Senator BINGAMAN. Regarding your comments on the undermining of the independent sunset review that you folks engage in. How many items come to you for that review each year?

Mr. MACRAE. This is off the top of my head, but I would imagine that most of our, say, two-thirds of the items that we get are recurring submissions.

Senator BINGAMAN. And how many, roughly? Could you give us an estimate as to how extensive that is, how much of your time is spent in reviewing recurring submissions?

Mr. MACRAE. Well-established, well-understood recurring submissions, probably don't get as much time as ones coming up for their first repeat.

Senator BINGAMAN. What we have done in the legislation, as I am sure you are aware, is we tried to retain a sunset provision but essentially delegate it to the agencies and have them self-certify and then give you authority within 30 days of the expiration of their control number, so that you would be able to go ahead and overrule them or step in and object or whatever you want to do.

That seemed to make sense to us from the point of view of practicality. Do you really have the people and time and all to do a meaningful in-house review of each of these things every year or so?

Mr. MACRAE. Well, one of the things that is going to help us, Senator, is that we expect to begin, and we are beginning, to see comments from the public because of the publication of the burden now on the forms. So we are getting responses. And of course, the most vocal one was the IRS Form W-4.

The problem we really have with S. 1742 is the very high standard we have to meet in order, if you will, to march back in and challenge the agency; and that is, the "arbitrary and capricious" standard that is set forth in S. 1742. We think that is too high.

Senator BINGAMAN. So you think a different standard that would give you more discretion would certainly ease some of your concerns?

Mr. MACRAE. Yes. I am not saying, and I don't think any agency has ever recorded, that we didn't have a basis, a legitimate basis based on complaints from the public, that when we challenged them at the renewal time that they didn't take it into account. That isn't to say that we challenge very often. Our overall disapproval rate for all paperwork is only 3 percent, less than 3 percent. Senator BINGAMAN. But a different standard, you think, might solve your concerns in that area?

Mr. MACRAE. I think it would help a lot, yes, sir.

Senator BINGAMAN. In the informal administrative agreement you have with some Members of the House, I guess by definition the provisions in there on ex parte communications, safeguards against delay, and disclosure requirements, are all acceptable to you, it's just that you don't want it in statute?

Mr. MACRAE. Well, I would say, first of all, that the principal change from the 1986 procedures, and we certainly recognize that there is legitimate concern, has to do with the question of delay. One of the things we have striven to do over the last couple of years is to really work off any backlog of regulations that are over

maybe is only 7 or 8 months old. That may sound like a long time, but it's very complicated, some of the regulations. But most of them we act on nowadays within 30 to 45 days. So I think that is important.

There are some discussions ongoing. The House agreement we are still discussing. It is a framework.

Senator BINGAMAN. Do you have any agreement on the delay, or the length of time that you can delay things?

Mr. MACRAE. Yes. I think we basically recognize that we need to express our views certainly within 60 days.

Senator BINGAMAN. Sixty days?

Mr. MACRAE. Yes.

Senator BINGAMAN. Is that for major regulations or is that for anything you think you should address?

Mr. MACRAE. Probably we should be able to do that for everything. But within 60 days. On the other hand it is important to have another delay-delay is a bad choice of words-another 30 days in extraordinary cases.

Senator BINGAMAN. Okay. So 60 days with the possibility of extending it another 30 where it is an extraordinary circumstance? Mr. MACRAE. Yes. And administratively I am trying to get to that point anyway.

Senator BINGAMAN. Right.

My time is up, Mr. Chairman.

Chairman GLENN. Mr. MacRae, you have indicated you have worked over at OMB for many years. You are a career person over there. Let me take your memory back into the Reagan years. Did you ever witness any OMB activity that was partisan or political? You indicated there was none.

Mr. MACRAE. No, I have not, sir.

Chairman GLENN. Did you read David Stockman's book?
Mr. MACRAE. I only read excerpts of it.

Chairman GLENN. Excuse me?

Mr. MACRAE. I only read excerpts. I did not read the entire book. Chairman GLENN. Well, he makes a pretty good case in there, which nobody has refuted as far as I know, that there was phony data generated at OMB to support supply-side economics and that it was done for political purposes to sell a program.

You didn't see any of that over there?

Mr. MACRAE. Well, at that time I was on the budget side, and I was the branch chief for the Treasury and GSA, and there weren't any data to cook there. I mean the deficit was whatever it was.

Chairman GLENN. You never heard any rumors going around that there was skullduggery afoot over there on the supply-side economics figures?

Mr. MACRAE. No, I did not. Regarding the statements that were subsequently made by Mr. Stockman, obviously he believes that to be the case. I don't.

Chairman GLENN. He directed it.

Mr. MACRAE. If this was the case.

Chairman GLENN. That's what he said. He says so in his book. Mr. MACRAE. But I don't necessarily believe that if that were the

Chairman GLENN. Well, I don't know where he was getting his figures. He was getting them from OMB-generated figures, if I understand his book correctly.

Mr. MACRAE. I have never known anybody to ask for us to phony up figures. I have never heard of us being asked to phony up fig

ures.

Chairman GLENN. Well, it has been some time since I went through his book, and I didn't read it in preparation for this morning.

Mr. MACRAE. I am not saying that once figures were provided, where we provided our best guess, that they may not have been manipulated. And that is always the possibility.

Chairman GLENN. Figures don't lie, but liars figure. [Laughter.] Okay. I don't have any more questions.

Senator BINGAMAN. Well, Mr. Chairman, we have a report dated February 21 which was prepared by the Library of Congress, the American Law Division and Mr. Morton Rosenberg on the issue of the constitutionality questions that the Administration raised on this proposed legislation.

I would like to include it in the record if we could.1
Chairman GLENN. Without objection, so ordered.
Thank you, Mr. MacRae.

Mr. MACRAE. Thank you, Mr. Chairman.

Chairman GLENN. We appreciate it. We may have other questions that other Committee members may wish to submit to you, and we would appreciate a response so that it could be included in the record.

Mr. MACRAE. Will do, sir.

Chairman GLENN. Good.

Our next witness is Milton J. Socolar, Special Assistant to the Comptroller General, General Accounting Office.

Mr. Socolar, welcome to our hearing this morning. We look forward to your statement.

TESTIMONY OF MILTON J. SOCOLAR, SPECIAL ASSISTANT TO THE COMPTROLLER GENERAL, U.S. GENERAL ACCOUNTING OFFICE, ACCOMPANIED BY ELEANOR CHELIMSKY, ASSISTANT COMPTROLLER GENERAL, PROGRAM EVALUATION AND METHODOLOGY DIVISION, AND JACK BROCK, DIRECTOR, GOVERNMENT INFORMATION AND FINANCING MANAGEMENT, INFORMATION MANAGEMENT AND TECHNOLOGY DIVISION

Mr. SOCOLAR. Good morning, Mr. Chairman.

I would first like to introduce my colleagues. On my left is our Assistant Comptroller General, Eleanor Chelimsky, who heads our Program Evaluation and Methodology Division. And on my right is Jack Brock, our Director of Government Information and Financial Management, for the work done by our Information Management and Technology Division.

We are pleased to be here to discuss S. 1742, titled the Federal Information Resources Management Act of 1989. I will just briefly

summarize my prepared statement, which I would ask to have introduced into the record.1

Chairman GLENN. Your entire statement will be included in the record.

Mr. SOCOLAR. S. 1742 basically reauthorizes the Paperwork Reduction Act of 1980 and is designed essentially to achieve efficient collection and use of information by the Government, and it also seeks to improve the Government's statistical and other information databases and provide for better access to that information by the public.

We support the thrust of the bill to reduce the burden of Federal paperwork and to change the way information is managed in the Government.

With regard to the management of information in the Government, you might say that Federal agencies are in somewhat of a state of crisis. We spend about $20 billion a year on information technology.

GAO has reviewed the development and implementation of a number of large data systems, and we have yet to find a system, a large system, that has met what we think are three pretty important criteria: We have not found any large system that has come in on time; that has been reasonably close to the estimates that were used to justify the project in the first instance; and perhaps most important of all, that fully satisfied the needs of users for whom the systems were designed.

We tried to get a handle on finding out what some of the underlying, fundamental reasons were for the current state of affairs. And in October of 1989 we organized a symposium at which we listened to members of the private sector who have been highly successful at putting together huge data systems that are comparable to the kinds that we use in the Government.

We had panels where personnel from the Legislative and the Executive Branch had an opportunity to discuss these issues. We listened to Members of Congress. And the symposium concluded with several key, fundamental aspects of bringing these kinds of systems on line that are largely not adhered to in Government systems.

As a matter of fact, we are releasing a report on that symposium today.

Some of these key attributes of successful systems are:

First, that there has to be full interest and commitment at the very top levels of an organization, together with a long-term continuity of project managers. They all, all of the successful systems, proceed under well-developed plans, plans that provide a clear and flexible architecture against which various aspects of any of these projects can be measured;

They all involve large networks of experts, people intra and external to the organizations involved who can provide advice as these projects go along;

And also, very important is that each of these projects has proceeded with service to the customer as one of the primary aspects that the projects are directed toward.

For example, in the Social Security Administration, the emphasis would really be on what it is that a person coming to the Social Security Administration for assistance would need that would drive and direct how the information management system in the agency should be developed. It is not so much designing a system to automate what is presently being done, but rather to use the technology to do business in a better and more efficient way.

Now, the bill, it seems to me, recognizes some of these aspects. The provision of a qualified information resource management official at each agency certainly should help. But I think here too we have to remember that the success of that individual is going to be highly dependent on how much support and interest and attention he can get from other top agency officials.

Program managers today must be capable of understanding and managing information resources and must be able to assist in developing those resources toward more efficient delivery of program services.

Turning now to OIRA, in the reviews that we have made of OIRA operations, we focused on the proper balance between minimizing burden on the public while seeking the best quality of information. We found that burden has remained essentially stable and perhaps with some increase.

We did find problems with the quality of information being collected by agencies and the timeliness of OIRA's clearance process. What we found is that, very often, information is being collected and the usefulness of that information is diminished in many cases because either proper information is not being collected or the information that is being collected isn't properly arrayed in the requests that underlie those information collection activities.

Reviews were slower than in earlier years. There was a large increase in the number of reviews exceeding the legally authorized time limits, and some OMB-approved requests, as I said, were technically flawed.

Finally, review policies were inconsistently applied.

But we concluded that there were changes in OMB's procedures that could lead to notable improvement in information quality.

We support the creation of a commission provided by the bill, which is designed to study the gathering, processing, analysis, and dissemination of information by the Federal Government.

We feel the commission's role should be expanded to include examining how well agencies use information internally to manage or to achieve their mission objectives. At the same time, we would have to recognize that the tasks already laid out for the commission are so broad that we don't think that they could be accomplished in a meaningful way by July 1991. We would suggest a later reporting date.

Finally, I would just make mention of the provision in the bill that calls for a GAO study. We would, in GAO, prefer not to have such a study mandated by the statute. We have already limited resources in the agency and, by mandating such a study into statute, we think that we might well not be able to do what is really required in the most efficient way that would be possible.

We would much prefer to work with the Committee to get at