Senator RIBICOFF. Elmer Staats, please?

Mr. Staats, your testimony is voluminous. It will all go in the record as if read. Senator Percy and I have read it. Copies are available to the press. I would appreciate if you would just emphasize the highlights so we will have an opportunity to ask you questions that we have prepared on the basis of your testimony. TESTIMONY OF ELMER B. STAATS, COMPTROLLER GENERAL OF THE UNITED STATES, U.S. GENERAL ACCOUNTING OFFICE, ACCOMPANIED BY ALLEN MENDELOWITZ, ASSISTANT DIRECTOR, PROGRAM ANALYSIS DIVISION; MARK NADEL, SOCIAL SCIENCE ANALYST, PROGRAM ANALYSIS DIVISION; JOHN LOVELADY, ASSISTANT DIRECTOR, GENERAL GOVERNMENT DIVISION; AND JANET LOWDEN, SUPERVISORY MANAGEMENT ANALYST, FEDERAL PERSONNEL AND COMPENSATION DIVISION

Mr. STAATS. You are quite right, Mr. Chairman. The statement is long. I do have a summary.

Chairman RIBICOFF. That is fine. We would appreciate it. We have your summary, but the entire statement will go into the record as if it was delivered in its entirety.

Mr. STAATS. Mr. Chairman, I am very happy to be here this morning to testify on both S. 262 and S. 755. I particularly want to take this opportunity to commend you for your leadership of this committee's efforts to address the problems associated with Federal regulation.

It is a subject I have been concerned with both in the executive branch as well as in my present position. I am well aware of how complex the problem is. It is becoming more and more complex and more and more difficult every day.

We agree with the observation in the committee's thorough study on Federal regulation that despite certain shortcomings, the Federal regulatory efforts have resulted in substantial improvements in the health, safety, and security of the American people.

We have a longer statement for the record, but what I will touch on today are four points: first, the location of the responsibility for supporting congressional oversight of regulatory reform; second, the need for regulatory analysis and evaluation; third, the need to improve the administrative law process; and fourth, the proposed role of the Administrative Conference of the United States.

The GAO strongly supports the general thrust of these two bills that regulatory agencies should carefully and comprehensively evaluate the effects of proposed and existing rules as has been required for executive agencies by Executive Order 12044.

Moreover, we believe that effective congressional oversight of this process is essential. However, we strongly recommend that the oversight role that S. 262 now gives to the Congressional Budget Office be assigned to the General Accounting Office. Sections 606 and 645 involve monitoring compliance with the law by regulatory agencies, evaluating the performance of those agencies and reporting findings to Congress. These are functions that Congress has already vested with the General Accounting Office by the Budget and Accounting Act of 1921, the Legislative Reorganization Act of 1970 nd the Congressional Budget Act of 1974.

Assigning this oversight role to CBO would duplicate GAO's responsibilities, would be wasteful, and would certainly prove confusing both to congressional committees and to the regulatory agencies.

GAO has extensive experience in reviewing agency compliance with legislative requirements, and we are increasing our capability in the area of program evaluation-which is, I know, a subject of longstanding interest to you.

Our work in that area, which includes a significant amount of economic analysis, now constitutes more than one-half of our work. Attached to our statement for the record is a list of 99 reports on regulation completed by the GAO in the last 3 years alone. We recommend, therefore, that the oversight responsibilities in sections 606 and 645 be explicitly assigned to the Comptroller General of the United States.

Now I would like to discuss the criteria for choosing which rules to analyze and the factors that ought to be included in the regulatory analyses. Both bills define a major rule as one that is likely to result in an effect on the economy of at least $100 million.

It is not clear why the monetary standard is set at $100 million except that this is the amount that was used in the economic impact statement program of the Ford administration and is also used in Executive Order 12044. One hundred million dollars may be too high or not high enough. Or it may be that no single dollar figure is appropriate. The purposes of the legislation may be just as well served by using qualitative standards.

If any specific dollar figure is to be used, the components of that figure need to be defined. As presently drafted, neither bill is clear about what is meant by a $100 million effect.

Does it include only direct compliance costs or does it also include indirect costs, the shift of existing costs, or the transfer of monetary income from one group or region to another? An economic effect could also be interpreted as the sum of the costs and benefits of a rule.

If you decide to retain a specific dollar criteria, we suggest that it be defined as the incremental costs of compliance to directly regulated industries or other entities. These projected compliance costs cannot be estimated precisely, but they are far easier to estimate in advance than any other specific economic effect. The ease of measurement is important because agencies should not be required to perform extensive analysis just to determine if another analysis is required.

We also believe that the purpose of this legislation will be better served by considering the differential impact of regulation as is done in the definition of a major rule in S. 755. A proposed rule that might fall far short of having a substantial impact nationally might still be of crucial importance to a small industry, a State, or a region.

Rules with such a concentrated impact should also be carefully analyzed. Therefore, we favor the more explicit language in S. 755 defining as major a proposed rule that may cause a substantial change in costs or prices for individual industries, geographic regions, or levels of government.

We support the approach implicit in S. 262 that the regulatory analyses should be more than simple cost-benefit analyses. The bill requires a detailed analysis of the projected economic effects and the projected health, safety, and other noneconomic effects.

There are qualitative benefits of regulation that reflect the values of our society. For example, how should we quantify the fear of parents for the long-term health of children who have been exposed to radiation or toxic chemicals? It is equally difficult to quantify the confidence in our financial institutions brought about by Federal regulation of banks. Providing security and peace of mind are important benefits. The fact that they are intangible does not make them any less important. Indeed, they constitute the primary objective of some Government intervention and therefore must be taken into account if the analysis is to be complete. Regulatory analyses should also focus not only on the magnitude of costs but on the distribution of these costs among different segments of the population. Many of the costs attributed to health, safety, and environmental regulation are not new but have always been incurred in various forms by different sectors of society.

For example, we as a nation have decided that firms can no longer externalize environmental costs by the free dumping of wastes in the environment. Similarly, reducing workplace hazards involves the shift of a cost of production from the worker-the expected loss from injury, illness or death-to the firm and the consuming public.

These cautionary notes on calculating economic effects are not meant to suggest that agencies should not use analysis to find the most effective and least burdensome regulatory strategy. My point here is that estimating the impacts of regulation is far from a precise science. We urge that attention be paid to the difficulty of estimating the costs and benefits of various alternatives. Because of these difficulties, it is important that agencies clearly state the assumptions on which their estimates are based.

Just as the projected effects of proposed regulations should be analyzed, the current effects of existing rules should also be evaluated in light of experience and changing circumstances. We have long supported the need for agencies to evaluate their own policies and programs. This is just as applicable to regulatory programs as to any other.

Both bills require continuing evaluation of past regulations, but they differ considerably in their standards for choosing matters to evaluate. We believe that the requirements in S. 755 are too rigid and burdensome and that the objective of requiring evaluation can better be met with the more flexible selection criteria in S. 262. Subjecting every major existing rule to a formal analysis on a 10year schedule may be undesirable because of changing circumstances. It may be too frequent or not frequent enough depending on the rule.

This committee is also considering S. 2, the sunset bill, which sets forth special procedures for regulatory agencies. It is important that the regulatory reform legislation be coordinated with the sunset process and in our statement for the record we offer suggestions on how this can be accomplished.

In imposing greater analytic requirements on regulatory agencies, it is important to recognize that this process is not costless. We have not seen any convincing hard numbers but have received estimates that the required regulatory analyses could cost up to a quarter of a million dollars for major rules.

It is better for the Federal Government to absorb these costs rather than imposing the greater costs of poorly formulated regulation on the economy, but the burden on the agencies should also be understood. Congress should be prepared to provide the added resources that may be necessary. Paradoxically, decreasing the costs of regulation to society may require larger regulatory agency budgets.

Similarly, GAO's role would involve the commitment of substantial staff resources and would require the authorization of additional staff by the Congress.

Both of these bills make extensive changes in administrative procedures and in the selection, evaluation and responsibilities of administrative law judges. As the committee is aware, last year we issued a report on the role of administrative law judges. We have just delivered today to you, Mr. Chairman, a followup report on that study.

Based on that work we support the provisions of the bills that clarify the power of the agencies to expedite the administrative proceedings and which give greater finality to the decisions of the administrative law judges.

In our statement for the record we offer more detailed comments on the changes in personnel procedures for administrative law judges. In brief, we support increasing the number of candidates referred to agencies for selection as administrative law judges. We also support assigning responsibility for performance appraisal to an organization outside the agencies, with the qualification that the agencies should retain a role in the appraisal process. To some degree, this is the case already in some agencies, and it can be done without endangering the administrative law judges' independence in making decisions.

Both bills assign responsibility for evaluation of administrative law judge to the Administrative Conference and S. 755 also assigns the Administrative Conference responsibility for administrative law judge recruitment. These functions, particularly the responsibility of the administrative law judge recruitment process, are far beyond the current mission of the Administrative Conference, which is basically a small research organization.

We recommend that responsibility for initial screening of administrative law judge candidates remain with the Office of Personnel Management in order to avoid wasteful duplication. If, however, Congress wishes to designate the Administrative Conference as the organization responsible for recruitment and/or evaluation of administrative law judges, it will be necessary to restructure and increase the resources of the Administrative Conference as contemplated by S. 262.

Currently the size of the staff and its research orientation would make it impossible for the Administrative Conference to accomplish the administrative law judge personnel responsibilities set forth in the administration bill.

Now I would like to discuss an issue raised by proposed subsection 593(4) of S. 262. That issue is the fragmentation of control over Federal information-gathering activities.

S. 262 provides that the Administrative Conference shall issue guidelines with respect to reducing paperwork and monitor compliance with such guidelines. The guidelines are to be consistent with the objectives of the act which established the Commission on Federal Paperwork. I was a member of that Commission, by the way.

This provision would further fragment control over Federal information-gathering activities which is already scattered among four Federal organizations. We believe these controls should be consolidated, preferably in OMB. Therefore, we would suggest that subsection 593(4) be dropped.

The objectives set forth in the act establishing the Commission on Federal Paperwork were to assure that Federal agencies and the public receive necessary and useful information, to guarantee appropriate standards of confidentiality, and to reduce duplication and other unnecessary burdens of Federal paperwork require

ments.

Responsibility for achieving these objectives is consistent with OMB's Federal Reports Act responsibilities for controlling the paperwork burdens on the public and is closely related to the Department of Commerce's responsibility for setting statistical policy with regard to information collected by the Federal Government.

We believe that progress toward achieving these aims is hampered because central management responsibility of the Governments' statistical and paperwork control activities is fragmented among four organizations-the Office of Management and Budget, the General Accounting Office, the Department of Commerce, and the Department of Health, Education, and Welfare.

Difficulties we have experienced in administering the review functions required by the Pipeline Act amendment support our position for consolidated, central management of the Federal Reports Act responsibilities. These difficulties are a result of ambiguities in both the original and amended Federal Reports Act and unclear jurisdictional lines between GAO and OMB.

Although there are several options for consolidating and restructuring Federal statistical policy and paperwork controls, we strongly favor consolidation within OMB. The Paperwork and Red Tape Reduction Act of 1979, H.R. 3570, was introduced in the House of Representatives by Congressmen Horton, Brooks, Steed, and Preyer, to reconsolidate the paperwork and statistical policy activities in an Office of Federal Information Management Policy in OMB.

The new office would be structured along the lines of the Office of Federal Procurement Policy. The bill would also amend the Federal Reports Act, strengthening and clarifying the authority of the central control agency, as well as creating a central locator system to aid in eliminating the collection of duplicative information. We have worked closely with Congressman Horton and Senator Chiles and Senator Bentsen, I might add, on the Senate side, in developing this bill. We hope that a similar bill will soon be introduced in the Senate. We believe it will be.