the science of risk assessment has improved dramatically in recent years, and there is a growing consensus about its capabilities and how they can be prudently used to look at potential carcinogenic and noncarcinogenic health threats. The net result of these developments is that risk assessment today must take full advantage of the best data and methodologies to produce the most accurate, and not just the most conservative explanation of public health threats posed by a particular substance, so that risk management decisions can be properly informed.

Thompson, "Federal Risk Management Policy: Where Are The Problems?" in RiskBased Decision Making in Water Resources VI 17 (Yacov Y. Haimes et al. eds. 1994) (emphasis in original). See also U.S. Dept. Energy, Choices in Risk Assessment: The Role of Science Policy in the Environmental Risk Management Process 247 (1994) ("Risk assessment is a valuable tool through which regulators can gauge the existence and severity of potential risks to human health and the environment * * * [R]isk assessment based on science policy can frame the debate about whether particular potential risks should be regulated and who should bear the costs of regulations."). Similarly, even those experts who recognize that the estimates involved in cost-benefit analysis must be made within a range of uncertainty also conclude that "evaluating the costs and benefits of proposed regulatory changes" is "critical to developing sound regulatory policy." Hahn & Hird, supra, at 259. The value of costbenefit analysis lies in the fact that:

[i]t can help decision makers and the public examine the advantages and disadvantages of regulatory options. It also can help decision makers make rational and informed decisions, although it cannot fully inform or precisely point to rational conclusions. Perhaps more importantly, it can encourage the decision maker to articulate policy preferences and demonstrate to the public how those policy preferences were applied in important rule-making initiatives.

McGarity, Regulatory Analysis and Regulatory Reform, 65 Tex. L. Rev. 1243, 1330 (1987).

The cost-benefit analysis provisions of S. 343 build and improve on the work done by the Senate in S. 1080. Several new features of the cost-benefit analysis in S. 343 warrant particular mention. First, the bill includes in the cost-benefit analysis consideration of market-based mechanisms as an alternative to command-and-control regulation. Proposed 5 U.S.C. §622(c)(2)(D). Second, the bill includes a section providing explicit "decisional criteria" governing the impact of the cost-benefit analysis on the promulgation of a regulation. In a related step for-ward, the bill requires an agency to prepare a cost-benefit analysis even where the enabling statute prohibits consideration of these decisional criteria. The agency is required to submit its final cost-benefit analysis in such a context to Congress, thereby allowing both Congress and the public to become aware of the costs and benefits of regulations promulgated under such an enabling act, and so possibly consider modifications to it.

One troubling feature of the cost-benefit analysis provisions of S. 343 is its apparent provision for interlocutory judicial review of the determination of whether a rule is a major rule, and so subject to cost-benefit analysis. See Proposed 5 U.S.C. § 624(b). I question whether interlocutory review of that determination makes sense. As provided in the bill, this interlocutory review must be initiated not later than 30 days after the publication of the determination that a rule is not major and that determination can be reversed only "upon a showing of clear and convincing evidence that the determination or decision not to designate is erroneous in light of the information available to the agency at the time the determination or decision not to designate was made." §624(b)(2). These provisions in practice would appear to be either unnecessarily disruptive or relatively meaningless. If the ultimate review of a rule includes review of the decision that the rule was not major, and if a conclusion by the court that the rule was in fact major can reverse the whole rulemaking, it would seem that interlocutory review of that decision is not necessary. At the same time, the standard for review of the major rule decision provided in S. 343 seems impossibly high with its requirement of clear and convincing evidence that the decision was erroneous in light of the information available to the agency at the time.

Finally, proposed § 625 takes the excellent step of allowing private parties to petition an agency or the President to perform a cost-benefit analysis of an existing rule.

S. 343-THE LEGISLATIVE VETO

Proposed 5 U.S.C. §626(b) provides for a legislative veto of major rules. The legislative veto device has a long pedigree, being first introduced in 1932 and subsequently employed by Congress in a wide variety of statutes. See Breyer, The Legislative Veto After Chadha, 72 Geo. L.J. 785, 786 (1984); Bruff & Gellhorn, Congressional Control of Administrative Regulation: A Study of Legislative Vetoes 90 Harv. L. Rev. 1369, 1371(1977). Up until the Supreme Court's decision in INS v. Chadha, 462 U.S. 919 (1983), both the constitutionality and the substantive merits of the legislative veto were the subject of debate. With the Court's ruling in Chadha, the constitutional question is settled, and any legislative veto of administrative action must conform to the Constitution's provisions governing legislation, that is, the veto must be passed by both Houses of Congress and presented to the President for his approval or veto. It is clear that the legislative veto provisions included in S. 343 conform to these constitutional requirements and so should raise no concern on that

score.

On the merits, the legislative veto is obviously designed to be a device to enhance agency accountability. As Justice Breyer wrote when he was on the First Circuit Court of Appeals, "The regulatory veto's focus, however, is upon a still different need for compromise, a need arising out of the classic conflict in the administrative state between political accountability and the necessary complexity of regulatory decision making. This complexity forces broad statutory delegations of power." Breyer, supra, 72 Geo. L.J. at 787-88.

Indeed, as an accountability device, the legislative veto has several principled advantages over judicial review. Obviously, the legislative veto, which relies on the actions of elected representatives of the people to assure agency accountability, is a more democratic device than relying on unelected judges for the same task. This is a particularly important point when one recognizes that rulemaking and other regulatory actions essentially embody the choices of government officials among reasonable alternatives, but choices that are ultimately shaped by the policy preferences of those officials. Clearly, review of policy-based choices by government officials is a matter not for the courts, but for elected representatives of the people, whether members of Congress or the President.

In addition, since administrative agencies act pursuant to congressionally-delegated power, Congress is, at least in a theoretical sense, the most likely entity to evaluate whether an agency is properly exercising the power that has been given to it. To be sure, this advantage also points to a possible, though probably remote, danger in the legislative veto device. A Congress that is considering a veto resolution of a particular regulation is unlikely to be exactly the same Congress that passed the enabling statute pursuant to which that regulation is being promulgated. If the regulation under consideration exceeds the agency's authority under its enabling statute, but the current Congress approves of the regulation, and so fails to veto it, has that Congress in effect amended the enabling statute? If so, the actual text of the enabling statute remains unchanged, but that text no longer conveys the law's true meaning. In such a situation, how is anyone to know what the law governing that agency really is? The problem becomes even more complicated when one recognizes that a subsequent Congress may itself not agree with the Congress that failed to veto the regulation in the first instance.

Of course, this concern may very well be merely hypothetical in view of the balance provided by judicial review, under which an agency is supposed to be held accountable to the terms of its enabling act as it is written. But this concern illustrates why other proposals for congressional oversight, such as Justice Breyer's suggestion "that Congress could * ** condition[] the legal effect of exercises of delegated authority on subsequent enactment of a confirmatory statute," are very problematic. See Breyer, supra, at 793. Indeed, Justice Breyer recognized this problem by also proposing that "each special confirmatory law include a clause rendering it ineffective unless the administrative action that initiated it would have been a valid exercise of the delegated authority, absent the requirement of a confirmatory statute." Id. at 795. As a legislative veto, the proposal made in S. 343 is unlikely to suffer from this problem.

Finally, I question whether this legislative veto should be limited to major rules. The analytical requirements proposed in this bill for major rules are obviously burdensome to an agency, and adding a legislative veto to the procedures affecting a major rule obviously creates a practical disincentive to declare that a particular rule is major. In addition, the determination of whether a rule is major remains in the hands of an agency or the President, and so limits congressional control over this device. Finally, it is very possible that many rules that are not technically "major" within the meaning of this legislation would be controversial or otherwise of interest

to members of Congress, and so appropriately within the scope of any legislative veto provision.

UPDATING THE INFORMAL RULEMAKING PROCESS

One subject that S. 343 does not address is the informal rulemaking procedure under the APA. By contrast, the Senate's earlier action, in S. 1080, made significant changes in these provisions in the APA and were among the least controversial subjects addressed in that bill. Because I believe that improvements to the informal rulemaking process are in many respects the most practical, if mundane, elements of sound regulatory reform, I would strongly recommend that the Subcommittee include in this legislation many of the changes to informal rulemaking set out in S. 1080.

In its work on S. 1080, the Judiciary Committee surveyed the various reasons why the APA rulemaking provisions need to be updated. See S. Rep. No. 284, supra, at 93-106. Those reasons for reform remain valid today and I will not reiterate them here. Indeed, the need for reform of the rulemaking process drove some courts to read new procedures into the APA on their own initiative. This judicial revision of administrative procedure, led in large part by the D.C. Circuit, was halted by the Supreme Court's quite correct conclusion in Vermont Yankee Nuclear Power Corp. v. Natural Resources Defense Council, Inc., 435 U.S. 519 (1978), that power to amend administrative procedure properly belongs to Congress, not the courts. And Congress has in a piecemeal fashion acted on this responsibility by supplementing the rulemaking procedures in specific enabling acts. See S. Rep. No. 284, supra, at 103-104. Yet such haphazard changes to administrative procedure make no sense. As then-professor Scalia observed over 20 years ago, "I fear that repeated Congressional attention to administrative procedure as a mere subsidiary issue in the context of more important substantive controversies will lead to an administrative process that is pointlessly diverse and frequently unsound.” 1973-74 Report, Administrative Conference of the United States 2 (1974).

In 1981, the Judiciary Committee concluded that "bare bones notice-and-comment procedures are no longer sufficient for resolution of the many complex, technical and factual issues arising in recent rule makings." S. Rep. No. 284, supra, at 106. If anything, this conclusion is more true today than it was in the early 1980's. It is absolutely essentially that the Senate not miss the opportunity of the popular support for regulatory reform to expand these rulemaking opportunity provisions of the APA. Reform of the rulemaking procedures will improve the reasoned decision making of an agency, give members of the public a better opportunity to interact with agency decision makers in formulating rules, and, by more clearly setting out the analysis and choices made by an agency, improve the quality of judicial review. Accordingly, the notice of a proposed rulemaking should include a fuller explanation of what the agency hopes to accomplish with the proposed rule, the authority under which the agency is acting, an identification of critical technical material on which the agency may be relying, an explanation of whether the proposed rule is a major rule, and if so, a summary of the draft cost-benefit analysis required for such a rule. When the final rule is published, the required statement of basis and purpose should be expanded to include a discussion of the issues raised by comments on the proposed rule, including a description of alternatives to the rule that had been considered, and an explanation of how any factual conclusions upon which the rule is based are supported in the record. These ideas were embodied in S. 1080, and I would recommend that similar provisions be included in this legislation.

MAKING JUDICIAL REVIEW OF AGENCY ACTION MORE RATIONAL

The existing statutory scheme for judicial review of agency action operates to cause three related, but conceptually distinct, problems for persons harmed or otherwise aggrieved by unlawful federal regulation:

(1) it is often impossible for such persons to obtain from the courts a full and complete remedy for the harms they have suffered due to the unlawful agency action; (2) even were it possible for such persons to obtain full relief, such relief often cannot be obtained from a single court or through a single lawsuit; and (3) the complexity of the current statutory scheme often forces persons bringing lawsuits challenging unlawful agency action to spend time and resources litigating over a threshold question-jurisdiction-before the merits of the claim are even reached. These problems result from the complicated interplay between the primary statutes governing judicial review of agency action-the APA, 5 U.S.C. §§ 702-706, and the Tucker Act, 28 U.S.C. § 1491-as well as from the large gaps left in the remedial scheme by these two statutes.

The APA waives the sovereign immunity of the federal government so as to allow judicial review of agency action in certain contexts and in actions seeking certain types of relief. 5 U.S.C. §§ 702, 704. Yet the APA does not allow a district court to provide complete relief to a litigant who has been harmed by unlawful agency action and who has appropriately invoked the APA's waiver of sovereign immunity. Most significantly in this regard, the APA does not allow claims for "money damages" against an agency, 5 U.S.C. §702, though the courts have differed on whether this proviso precludes any monetary award at all. In Bowen v. Massachusetts, 487 U.S. 79 (1988), the Supreme Court held that the APA's reference to "money damages" was intended to draw a distinction between "compensatory" and "specific" remedies, not a distinction between monetary and non-monetary remedies. Applying this distinction between compensatory and specific relief, the Court held that the relief that Massachusetts sought-i.e., reimbursement of monies that it claimed it was entitled to under the Medicaid statute constituted specific relief rather than compensatory relief. Id. at 910.

But the Supreme Court's decision in Bowen v. Massachusetts has not removed all of the uncertainty over whether a request for monetary relief in a particular case amounts to "specific" relief or "compensatory" relief. See. e.g., Transohio Savings Bank v. Director, OTS, 967 F.2d 598 (D.C. Cir. 1992). Compare Hubbard v. Administrator, EPA, 982 F.2d 531 (D.C. Cir. 1992) (en banc) (holding that back-pay award to unsuccessful applicant for federal job is compensatory relief and not specific relief); Ulmet v. United States, 888 F.2d 1028 (4th Cir. 1989) (holding that back-pay award is specific relief). Moreover, whatever the precise parameters of the "money damages" proviso, Bowen v. Massachusetts did not alter the fact that the APA does not waive sovereign immunity for most district court suits seeking compensation for harm suffered as a result of unlawful agency action. The most the APA offers in such a case is an injunction designed to stop the unlawful agency action.

Jurisdictional questions under the APA are further complicated by two other, somewhat ambiguous, provisions in that statute. First, section 702 provides that the APA does not "confer[] authority to grant relief if any other statute that grants consent to suit expressly or impliedly forbids the relief which is sought.” 5 U.S.C. §702. In addition, section 704 precludes APA review of "final agency action for which there is [an] * * * adequate remedy in" another court. 5 U.S.C. §704. The parameters of these two further limitations on judicial review under the APA are unclear, which has resulted in even more litigation on the threshold issue of jurisdiction. Compare Bowen v. Massachusetts, 487 U.S. at 904-08 (discussing scope of "adequate remedy" provision of section 704 and concluding Claims Court action would not be an adequate remedy in that case) with id. at 922–30 (Scalia, J., dissenting) (reaching opposite conclusion).

The Tucker Act, which defines the jurisdiction of the Court of Federal Claims, fills in some of the gaps in the remedial scheme left by the APA judicial review provisions. However, the Tucker Act does not fill in all of the gaps, and has some further limitations of its own that make a Court of Federal Claims lawsuit under the Tucker Act less than a complete remedy for persons harmed by unlawful agency action. To begin with, with certain minor exceptions, the Court of Federal Claims does not have the power under the Tucker Act to award injunctive or other prospective equitable relief. See. e.g., United States v. King, 395 U.S. 1, 4 (1969); Security Savings & Loan Ass'n v. United States, 26 Cl. Ct. 1000, 1003 (1992).1 Thus, in most cases the Court of Federal Claims is limited to awarding monetary relief.

Perhaps even more significantly, the Supreme Court and the Court of Federal Claims have made clear in interpreting this provision that not all non-contractual claims involving the application of federal regulatory, statutory, or constitutional provisions are cognizable under the Tucker Act. Thus, the Supreme Court's decision in United States v. Mitchell, 463 U.S. 206, 216-17 (1983), contains the following discussion summarizing prior case law with respect to these points:

It ✶✶✶ remains true that the Tucker Act does not create any substantive right enforceable against the United States for money damages. A substantive right must be found in some other source of law, such as "the Constitution, or any act of Congress, or any regulation of an executive department." 28 U.S.C. § 1491. Not every claim invoking the Constitution, a federal statute, or a regulation is cognizable under the Tucker Act. The claim must be one for money damages against the United States, and the claimant must demonstrate that the source of substantive law that he relies

1 The Tucker Act does contain an exception allowing the Court of Federal Claims to grant injunctive and other equitable relief in order to "afford complete relief on any contract claim brought before the contract is awarded." 28 U.S.C. § 1491(a)(3).

upon can fairly be interpreted as mandating compensation by the federal government for the damage sustained.

(citations and internal quotations omitted). See also United States v. Testan, 424 U.S. 392 (1976); Eastport S.S. Corp. v. United States, 372 F.2d 1002, 1008 (Ct. Cl. 1967).

Whether a regulatory, statutory, or constitutional provision can "fairly" be interpreted to "mandate" compensation in the event of its violation is an issue that is itself not free of difficulty, and one that has therefore generated a great deal of litigation. Compare Bowen, 47 U.S. at 905 n.42 (majority opinion expressing doubt whether claim at issue would be cognizable under Tucker Act) with id. at 923-25 (Scalia, J., dissenting) (opining that claim would be cognizable under Tucker Act). In any event, Tucker Act noncontractual jurisdiction has been interpreted rather narrowly, and the Tucker Act thus does not allow most suits seeking damages for harm incurred due to unlawful agency action. In addition, the Court of Federal Claims and its predecessors have made clear that not every constitutional violation can be remedied through a Tucker Act suit for damages. Thus, to take one example, the Court of Federal Claims and its predecessors have held that the Fifth Amendment due process clause is not a provision which mandates the payment of compensation for its violation. See, e.g., Froudi v. United States, 22 Cl. Ct. 290, 296, recons. denied, 22 Cl. Ct. 647 (1991); Wright v. United States, 20 Cl. Ct. 416, 420 (1990); Castillo Morales v. United States, 19 Cl. Ct. 342, 345 (1990); Alabama Hosp. Ass'n v. United States, 656 F.2d 606, 609 (Ct. Cl. 1981), cert. denied, 456 U.S. 943 (1982).

As a result of the limitations of the APA and Tucker Act judicial review provisions, persons who have suffered economic injuries or other harm to their legitimate business or property interests due to unlawful agency regulation or other action may find it impossible to obtain complete relief from the federal courts. While such a person may be able to obtain injunctive relief, under the APA, to stop the unlawful agency action, the APA's "money damages" proviso prevents a district court from also awarding the person damages to compensate the person for the injuries suffered during the period before the issuance of the injunction. In addition, the requirement that the substantive provision at issue be "fairly" interpreted to "mandate compensation" for its violation before Tucker Act jurisdiction will obtain will most likely preclude such a person from being able to recover such damages through a Court of Federal Claims lawsuit. As a result, such a person is left without a full and complete remedy for the harm caused by the illegal actions of the federal agency.

In addition, even where the aggrieved person can, as a legal matter, obtain both injunctive relief and damages from the federal courts, it is likely that the person cannot obtain both forms of relief from the same court or through the same lawsuit. Because a district court cannot, in an APA lawsuit, award relief that amounts to "money damages," and because the Court of Federal Claims, in a Tucker Act lawsuit, cannot award injunctive relief, a person who desires to both put a halt to the unlawful agency action and to obtain compensation for the harm he has suffered before the action was halted must file two suits-one in the district court and one in the Court of Federal Claims.

To add insult to injury, even those persons who attempt to obtain full relief by seeking both injunctive and monetary relief are often prevented, under current law, from maintaining concurrent lawsuits in both a district court and the Court of Federal Claims. This is because another statute, 28 U.S.C. § 1500, provides that the "United States Court of Federal Claims shall not have jurisdiction of any claim for or in respect to which the plaintiff or his assignee has pending in any other court any suit or process against the United States * * *" This provision has been broadly interpreted to preclude Court of Federal Claims jurisdiction over any claim based on the same operative facts as a claim also pending in a district court, regardless of whether the separate suits are based on different legal theories or seek different types of relief. See. e.g., UNR Industries v. United States, 962 F.2d 1013 (Fed. Cir. 1992) (in banc), aff'd sub nom. Keene Corp. v. United States, 113 S. Ct. 2035 (1993). Thus, this statute operates to force persons who have been aggrieved by unlawful agency action and who desire to obtain a full and complete federal court remedy not only to file two lawsuits in order to obtain such a remedy, but to file them sequentially rather than concurrently. Indeed, it is more than a theoretical possibility that these jurisdictional complexities may require a litigant to shuttle back and forth between district court and the Court of Federal Claims. The statute thus places another needless obstacle in the path of full relief, and, at the very least, forces parties