people in leadership positions, it isn't going to work. So I think that is absolutely essential, and I think you've got to stay more than 18 months. In other words, people who come in and just stay in these political appointee jobs for a few months just aren't getting the job done for the American taxpayer.

I think it is essential to stay at least 4 years, and I think if we had more people staying 6 and 8 years, we'd see a lot more improvement.

Chairman GLENN. Should we require a longer term commitment before we confirm people?

Mr. BowSHER. Well, I have always thought that that should be part of the confirmation process, that is, asking the question how long do you plan to stay, and putting an emphasis on encouraging the people to stay for a reasonable time, because there is no large organization where you can make good progress with leaders turning over as fast as we turn them over here in the Federal Government.

Chairman GLENN. Is 18 months the current estimate, do you know?

Mr. BOWSHER. I think that's in the ball park, Mr. Chairmanand it has been that way now for quite a long time. We have had a turnover. And you know, I have a unique position of a 15-year term. What I have been doing is inviting all the new heads of agencies and Cabinet members over to GÃO to meet with us, have lunch with us, get to know the GAO-and I'm on my 4th and 5th and 6th person in 11 years in some of these departments and agencies. And it is discouraging, it is very discouraging.

In fact, one Cabinet officer told me a marvelous story. He was on an elevator with some of his people, and they asked him how he liked his job, how he was doing, and he said he decided to play Mayor Koch and turn it back on them, and he said, "How do you think I'm doing? You tell me." And the person came back and said, "It doesn't make any difference-you aren't going to be here long enough.'

That's a terrible thing when the people down the line view the top person in that light.

Chairman GLENN. That's one of the sickest jokes I've heard recently.

Mr. BOWSHER. Yes, it is really sad.

Chairman GLENN. I think that's an area that we've never really addressed. I don't know how you address it. I don't know what sort of restrictions you could put on the President's right to appoint who he or she may want in public office, and yet it seems to me it is so key, and we sort of dance around it as being unsolvable, and we try to set up frameworks of other legislation to address it. They aren't going to work unless you have good management that is there long enough to do something and be responsible.

Mr. BOWSHER. One thing that other countries have been trying to do and Australia is a good example-is to get more of a defined role for the senior civilian executives as to what they are responsible for carrying out. I have always felt that we sometimes don't expect our senior civilian executives to carry out as big a role as they are very capable of carrying out, and that's an area where I

The Volker Commission spoke to that; they raised the question whether we needed as many Presidential and political appointees as we have, and whether we could do a better job of managing these departments and programs. I think that is an area where we could make a lot of improvement.

Chairman GLENN. The President's Commission on the Federal Appointment Process criticized the lack of a comprehensive list of political appointee positions which would detail job descriptions and needed qualifications. The President's Commission recommended such a list would help and "present and future Presidential personnel offices in recruiting nominees of the highest caliber." I think the Commission very rightly says that this should be "regularly updated public information.'

Should we go ahead and incorporate the Commission's recommendations with respect to political appointee job descriptions and qualifications in the legislation? Maybe S. 20 would be a good vehicle to incorporate that into.

Mr. BOWSHER. I'd have to think about that.

Chairman GLENN. That would be a big step, an enormous step. Mr. BowSHER. Yes.

Chairman GLENN. Senator Roth?

Senator ROTH. I think the question of personnel is critically important, but I think we'd better reflect carefully as to what we think needs to be done in this area.

I think one of the problems is that it is very difficult to attract people from the private sector to Government. I think the problems of trying to flyspeck an individual's reputation from the day he was born forward makes it a very unattractive offer to many people.

But secondly, we have a political process. It's all very well to say everything ought to be covered by civil service, but the purpose of the election this fall is to decide in what direction this country is going, and whoever is elected President should have the opportunity to put into effect the kinds of policies he espouses; otherwise the political process is a mockery.

I agree that this question of people serving too briefly is a critical one, but I think at the same time Congress as well as the executive branch had better look at the role it has played in making it unattractive today for good people to want to get involved in Government.

The other comment I would make is that when you do have a turnover-and you always have, as long as I have been down here, whether it is Republican or Democrat-it makes it all the more important that you have the institutions or procedures in place that will help move the program forward. It seems to me even more important that we have in place performance procedures that will function. Normally, your cabinet secretary sets broad policy; he really isn't getting into the details of every program, except to look at the overall effect and the way they are operating.

So I share the concern expressed about people coming and going, but I think the two problems are separate and should not be intertwined. As I said, I think the people want some results now, and they are weary of the inability of the executive branch and Con

about today is one step that can hold out some hope. My main concern is the time it will take to implement, which I think you have to be careful about.

Let me ask one final question. Can this program be done without substantial additional expense, and how can we move forward and accomplish what we want, but not make it expensive either from the standpoint of personnel or cost?

Mr. BOWSHER. It shouldn't be that expensive. In other words, your expenses should be some additional training, which I think is an important area when you are trying to make changes like this, and some improvements in some of the systems. But those are essential. In other words, what we really should have is Government departments that are running successful programs. And, as you say, Senator Roth, when a new President is elected, they should be able to concentrate on policy, but I think now a tremendous amount of their time is taken up by the fact that the current process is not working very well, so they are constantly spending a lot of time bailing water, you might say, in these agencies, trying to keep things going.

So I don't think the cost should be all that great; I really don't. Senator ROTH. Thank you, Mr. Chairman.

I would ask your organization, Mr. Bowsher, to give us the advantage of any advice and comments. This is an area where I think your organization has and should have special expertise. It's the kind of reform that I think is critically important if people's faith in Government is to be restored.

Mr. BOWSHER. We'd be pleased to work with the Committee, Senator.

Chairman GLENN. I agree. We have to move on shortly, but I just want to say that I really think this idea of personnel is very important. The best organization-and this may be a little bit of a purist view, and I know what we're trying to come up with here is something in the middle-but I just think the best organization isn't going to work without good people. We can organize from now until forever, but without good people in there, it isn't going to work; and then, on the other hand, if you have really good people to begin with, we probably don't need all this much organizational advice to tell them how to do the job if they are going to be there long enough to do it. I just think that the level of expertise of a lot of people going into departments and agencies, and in diplomatic posts as well, shouldn't be judged more by their fundraising ability in the last election than on their job expertise.

I think that the willingness to serve long enough to be effective and the willingness of the administration to pick really good people is key here. We can give them some organizational tools, and we can give them all the studies and pilot projects you want, but until we can get some really good people in here, we aren't going to solve this problem.

We do have to move on, so we appreciate your being here, and we'd appreciate your willingness to answer any additional questions we might submit to you in writing that will be included in the record.

Thank you all for being here this morning.

Chairman GLENN. The next witness is the honorable Frank Hodsoll, Deputy Director for Management, Office of Management and Budget.

Frank, welcome to the hearing this morning. We look forward to your comments.

TESTIMONY OF HON. FRANCIS S. HODSOLL,1 DEPUTY DIRECTOR FOR MANAGEMENT, OFFICE OF MANAGEMENT AND BUDGET Mr. HODSOLL. Thank you, Mr. Chairman.

It's a pleasure for me to be before the Committee once again. I have a statement which, if I may, I would submit for the record. Chairman GLENN. Without objection, it will be included in the record in its entirety.

Mr. HODSOLL. Thank you, Senator.

It is a particular pleasure for me to follow the distinguished Comptroller General. Let me just say as he is leaving the room, that GAO has been an enormous help to us as we have proceeded with the CFOs Act and a number of other matters as well.

I would agree with the Comptroller General and John Hill that the issue of performance measures needs to be broader then financial. Indeed, the guidance that we issue presages that. It was our view that the link to financial statements was the first phase, and we said at the end of our statement that we were planning to proceed beyond that after getting the first phase started. So in that sense we do agree.

First, if I could, let me congratulate the Committee, particularly you, Mr. Chairman, and Senator Roth, for taking the lead in trying to develop a better framework for measuring the Government's performance. As you, Senator Roth, noted both here and on the Floor recently, we have a situation which is most unsalutary: that is, we have all-time lows in terms of confidence in Government. You mentioned the point about how the average American believes roughly half of every dollar in Federal taxes, is wasted. One might also note that, in a recent poll, 78 percent of Americans said they were dissatisfied with the performance and functioning of the Federal Government.

There are many reasons for this; but surely, we would argue, a contributing factor has to be our general inability to demonstrate performance consistently and accurately. One might ask the rhetorical question: what private entity could survive if it didn't know how it was doing and couldn't show others how it was doing. I would suggest they'd have great difficulty tapping the capital markets in that kind of condition.

This is not to say, however, that there is in our view no measurement of performance going on. Michael McGinnis and Raymond Uhalde will tell the Committee of important efforts in the Public Health Service and the Employment and Training Administration. As the Comptroller General has noted, three-quarters of the 103 agencies surveyed do indeed have performance data of some kind. The problem is that, again as the Comptroller General has pointed out, only nine of these agencies have an administrative struc

ture in place for developing and reporting results so that policymakers can see them. Most of the data is reported internally, and policymakers are unaware of it. And a recent Department of Labor study involving education and training programs indicated that outcome data from these efforts was largely informational.

There is no question in our view that we need better and more systematic performance measurement. The CFOs Act envisions this. That is, incidentally, how the United Kingdom began, with the equivalent of a CFOs Act in the beginning of the 1980's. The strong movement toward quality management, which the Comptroller General also referred to, must of necessity include performance measurement.

Private entities do it as a matter of course. State and local governments, it has been pointed out, are exploring the idea. Some, like Sunnyvale, have been doing it for years, and of course, John Mercer, who is on the Committee staff, was mayor out there, and is to be congratulated for what he achieved in that city. Some other countries, as has been pointed out, have also been doing it.

The only question, it seems to me, is how do we do this successfully, and how do we surely-and I will underscore the word "surely"-move forward.

In my prepared testimony, I had eight principles. I'd like to recombine them, and somewhat add to them, and suggest five . . . propositions, let's call them.

First, it seems to me that performance-that is policy and program outcomes, as well as inputs and dollars-must become a part of the political and policy debate, to become unavoidable by policymakers. This hearing, and your consideration of legislation, helps with that.

Second, we must agree on what to measure, program by program. It cannot be done across the board. We must have objectives agreed upon, the objectives of the legislation as well as key indicators, and the definitions at the nitty-gritty level must be solid.

Third, we must keep it simple-a few indicators for each program-and then start with the easier programs so that we can build some confidence that we will have a system that will actually work.

Fourth, we must link program and financial and budget data. Fifth, we must involve the people in the trenches, and by that I mean the senior civil servants, and build incentives for them to be a part of this, as opposed to a paper exercise. That means trading, in my judgment, flexibility for accountability; that's what total quality management is really all about.

One last point before concluding. We need to apply performance measurement to entitlement programs. We often think of this mostly in the discretionary area, but entitlement programs are critical here, it seems to me. They are over half the budget and growing.

Let me give you one example from the health care area. After deposit insurance, Medicare and Medicaid lead the pack in terms of annual growth during the last 10 years-Medicaid, at over 15 percent a year, and Medicare at roughly 10 percent. The Federal Hospital Insurance Trust Fund, just to take one example, will be in a deficit condition in 10 years; the actuarial shortfall is 4.2 percent of