6.

because it comes from a source with an interest in the outcome of the rulemaking. Agencies should reduce the impact of bias by:

(i) consulting, whenever possible, multiple sources of information in preparing regulatory analysis documents;

(ii) carefully citing in regulatory analysis documents all information upon which the analysis draws, and making the information available for public scrutiny at convenient times and places;

(iii) actively soliciting comment and criticism from acknowledged experts in the fields that the documents address.

e. Retrospective

Assessments of Previous Analyses. Agencies should regularly per for m retrospective assessments of the predictions made previously in regulatory analysis documents. Retrospective analysis can provide information on the accuracy of past agency predictions and thereby enable an agency to increase the accuracy of future predictions or make judgments about the value of regulatory analysis to its regulatory effort.

Use of Consultants in Preparing Regulatory Analysis
Documents

Agencies can benefit from entering into consulting contracts with qualified experts to aid in gathering and analyzing information for regulatory analysis documents. However, agency personnel should retain the ultimate responsibility for the contents of regulatory analysis documents and guard against consultant conflict of interest. To these ends, agencies should ensure that: (1) agency employees, not consultants, draft regulatory analysis documents, and (2) when a regulatory analysis document relies upon consultant reports, the reports are placed in the public file of the rulemaking proceeding, even if the Freedom of Information Act's exemption for intra-agency memoranda, 5 U.S.C. $552(b)(5), might apply to portions of the reports.

7.

The Scope and Limits of Regulatory Analysis

a.

Cost-benefit analysis is an effective tool for marshalling and analyzing information and for establishing regulatory priorities.

b. Other analytical techniques, such as cost-effectiveness analysis and multi-objective analysis, are also useful for rulemaking that involves health, environmental, historical, artistic, and aesthetic considerations for which markets do not exist.

c. Agency rulemaking decisions must take into account the limits of the agency's statutory authority and its overall policy goals, as well as the limits of the methods and data used in the regulatory analysis.

d. The same criteria should be used in granting exemptions from regulatory analysis requirements, irrespective of whether the proceeding has been commenced to formulate new rules or to amend or repeal existing rules.

Informed observers generally agree that the rulemaking process has become both increasingly less effective and more time-consuming. The Administrative Procedure Act does not reflect many of the current realities of rulemaking. The APA's cumbersome "formal rulemaking procedures are rarely used except in some adjudicative-type rate proceedings. Meanwhile, the APA's simple "informal rulemaking" procedures (set forth in 5 U.S.C. §553) have been overlain with an increasing number of constraints: outside constraints imposed by Congress, the President, and the courts, and internal constraints arising from increasingly complex agency management of the rulemaking process. As a result, many federal agencies, faced with unsatisfactory rulemaking accomplishments in recent years, have turned to alternatives such as less formal policy statements or adjudicative orders to achieve regulatory compliance.2

The Conference believes that the environment for agency legislative rulemaking can be improved. This recommendation sets out a coordinated framework of proposals aimed at promoting efficient and effective rulemaking by addressing constraints on the current process that derive from a variety of sources. We present an integrated approach for improving the rulemaking environment in order to relieve agencies of unnecessary pressures and disincentives relating to rulemaking. We also identify desirable revisions of section 553 relating to legislative rulemaking. In doing so, this recommendation both presents new proposals and incorporates previous Conference recommendations.

Presidential Constraints

We continue to support presidential coordination of agency policymaking as beneficial and necessary. 3 We are concerned, however, that, unless properly focused, this additional review may impose unnecessary costs. All recent presidents have undertaken some level of review and coordination of agency rulemaking. Presidential review of rules, as undertaken under various executive orders applied by the Office of Management and Budget and other White House entities, has often required agencies to submit nearly all proposed and final rules to a review process in which the rules are screened and analyzed for consistency with presidential objectives. Some of these objectives have been incorporated into analytical

See generally McGanty. Some Thoughts on "Deossifying the Rulemaking Process, 41 Duke L. J. 1385 (1991). 2See Conference Recommendation 92-2, Agency Policy Statements, 1 CFR § 305.92-2 (1993), which distinguished "legislative" rules, normally promulgated through notice-and-comment procedures, from interpretive rules and policy statements, which are exempt from such procedures The present recommendation addresses legislative rulemaking.

3See Conference Recommendation 88-9, "Presidential Review of Agency Rulemaking 1 CFR § 305.88-9 (1993) (applying Presidential oversight to both executive branch and independent agencies).

requirements found in separate executive orders. This screening process can unduly slow the entire system of rulemaking; it can inhibit the growth of the promising consensus-based alternative of negotiated rulemaking; and it can create undesirable tensions between the reviewing entities and agency policymakers. While these analytical emphases can be rationalized individually, in the aggregate, they can result in redundant requirements, boilerplate-laden documents, circumvention, delays, and clutter in the Federal Register. Although specific presidential review policies have varied among Administrations, these recommendations set forth principles that the Conference believes generally should govern presidential review of rules.

We therefore recommend that presidential oversight and review be reserved for the most important rules and that the agencies be given clear policy guidance in a directive, approved by the President, specifying what is required. In addition, the reviewing or oversight entity should avoid, to the extent possible, extensive delays in the rulemaking process. The review process itself should be open to public scrutiny-following guidelines previously developed by the Administrative Conference The President's policy should encourage planning and coordination of regulatory initiatives, and early dialogue between agencies and the reviewing entity. To this end, the concept of a unified agenda of regulations is a useful tool and should be preserved. We also believe that additional non-APA analytical requirements should be kept to a minimum. The cumulative impact of such requirements on the rulemaking process should be considered before existing requirements are continued or additional ones imposed. We also believe it is useful to periodically reassess the continued viability and relevance of the various presidential directives."

Legislative Constraints

Congress should similarly review and rationalize legislatively-mandated rulemaking procedures. Specifically, we recommend that it refrain, as it generally has done since the 1970s, from imposing program-specific rulemaking requirements that go beyond the APA's basic notice-and-comment procedures. Statutory "on-the-record" and "hybrid" rulemaking provisions that require adjudicative factfinding techniques such as cross-examination, or more stringent provisions for judicial review (in particular, use of the "substantial evidence" test instead of the normal "arbitrary and capricious" test), can be unnecessarily burdensome or confusing and should be repealed. Although additional procedures can sometimes be beneficial-see, e.g., §307 of the Clean Air Act (providing additional safeguards for rulemaking with significant economic and competitive effects)10--they should be imposed only after careful review and attention by Congress to possible unintended consequences. Otherwise, such additions generally should be left to the discretion of individual agencies.11

Similarly, legislatively-imposed time limits on rulemaking, while understandable, can be unrealistic. resulting in either hastily-imposed rules or missed deadlines that undermine respect for the rulemaking process. 12 Legislative deadlines backed by statutory or regulatory "hammers" (mandating, for example,

*Among the mandates reflected in these executive orders are requirements that agency rulemakers include costbenefit estimates and analyses of the proposed and final rule's impact on federalism, family values, and future litigation, of whether it effects a "regulatory taking." and of other matters. The Conference of course takes no position on the merits of the values underlying these executive orders.

$See Conference Recommendations 82-4 and 85-5. "Procedures for Negotiating Proposed Regulations, 1 CFR §§ 305.82-4, 305.85-5 (1993);" Negotiated Rulemaking Act of 1990, 5 U.S.C §§561-69.

"See Conference Recommendation 88-9, "Presidential Review of Agency Rulemaking,* 1 CFR § 305.88-9 (1993) at 14.

"While the most recent executive order of presidential review of rules generally reflects the views set forth in this recommendation, see Executive Order 12866, 58 Fed. Reg. 51735 (1993), the Conference takes no position on the specifics of that order.

See Conference Recommendation 76-3, "Procedures in Addition to Notice and the Opportunity to Comment in Informal Rulemaking," 1 CFR § 305.76-3 (1993).

9See Conference Recommendation 80-1, Trade Regulation Rulemaking Under the Magnuson-Moss Warranty Federal Trade Commission Improvement Act, 1 CFR § 305.80-1 (1993).

1042 U.S.C. §7607.

11 See Conference Recommendation 76-3, "Procedures in Addition to Notice and the Opportunity for Comment in Informal Rulemaking," 1 CFR § 305.76-3 (1993).

12See Conference Recommendation 78-3, "Time Limits on Agency Action, 1 CFR § 305.78-3 (1993).

that the proposed rule or some other policy change13 automatically take effect upon expiration of the deadline) are particularly undesirable and often counter-productive; 14 they are generally less desirable than the alternative of judicial enforcement of deadlines. 15

Finally, legislation ancillary to the APA that creates additional rulemaking impediments should be reconsidered. Statutes such as the Regulatory Flexibility Act, which requires a special analysis of virtually all rules' effects on small business, may have laudable intentions, but their requirements are often both too broadly applicable and not sufficiently effective in achieving their goals. If such requirements are imposed, Congress should focus them more narrowly, by, for example, confining their application to significant rules or particular categories of rules.

Judicial Constraints

Other constraints on rulemaking that warrant similar reconsideration have been imposed through judicial review. The APA, in section 706, provides that agency rules may be set aside if they are "arbitrary or capricious," represent an "abuse of discretion," or are "otherwise not in accordance with law. The evolving scope of judicial review of agency rules, along with the timing of much such review at the preenforcement stage, has contributed to what is sometimes an overly intrusive inquiry. This, in turn, has led agencies to take defensive measures against such review. While some tension is an inevitable adjunct of the process of judicial review, we believe that steps can be taken to lessen some of the burdens without loss of effective outside scrutiny of agency rules.

The tendency of some courts to require extra-APA procedures in rulemaking was arrested by the Supreme Court's Vermont Yankee decision in 1978.16 Nevertheless, while the prevailing judicial interpretation of the arbitrary-and-capricious standard of review (which became known as the "hard look doctrine") has promoted reasoned decisionmaking, courts have not infrequently remanded rules on the basis of an agency's failure to respond adequately to comments, consider relevant factors, or explain fully the bases for its rule. Courts should be sensitive not to require greater justification for rules than necessary; a reasoned statement that explains the basis and purpose of the rule and addresses significant issues raised in public comments should be adequate.

Preenforcement review, expanded by the Supreme Court in the 1967 Abbott Laboratories cases, 17 endorsed by the Conference in various recommendations, 18 and codified in numerous rulemaking programs, has the virtue of settling legal issues early and definitively. When overused, however,

preenforcement review can have the negative effect of inducing precautionary challenges to most rules and the raising of as many objections to a rule as possible, including somewhat speculative challenges pertaining to the rule's potential application.

Under the Abbott Laboratories standard, challenges to a rule are permitted where issues are appropriate for judicial review and where the impact on a challenger is direct and immediate. The Conference believes that the Abbon Laboratories standard strikes a sensible balance, and that preenforcement challenges generally are appropriate where the administrative record provides a sufficient basis for the court to resolve the issues before it. Thus, a preenforcement challenge to a rule based on the procedures used in the rulemaking should normally be permitted. Preenforcement review that involves a facial challenge to a rule's substantive validity (whether because of a conflict with a statute or the Constitution, or because of the inadequacy of the facts or reasoning on which it is based) should also

13 See, e.g., Conference Recommendation 90-8, "Rulemaking and Policymaking in the Medicaid Program, I CFR 305.90-8 (1993).

14Where the "hammer" applied because of a failure to meet a deadline is that a proposed rule becomes effective, the anomalous result is that a policy that has withstood no public aining will be implemented.

15 Courts should continue, where appropnate, to consider whether agency action in a rulemaking is "unreasonably delayed. See 5 U.S.C. §706(1); Telecommunications Research and Action Center v FCC, 750 F.2d 70, 80 (D.C. Cir. 1984).

16Vermont Yankee Nuclear Power Corp. v. NRDC, 435 U.S. 519 (1978).

17 Abbott Laboratories v. Gardner, 387 U.S. 136 (1967); Toilet Goods Ass'n v. Gardner, 387 U.S. 158 (1967). 18 See Conference Recommendation 74-4, "Preenforcement Judicial Review of Rules of General Applicability," 1 CFR § 305.74-4 (1993); Conference Recommendation 91-5, "Facilitating the Use of Rulemaking by the National Labor Relations Board," 1 CFR § 305.91-5 (1993).

35-770 96-2