Clinton Regulation Will be "Rational"

Interview with Sally Katzen, the New OIRA Administrator

On June 29, 1993, Regulation interviewed Sally Katzen, the new administrator of the Office of Information and Regulatory Affairs (OIRA) at the Office of Management and Budget (OMB). Ms. Katzen is the first political appointee to serve in this position since 1989 and comes to office with a distinguished record in administrative and communications law. The interviewer was William Niskanen, the editor of Regulation.


Regulation: Ms. Katzen, you have been appointed the White House traffic cop on regulation in an administration that does not seem to give that role a very high priority. Is that perception correct?

Katzen: I don’t think so. In fact, I would say my selection is evidence that the administration gives that role a high priority. It was one of the positions identified early by White House personnel, by the director of OMB, and by the White House itself. While many other senior positions were considered early in the administration, I was one of the first at the subcabinet level to be identified and appointed. The administration put emphasis on getting a prompt confirmation, which we were successful in achieving. Also from the time that my appointment was announced, the White House has made it abundantly clear that this was one of the most important positions in the administration and that regulation is an important component of economic policy, which was President Clinton’s major issue during the campaign. If there is the perception that regulatory review is a low priority, I think with time it will change.

Regulation: Good. To whom do you report on the substantive issues? Where do you get your guidance on substantive issues?

Katzen: The leadership team at OMB consists of Leon Panetta, Alice Rivlin, and Philip Lader. Leon Panetta has been legitimately occupied with putting together and getting Congress to pass the economic program, but he has taken every opportunity to ask about OIRA’s issues, what I’m working on, etc. When I have wanted guidance, as I have, he has been available to me and has been very forceful in his views about the importance of OIRA’s mission. I’ve also spoken with Alice Rivlin on issues on which I know she had an interest or some past involvement. I also report to Phil Lader, the deputy director for management. He brings a fresh perspective and eminently sound judgment. He is particularly interested in those issues that are otherwise within his jurisdiction, such as personnel or procurement. He has been extraordinarily helpful on those matters. At any time, I might meet with one, two, or all three to discuss OIRA’s issues.

Regulation: For the past 12 years the vice president has had a special role in reviewing regulations and setting regulatory policy. What is expected to be the particular role of Vice President Gore in this regard?

Katzen: There has recently been a lot of discussion in the press about the vice president’s role in regulatory policy. I can say unequivocally that most of what has been said is wrong. By the time this interview is published, the executive order we are now working on, which will replace the several executive orders on regulatory management that are now in effect, will be released and that will put the misperceptions to rest.

One of President Clinton’s first acts was to announce the end of the Competitiveness Council. During the campaign he had criticized the Council for being a backdoor way to secretly affect regulations at the behest of private interests. He does not want to recreate a body that would be available or responsive to special interests. No one in the administration wants to recreate that.

At the same time, it is clear that Vice President Gore has very important experience and excellent credentials in the regulatory area, and not surprisingly the president identified that as an area in which he would like to have the vice president’s assistance. The new executive order will clarify how the vice president will assist in the regulatory area. One way will be in those instances, hopefully few and far between, when I will not be able to resolve disputes among different agencies or departments or between OIRA and an agency or department. In those infrequent instances, there will have to be some mechanism for resolving the dispute. For example, there could be two cabinet secretaries, each of whom is legitimately focused on his or her own programmatic missions, and one plans to issue regulations that could have an unintended but real adverse effect on the other’s mission. Both of the secretaries were appointed by the president, who is also the head of the executive branch and hence ultimately responsible for the actions taken. How the president will structure his decisionmaking process to resolve such a dispute is the issue that we are discussing. I think it is fairly obvious that the vice president will have a significant role in helping the president resolve such disputes.

Regulation: What are the general instructions from your administration superiors for the activities of OIRA? How do you interpret your mandate?

Katzen: Your question presumes that instructions are issued, whereas there are no formal instructions. The content of my job, the job description if you will, is something that Leon, Alice, Phil, and others in the administration all discussed with me, and those discussions produced a consensus as to what is expected of me and what I expect of myself.

With that qualification, the area that has been the subject of the most controversy in the past has been regulatory review. That function is extremely important. While the president condemned the activities of the Competitiveness Council, he confirmed his support for centralized review of federal regulations and has indicated that he expects that OIRA would be the entity that carries out that function. It is up to me to shape how we would go about doing that.

There are two other aspects of OIRA that I want to mention in response to the question. One is the paperwork reduction function involving information collection request clearances. The other is what the "I" in OIRA stands for: information—information resource management, information policy, and information technology. Some in Congress have pointed out that information has not had the visibility that it should. I hope to rectify that. So there’s a mix, actually, of three different functions for OIRA.

Regulation: On the basis of your meetings with members of Congress during your confirmation process, what are their major concerns about the activities of OIRA?

Katzen: I would summarize the major ones as follows. First, there is a concern about the amount of regulations that exists in this country—the sheer number and cumulative burden. All of them, even the traditionally more liberal senators, have heard from their constituencies that there is an enormous amount of regulation. Small businesses and state and local governmental entities that are themselves subject to federal regulations are beginning to be very vocal in their opposition. Those voices are being heard. Second, there is tremendous concern with paperwork. The classic case is the dry cleaner who works six days a week and on the seventh he fills out all the forms that the federal government requires—that or hires someone just to do the paperwork. Finally, I heard a great deal of concern about the integrity and accountability of the regulatory review process—fear, almost, that the activities of the Competitiveness Council cast the function of regulatory review in such controversy that the legitimacy of the function has been impugned. Above all I heard the importance of openness and accountability—that decisionmaking should be on the record in some sense.

Regulation: I understand that you are reviewing the several executive orders under which OIRA operates. Can you give me some sense about the status of the review and the probable outcome?

Katzen: Yes. Unfortunately, this is a quarterly magazine instead of a weekly, so this may well be public knowledge by the time of publication.

Regulation: It would be useful to explain at least the reasoning that went behind it.

Katzen: We assembled all the existing executive orders on regulatory management and have tried to create a single coherent document to replace them. We have several principal objectives. One is to restore the credibility of centralized review. Another is to establish a system that is more open and more accountable. A third is to speak to existing regulations, as well as to proposed regulations. A fourth is to send the right signals—to create the right incentives—for more and better public involvement. We can accomplish that by enabling those who expect to benefit from the regulation, as well as those who might be burdened by it, to be consulted before a proposal becomes an actual rulemaking. We would also encourage consensual-based rulemakings. A fifth objective, which is very important, is to structure a system in which potential disputes would be identified early, so that they can be resolved early.

This leads us to create—or resuscitate perhaps would be a better word—a meaningful planning process, so that agencies would be aware of the administration’s priorities, would align their plans to the extent permitted by law with those priorities, and the public would have a sense of a coherent executive branch philosophy. We hope to provide guidance to the agencies as to what is expected of them so that there will not be any surprises. We hope to create a forum in which cross-cutting issues, such as methodologies or procedures, could be discussed. The forum would consist of agency heads, or their appropriate designees, and could meet as a group or in subgroups. One such subgroup, for example, could consist of the risk-based agencies, who could work together on how risk assessment is done and whether it could be made more consistent across agencies, to the degree permitted by applicable legislation. And to the extent existing legislation precludes rationalizing risk assessment, then the subgroup could present to Congress its findings as to the state of risk assessment, and any suggestions for legislative changes that might improve the situation. Another topic for a subgroup could be small business exemptions and how different agencies respond to their particular concerns.

But whether we have a regulatory program calendar, a forum, or just consultation early and often, the notion is that we should not wait until the last minute—when the agency is about to publish its regulation—and say, "Oh, gee. We have a few ideas about how you should have structured this rule." After an agency has invested its time and effort to develop a regulation, it is difficult for the agency to change it. That is understandable. That is why we will attempt to be apprised earlier of the direction in which the agency intends to proceed and to provide guidance and assistance at that point.

Those are the five objectives that guide our thinking as we draft the replacement executive order for regulatory review.

Regulation: That leads to my next question. In 1992 the Bush administration considered issuing an executive order on risk assessment. You seem to be approaching that by setting up a forum or council to address differences in risk assessment procedures or criteria but without spelling out the procedures or criteria for risk assessment.

Katzen: I think the issue is complicated and involves a number of constituencies, and thus may not lend itself to a prescription set further than simply an executive order. Any real progress in this area has to involve the full Congress, because so much of this is driven by legislation, and it has to involve the agencies that are implementing that legislation. The way to proceed, I think, is through a forum—a council or working group—which hopefully would come to some consensus. Then it may be possible to ask the Senate Committee on Government Affairs and the House Committee on Government Operations to act as honest brokers on the Hill since they are more process oriented, or we may go directly to the substantive committees with jurisdiction over the areas which are out of sync or we think should be revised. I do not want to prejudge the outcome.

It also seems to me that part of the problem with the Bush administration’s proposed executive order on risk assessment was that it was seen less as an objective exercise in rationality than as a way to impose a preconceived ideology on the process. I am a process person. I prefer to establish a process by which we can achieve consensus of all the constituencies. While it may be more efficient to issue an edict, in a democracy it is probably more effective to have a town meeting first with the interested parties. Such a process can educate them, and us as well.

Regulation: What are the most important changes to improve the quality of scientific input to regulatory decisions? Specifically, do you support the recommendations of the Environmental Protection Agency task force in that regard?

Katzen: I am not sufficiently familiar with the specifies of the EPA’s task force recommendations to respond to the second part of your question. As to the first, I believe that it is universally accepted that decisions should be based on the best available scientific information. Unfortunately, however, here, as elsewhere, the devil is in the details, because what is one person’s best available scientific data is another person’s skewed, ideologically driven predisposition to manipulate the process. I think that the discipline of looking for data, analyzing that data, and testing the validity of that data is wholly salutary. So I support the notion of using the best reasonably available scientific, technical, and economic data. But I cannot be more specific at this time on how that should be done.

Regulation: What are the most effective ways to improve the quality of economic analysis?

Katzen: The staff of OIRA includes a number of extraordinarily able people who have worked extensively in that area. They are the beneficiaries not only of their own training but of having been at OMB, where they work with a number of different agencies and can identify the best practices and also be alerted to missteps along the way.

One of the items that I did not mention in our earlier discussion on the regulatory working group is to use that forum to think through ways in which we would provide additional meaningful guidance on how to do a cost-benefit analysis or how to identify cost-effective alternatives. I would like to use the OIRA staff as the leading edge in providing guidance in this area. I would also like to include the staff of the various agencies’ offices of policy and plans, and the chief economists of other agencies, who have the same kind of expertise. Again, we can use the regulatory working group or forum as a place where we can discuss those issues, and it might produce more standardization, if you will. I think that might be highly salutary.

Regulation: What are your views about the risk-risk type of procedure that your own staff has promoted?

Katzen: I was questioned about that during the confirmation hearing, and I have been questioned about that ever since. I have generally tried to sidestep the question.

Regulation: I can understand why.

Katzen: I believe that there is merit to the notion that one should consider the risk of taking a particular action. Identifying the risk that one incurs in complying with a particular regulation (e.g., taking asbestos out of school buildings) and the risk of not complying with it (e.g., leaving the asbestos in place) is a risk-risk analysis. Another kind of risk-risk analysis is almost a comparative analysis: if we spend money doing something, how much good will we achieve, compared with how much we might achieve from spending the same dollars elsewhere. These types of analyses are, I think, noncontroversial.

What was at the heart of the controversy you were referring to is something different—a pairing of two basic, theoretically sound propositions. One is that some regulations will impose costs on business entities which the business entities may not be able to pass through and therefore will have to absorb. The increase in operating expenses might ultimately be reflected in lesser increases in workers’ wages or benefits, or even reduced employment. The second proposition is that there is a general correlation between health and income, so that one could say that the lower the income, the more at risk is the health status. When the staff paired these two propositions, it produced a thesis that a costly regulation which was designed to improve health might in fact result in reduced health. Even though each proposition is independently supportable, the result of the combination might not be.

Apparently the OIRA staff raised this issue with the Department of Labor because there is some support in the literature for this type of analysis. But the reaction at Labor and elsewhere was extremely adverse and the whole situation is, I think, most unfortunate. Prudence would dictate a very careful analysis of the approach, and its application only in those instances, if at all, where the specific application can be independently validated.

Regulation: I happen to agree with you on the issue. I interpreted the approach as a way around legislative constraints on formal benefit-cost analysis, to try to do regulatory analysis in terms of the same metric of lives versus lives rather than money versus lives. But going through money is the way of getting to a life-life comparison, a risk-risk comparison. I understand and sympathize with the objectives of what your staff had done in this regard, but I think the better part of this is to take on the basic problem rather than what looked to me as being an end run.

You have made a substantial reputation as a communications lawyer. Would you like to share your views with us, for example, as to whether the Bell operating companies should be allowed to provide cable television or on the monopoly status of INTELSAT?

Katzen: I must for obvious reasons decline to comment on those issues. I should add that though my practice has been primarily in the area of communications law, I consider myself to be an administrative lawyer, having done substantive work in a lot of other fields as well. And what I thought made me particularly well-suited for this particular position is that I do not have a substantive agenda.

I have come to OIRA as a process person. I place great value on the integrity of the process and am concerned with how decisions are made rather than where they come out. With respect to substance, I do have views on the subjects about which you asked, but most of those issues are in the hands of the Federal Communications Commission, an independent agency—and under the existing executive orders, OIRA does not review the regulations of the independent agencies.

Regulation: Will you be part of any review process that involves major actions by the Antitrust Division of the Justice Department?

Katzen: Such as the modification of final judgment, where I also have views? I cannot tell you what I will be working on in two hours, let alone two months or two years. I have noticed that this administration is interested in using the talents of the people they have brought on board wherever they may apply, so that job titles are less meaningful here than elsewhere. I suspect that if an issue comes up on which people think I can make a contribution, they will ask for my advice. If they do, I will give it. Beyond that I cannot really say what lies ahead.

Regulation: What is your own judgment about the appropriate balance between regulation and tort law on matters that involve negative externalities? How much should we rely on regulation versus tort law? Have we moved too far in one direction or another? Are there any conditions that would lead to sorting out which activities we should handle by one process or the other?

Katzen: It is hard to answer that question generally because in some areas one approach may be more effective than another, but the opposite may be true in another instance. I am a lawyer. I believe in our adversarial system. I believe in the judicial system. I understand that tort law provides a solid basis for some allocation of risk. But I also believe that litigation is one of the least productive uses of our energies and resources, and that there are a number of instances where regulations would work more effectively and more efficiently in achieving our shared societal goals.

Regulation: Will OIRA have a role in reviewing proposals for changes in legislation, such as when the Clean Water Act comes up for reauthorization?

Katzen: OMB has a legislative referral division. When legislation involves regulations, we are included in the process. So the answer would be yes.

Regulation: Do you expect any of the major environmental reauthorization issues to come up within the next few months?

Katzen: Yes. Reauthorization of some of the major legislation is coming up in the next several years, and the administration obviously has a great interest in those issues. Several may be controversial, and I think it will be an interesting time.

Regulation: I am sure that is true. Much of the deregulation that was implemented in the Reagan and Bush years was actually started in the Carter years. Can you give me a sense of what you expect the regulatory record of the Clinton administration to be?

Katzen: Rational.

Regulation: What does that mean?

Katzen: I think that we will be making good regulatory decisions—not pro-regulation or anti-regulation, but smart regulations. I was in government during the Carter administration. I was general counsel of the Council of Wage and Price Stability when the regulatory review group was housed there. That staff ultimately became the staff of OIRA. During the Carter years, I supported the deregulation effort. It started as an effort to see whether or not (and there was no predetermined answer) existing regulation made sense—whether regulations that were on the books had costs that had not been anticipated or failed to achieve the purposes for which they were originally introduced.

The deregulatory effort of the Reagan and Bush years—particularly the Reagan years—took on a slightly different cast, at least to those of us on the outside. The Republican deregulatory effort was basically motivated by the notion that "Regulation is bad because government is bad," or maybe they would start with "Government is bad, government produces regulations, regulations are bad." That philosophy was coupled with the firmly held view that "The marketplace, not government, is the best determinant of resource allocation, etc." Having those views meant that the Republican deregulator puts weights on the scale against regulation. There was a philosophical or ideological predisposition against regulation. So there was a presumption—deeply held predisposition—that one had to overcome to get a regulation in place.

It is true that free markets—private markets—are a great engine for growth and are effective in solving all sorts of problems, if the markets operate correctly. But our markets are imperfect. Sometimes consumers have insufficient information to make the right choices. Sometimes the markets are such that one or more entities exercises market power. There are lots of different ways in which markets are less than perfect. In those cases, regulation can be highly salutary. If someone walks into the situation saying, "Good God! Regulations! They’re bad!" then I think he is skewing the decisionmaking.

When I said earlier that the Clinton approach would be "rational," I meant it would be thoughtful, in the merits of each case. It would be fair and dispassionate. The Clinton approach would be to consider all the implications and do what makes sense in the particular circumstance.

The other characteristic of the Republican deregulation movement, particularly in the Bush era, was its tendency to clean out the books like one would clean out a closet. "Go back and look at everything on your books. If you have not used this regulation in the past four years—as if you have not worn this suit in the past four years—you probably won’t need—use—wear it next year. Take it off the books—toss it out of the closet."

While that cleaning could be useful, there was an enormous amount of effort that went into the exercise that I believe was not extremely productive, particularly in cases where a regulation, once in place, has no ongoing costs. Take, for example, regulations that require changes in an assembly line. The regulation increases capital costs. But assume there is no record keeping, and there is no reporting. Since the capital costs are already sunk, does it make a lot of difference if we get rid of that particular regulation, unless, of course, someone is about to build a new plant?

Ultimately I think that the exercise was more an inside-the-Beltway attempt to clean up the books. Moreover, there was a tendency to equate the extent of regulation with the number of pages in the Federal Register. I do not believe that the number of pages in the Federal Register is the best performance measure or the way to judge the efficacy of regulations.

Regulation: The number of pages in the Federal Register has always been recognized as only a proxy. Is there clearly a better measure? I am personally skeptical of the potential for a regulatory budget, for example.

Katzen: The idea of a regulatory budget is theoretically attractive, but how it would work, I think, would engender as much controversy and divisiveness as anything else. You raise an excellent question: what is a good performance measure? Part of the answer may be a little bit like Justice Stewart’s definition of obscenity: you know it when you see it. Those who live with the regulation (those who are the intended beneficiaries and those who are burdened by it, as well as those who monitor or enforce it) can say whether it is making sense or not, and the beauty (or efficacy) is in the eyes of the beholder. I suspect, however, that a qualitative measure may be difficult to agree on.

Regulation: We will probably expect a substantial amount of new labor regulation and labor legislation: minimum wage and striker replacement. The Electromation case in the National Labor Relations Board is very likely to be reviewed. Are you likely to be part of the process in reviewing matters of that nature?

Katzen: If there are new regulations from an executive branch agency, I expect to be part of the process.

Regulation: On the basis of how Panetta and Rivlin have drawn upon you so far, do you expect to be part of any legislative review process on matters affecting the basic regulatory legislation?

Katzen: Yes, although it remains to be seen to what extent I am involved with any particular piece of legislation. Although I have been on the job for only three weeks, it seems clear that I will be drawn into all sorts of things. It is also clear that there is only so much that can be done. If I want to make a meaningful contribution, I have a better chance of succeeding if I focus my energies on those things where I can make a difference and leave some of the others that are intriguing, seducing, interesting in any way, shape, or form, for others who can make more of a contribution. I have not drawn any such lines yet. I am still testing the waters, but I hope that by the fall I will have a clearer vision of how I can most productively spend my time, effort, and energy.

Regulation: You paid special attention to the word rational as a description of the Clinton administration’s approach to regulatory process. Can you give me some sense of what you mean by the use of that word? I suspect most people think of their own behavior or the behavior of their organization as rational and want to be judged on that basis. How do you use the word?

Katzen: I was using the word as a surrogate for a description of a process that produces sound decisionmaking.

Regulation: Is the word descriptive of the process or the outcome?

Katzen: Both, because a rational decisionmaking process should produce a rational decision. I use the word rational to distinguish between a predetermined, ideological, result-oriented process and a process in which one gathers information, analyzes all of the implications, weighs the various considerations, and then makes the best judgment possible, based on the available information. That is the distinction I would draw.

Regulation: I understand, however, that Sally Katzen does not have 400 hours a day. So you have to bring some priorities to bear on what you choose to address yourself to. What are those priorities? What criteria do you use to determine where it is important to spend your own time?

Katzen: Throughout my tenure here, I will routinely reexamine my priorities. Coming into the job, I determined that the most important thing is to reestablish the credibility and the integrity of this office. The controversy over the activities of OIRA unfortunately obscured the very valuable work that the office has done. Many of the people who are on the staff are extraordinarily talented, and it is important that others realize that the work they do is very important.

The second thing that I wanted to do coming into the job was to resuscitate the "I" in OIRA and to focus more attention on information policy and technology. The state of information technology in the federal government is appalling. We spend $20 billion to $425 billion a year, but the agencies cannot speak to one another on electronic mail. Everybody else in the country can communicate on electronic mail, but we cannot. Also, as we move to an electronic form of government, there are important policies involving privacy and security on the one hand, and disseminating information on the other, that must be addressed.

Those are two things I have started to work on. I hope that in the fall, perhaps late fall, I can sit down, take a few deep cleansing breaths, and see what I have accomplished and what I want to do next. I hope to do that every several months so that I keep focusing on where I should be spending my time—on what my priorities should be. I hope that they will change as I complete some projects and can move on to other things.

Regulation: Thank you. We wish you the best.


Regulation: The Cato Review of Business & Government, (1993 #3)

Regulation is published four times a year by the Cato Institute. Editorial and business offices are located at 1000 Massachusetts Avenue, N.W., Washington, D.C., 20001. For subscription information, please write to Circulation Department, Cato Institute, same address, or call (202) 842-0200. Send email inquiries to, or subscribe online via the World Wide Web at:

| Regulation | Home | Order Regulation | Publications | Search