conduct) or "jurisdictional" rules (which tend to say that states may not regulate primary conduct, or at least may not regulate primary conduct in certain ways).' But however one classifies a particular rule, the rule—if it is within Congress's constitutional power to establish—will displace whatever state law it contradicts. The Supremacy Clause says as much.

Federal statutes, of course, can establish rules by implication as well as expressly. In principle, this is no less true of "jurisdictional" rules than of "substantive" rules. It is possible, in other words, for a federal statute to imply a preemption clause, just as it is possible for a federal statute to establish a substantive rule that it fails to state in so many words.

For a clear example, consider the federal Ports and Waterways Safety Act of 1972, which was at issue in the Supreme Court case of Ruy v. Atlantic Richfield Co. Title I of the Act authorized the Secretary of Transportation to promulgate safety rules for both vessel traffic and shore structures, and a savings clause declared that nothing in the title prevented individual states "from prescribing for structures only' higher... safety standards than those which may be prescribed pursuant to this title." In the overall context of the title, this provision implied that states could nor supplement the safety standards that the Secretary promulgated for vessel traffic.

To acknowledge that a federal statute can imply a preemption clause, however, is not to suggest that this inference should be drawn too quickly. The rules of construction that the Supreme Court has articulated for preemption cases err in that direction; they risk making judges

too quick to infer broad preemption clauses.

-3

In the past, the Court suggested that the mere comprehensiveness of a federal statute could imply that Congress intended to preempt all state regulation of a particular field." The Court now recognizes that modern federal legislation in complex fields is likely to be detailed whether or not Congress wants to preempt supplementary state legislation, and so-called "field preemption" accordingly has been on the decline.' But much of the work that used to be done under the rubric of "field preemption" can still be done under the rubric of what the Court calls "conflict preemption."

..

Suppose that a federal statute does not contain an express preemption clause. The statute unquestionably will still have preemptive effects; assuming that the Constitution gives Congress

the

power to enact it, the statute will displace whatever state law it contradicts. But the Supreme Court's current doctrine of "conflict preemption" goes farther. The Court says that "[i]n the absence of explicit statutory language signaling an intent to pre-empt, we infer such intent where ...the state law stands as an obstacle to the accomplishment and execution of congressional objectives." Under this rule of statutory interpretation, federal statutes reflect an "implicit[] preemptive intent" whenever state law gets in the way of their "full purposes and objectives.""" In effect. then, the Court reads all federal statutes that do not expressly address preemption as if they contained the following implied preemption clause: "No state may enact or enforce any law or policy, of whatever type, to the extent that such state law or policy stands as an obstacle to the accomplishment of the full purposes and objectives behind this statute."

The Court sees this test as a means of giving effect to congressional intent. If such a

preemption clause were actually to appear in a proposed bill, however, I suspect that many