work subject to periodic review. This is an important and sensitive human problem. I would urge the committee to obtain the views of the Administrative Conference. For myself, I believe that periodic review of the quality of work is warranted and should be embodied in your bills.

Senator RIBICOFF. There have been several suggestions to require agencies to provide advance notice of proposals and to allow an opportunity for public comment before issuance of the actual notice of proposed rulemaking.

Is this a good idea? Or would it just add another layer of procedure to the rulemaking process?

Judge FRIENDLY. I think, clearly, it would just add another layer of procedure to the rulemaking process.

The whole genius of section 553 of the Administrative Procedures Act is to give an opportunity for comment before the agency has acted. I can't see why there is any need of inserting another period for comment before it even gives notice of the proposal. It would just add another element of delay without accomplishing anything. Senator RIBICOFF. Judge Friendly, on pages 5 and 6 of your statement, you said you have no serious objection to the requirement of the new section 553(C) that in promulgating the rule, the agency shall respond to the significant issues raised by the comments.

But then you say you wonder whether the game is worth the candle.

Can you tell us what you mean by that?

Judge FRIENDLY. I am glad you asked that, Senator, because I think it is not clear.

I really meant two things.

One is that if you are not amending section 553 in any other respect, I wondered whether this was important enough to do, since, in practice, almost all of the agencies follow this course, anyway. And if they don't, the courts usually remand when any serious issue has been raised.

Another thought that I had, which perhaps wasn't clearly expressed, was that the more requirements that you put into a statute, the more temptation there is for people to seek judicial review. Or if they would have sought one anyway to have that as another point in their brief. If that is a legitimate objection, it should have attention, and it has been getting it.

When you put it in black letters, it encourages making the argument in cases where it doesn't have any real merit.

But I really am perfectly happy to see that amendment made. Senator RIBICOFF. Judge, you found in the Federal Maritime Commission v. New York Terminal Conference, in which the agency sought to enforce the subpena, that the proceeding was stalled in its tracks for 8 months quite needlessly, as this opinion shows. Then you went on to say, in reference to the validity of the subpena: "All of this is so self-evident that it is difficult to believe that the experienced businessmen and hard-nosed lawyers require judicial instruction."

Do you favor the legislative instruction for subpena enforcement, which S. 262 provides? Would this subpena-enforcement provision

help eliminate the needless delay that often arises in subpenaenforcement efforts?

Judge FRIENDLY. In answering that, Senator, I would like to say at the outset that here I am speaking only for myself, because I know there is a division of views within our committee.

We didn't address this problem specifically. But there was some expression about it.

Speaking for myself, I have long been in favor of speedier enforcement of agency subpenas. I see no reason why an agency should not be allowed to enforce a subpena subject, of course, to the right of the citizen if he believes the subpena is unjustified, beyond the authority of the agency, unduly broad, or harassing to go to court and attack it.

So, I do support the portions of your bill that do that.

The part that is not so clear is the portion that gives the administrative law judge the powers that district judges have to impose other sanctions, which can be quite drastic.

I really would like to think that one over and get the benefit of some other views before I express my own.

But so far as the specific question that is here contained, I do favor the provision of your bill.

Senator RIBICOFF. Gentlemen, I do appreciate your giving us the benefit of your views. I have found them especially valuable and very helpful.

I want to thank you, too, for all of your cooperation with the committee staff. I hope that they and I and the committee can call upon your wisdom and experience in the few months ahead of us. Thank you very much.

Judge FRIENDLY. Thank you, Senator.

Mr. GINSBURG. Thank you, Senator.

Senator RIBICOFF. It is good to see both of you again.

Dr. Nicholas Ashford?

Our next witness is Dr. Nicholas Ashford.

Dr. Ashford is assistant director of the Center for Policy Alternatives, Massachusetts Institute of Technology. He is also professor of technology and policy at MIT's School of Engineering.

The committee is looking forward to working with you, Dr. Ashford, on a study that you have proposed, which will gather information on the benefits of regulation. This is something that hasn't been done before.

We do have increasingly better information on the regulatory costs. But the economic and other achievements of regulation haven't been a subject of careful study.

We welcome your comments on this.

STATEMENT OF DR. NICHOLAS A. ASHFORD, ASSISTANT DIRECTOR OF THE CENTER FOR POLICY ALTERNATIVES, AND PROFESSOR OF TECHNOLOGY AND POLICY, MASSACHUSETTS INSTITUTE OF TECHNOLOGY

Dr. ASHFORD. Thank you, Senator.

My name is Nicholas A. Ashford. I am associate professor of technology and policy and assistant director of the Center for Policy Alternatives at the Massachusetts Institute of Technology. I serve as a public member and chairman of the National Advisory

Committee on Occupational Safety and Health, which advises and monitors the actions of both the Department of Labor and the Department of Health, Education, and Walfare in carrying out the Occupational Safety and Health Act. I also serve on the Science Advisory Board of the Environmental Protection Agency, which serves a similar function for that agency in carrying out many pieces of legislation including the Clean Air and Water Acts, the Pesticide Act, and the newly enacted Toxic Substances Control Act. Today, I am appearing in a personal capacity and not on behalf of MIT or the advisory organizations.

I am particularly pleased to be here today to address the usefulness of regulatory impact analysis in decisions concerning health, safety, and the environment. In the last few years, I have carefully examined the records of the regulatory agencies concerned with those problems and have developed a detailed methodology for evaluating the impact of regulatory decisions under the sponsorship of the Council on Environmental Quality and the Environmental Protection Agency. I would be happy to provide you with a copy of that research.

Although I have spent much time developing a regulatory impact methodology for environmental decisionmaking, I have come today to offer words of caution about the potential misuse of analytical techniques in this area. My remarks will address both methodological flaws in the state of the art of cost-benefit analysis and the possible political misuse of the results.

In my comments today, I would like to make the following six major points.

First, there are important differences between economic regulation and environmental or safety regulation, which must not be overlooked in constructing evaluative criteria.

Second, there are serious problems in calculating both the costs and benefits of proposed regulations but the problems are different. Third, the comparison of costs and benefits in a cost-benefit framework is beset by serious difficulties. Cost-benefit analysis deceptively appears as a neutral methodology.

Fourth, it is crucial to distinguish between using cost-benefit analysis as a decision tool and using it as a decision to rule, to guide agency action or evaluate agency performance.

Fifth, it is important to realize that some regulatory reform efforts are undemocratic attempts to reorient legislative mandates in which the balance between environmental protection and economic burden has already been struck by the Congress.

Finally, it is essential to recognize the hidden dangers in the present congressional proposals to require regulatory impact analysis, even those which are motivated by a "pure heart."

Differences between economic regulation and regulation of health, safety, and the environment.

Economic regulation seeks to improve the working of the market for goods and services by encouraging competition, economic efficiency, and the diversity of available goods and services. Regulation addresses itself to this goal by attempting to insure that the price mechanism operates efficiently to allocate goods and services properly among economic sectors and between producers and consum

ers.

Economic regulation, properly carried out, thereby is expected to reduce the price of goods and services it seeks to regulate.

Health, safety, and environmental regulation attempts to ameliorate the adverse consequences of market activities-and technology in general-by reducing the attendant social costs. This regulation attempts to internalize the social costs of production by insuring that the prices of goods and services reflect the true costs to the society.

This means that prices in many cases can be expected to go up. Charles Schultze, in his now famous work with Alan Kneese entitled, "Pollution, Prices, and Public Policy," has cautioned us not to regard price rises which internalize social costs as inflation

ary.

The requirement that all price rises be viewed as inflationary by the early Presidential directive of Gerald Ford ignored the crucial distinction between economic and environmental regulation.

With economic regulation, price rises may well be inflationary and an indication that the Government's efforts need to be examined.

On the other hand, with environmental regulation, price rises may be a measure of success.

We must be careful not to use inappropriate criteria as facile indices of regulatory activity.

Environmental regulation is not really an instrument of economic policy. It is an instrument of social policy concerned with the distribution of effects that result from industrial activity.

Therefore, environmental regulation cannot be judged by economic criteria alone.

Problems with estimating the costs of regulation: It is often assumed that, because the costs of complying with regulation can be easily monetized, they are reliable estimates of true costs.

Unfortunately, there are many instances in which the costs are not only uncertain but, also, unreliable.

Agencies depend to a large extent upon industry data to derive estimates of compliance costs. I do not believe I am being too unkind in questioning the bias of those estimates.

The regulatory agencies themselves do not have access to the information concerning alternative products and processes and the resultant costs which will enable them to come up with the best estimates of the costs of compliance.

In addition, compliance cost estimates often ignore two crucial issues: One, the ability of a regulated industrial segment to learn over time to comply more cost effectively-what the management scientists call the learning curve-and, two, compliance costs based on present technological capabilities ignore the crucial role played by technological innovation, which yields benefits to both the regulated firm and to the public intended to be protected.

Indeed, environmental regulation has been called technologyforcing by the courts and by analysts.

The costs of compliance should not be based on static assump tions about the firm and its technology. Otherwise, a large overestimation will result.

Problems with estimating the health, safety, and environmental benefits of regulation: The state of the art in estimating the num

bers of cancers or cases of chronic disease prevented-or even injuries-is in its infancy.

Many health professionals believe that there is no safe exposure to a carcinogen as a result of the accepted view of the mechanisms of cancer causation.

Safe levels for chronic toxins which are not carcinogens are often derived from either acute human exposures or high-dose animal experiments. The extrapolation techniques to lower doses for chronic human exposure are imperfect. Therefore, benefit calculations for a particular maximum exposure level allowed by regulation are often not very meaningful.

Theories of accident prediction do not serve us much better. We scarcely need reminding of the unanticipated risk which attended the incident at Three Mile Island.

It is fair to say that the state of the art in benefit estimation is much less developed than the methodologies for calculating compliance costs. In addition, there is no organized interest group which systematically pursues the benefit estimations in the same way in which the costs of compliance are researched.

The tendency by analysts to rely on hard numbers places the estimation of benefits on insecure ground.

Finally, it must be realized that the benefits derived from direct regulation are only a part of the benefits that can be derived from the regulatory process. Indirect or leveraged, benefits are derived from the pressure of regulation to induce industry to deal preventively with unregulated hazards, to innovate, and to find ways to meet the public's need for a cleaner, healthier environment while maintaining industrial capacity.

To put it another way, the positive side effects accompanying regulation need to be included in a complete assessment of the effectiveness of the regulatory agency's strategies.

Problems in comparing costs and benefits within a cost-benefit framework: Even if we could accurately estimate the amount of disease or injury prevented by regulation and the compliance costs of doing so, the tasks of, one, monetizing health benefits which may accrue far into the future-or even monetizing current safety benefits from reducing accidents-and two, comparing those benefits to current compliance costs are fraught with difficulty.

A human life or a lost limb does not have an established market value. Payments to workers to assume risky occupations prior to their being injured are different than the values placed on the injured workers by their families after the injuries have occurred. Some analysts today still insist on expressing health, safety, and environmental benefits in monetary terms, although this is chang

ing.

The successor index to evaluating a change of net social welfare in dollars is the benefit-to-cost ratio for example, the number of fatalities prevented per dollar expended. The problem with this index is that it can never really be applied.

The benefits of regulation include: Deaths prevented, diseases and injuries prevented, pain and suffering prevented, hospital costs prevented, et cetera. The benefit side of the equation itself is composed of many elements of different character.